Title: A Secret History of the IRA
Author: Ed Moloney
Date: October 2002
SKU: ZH12

A Secret History of the IRA

Second Edition

Ed Moloney

PENGUIN BOOKS

Published by the Penguin Group
Penguin Books Ltd, 80 Strand, London WC2R 0RL, England
Penguin Group (USA) Inc., 375 Hudson Street, New York, New York 10014, USA
Penguin Group (Canada), 90 Eglinton Avenue East, Suite 700, Toronto, Ontario, Canada M4P 2Y3
(a division of Pearson Penguin Canada Inc.)
Penguin Ireland, 25 St Stephen’s Green, Dublin 2, Ireland
(a division of Penguin Books Ltd)
Penguin Group (Australia), 250 Camberwell Road, Camberwell, Victoria 3124, Australia
(a division of Pearson Australia Group Pty Ltd)
Penguin Books India Pvt Ltd, 11 Community Centre,
Panchsheel Park, New Delhi – 110 017, India
Penguin Group (NZ), 67 Apollo Drive, Rosedale, North Shore 0632, New Zealand
(a division of Pearson New Zealand Ltd)
Penguin Books (South Africa) (Pty) Ltd, 24 Sturdee Avenue, Rosebank,
Johannesburg 2196, South Africa

Penguin Books Ltd, Registered Offices: 80 Strand, London WC2R 0RL, England

www.penguin.com

First published in the United States of America by W.W. Norton and Company Ltd 2002
First published in Great Britain by Allen Lane The Penguin Press 2002
Published in Penguin Books 2003
Second edition published in Penguin Books 2007
1

Ed Moloney, 2002, 2007

Map on page 135 courtesy of Bob E. Hall, Center for Earth and Environmental Science, Indiana
University–Purdue University, Indianapolis. Maps on pages 36, 76, 94, 351 reprinted from
Provisional Irish Republicans by Robert W. White. Copyright 1993 by Robert W. White. Reproduced
with permission of Greenwood Publishing Group, Inc., Westport, Conn.

Except in the United States of America, this book is sold subject
to the condition that it shall not, by way of trade or otherwise, be lent,
re-sold, hired out, or otherwise circulated without the publisher’s
prior consent in any form of binding or cover other than that in
which it is published and without a similar condition including this
condition being imposed on the subsequent purchaser

EISBN: 978–0–141–90069–8

This book is dedicated to
all the people who lost their lives
in the Northern Ireland Troubles.

History, Stephen said, is a nightmare
from which I am trying to awake.
—James Joyce, Ulysses

Preface to the second edition

It is to the enormous credit of Penguin Books and my editor there, Simon Winder, that I have been given the opportunity to update A Secret History of the IRA. The first edition ended just after the IRA’s first act of decommissioning in the autumn of 2001, following the September 11 attacks in the United States and the arrest of IRA personnel in Colombia a month beforehand. This edition brings the story to what I would argue is the definitive end of the Provisional IRA as an instrument of armed and revolutionary resistance to British policy in Ireland. With that event, I believe, it is now also possible to say that the Troubles have ended.

The ending was signaled in two ways. First, by the IRA statement in July 2005 that formally announced that its armed struggle against Britain was over; and secondly by the completion of the decommissioning process the following September. Doubtless the IRA retained some weaponry to protect its leadership and key members against rivals and enemies but the destruction of its Libyan-supplied arsenal robbed the IRA of the capacity to wage war. Not only that but the act itself was replete with symbolic meaning, indicating a wish to eschew armed struggle in favour of political methods from thereon. The event was unprecedented in the history of Ireland’s struggle for independence.

This point came after five long and often turbulent years which saw the downfall of the Ulster Unionist leader, David Trimble, and his replacement as unionism’s leader by his party’s long-time loyalist critic and rival, Ian Paisley, leader of the Democratic Unionists and permanent Moderator of the Free Presbyterian Church. Nationalist politics were refashioned in an equally radical way with the IRA’s political partner, Sinn Fein, replacing the SDLP as the dominant party and Gerry Adams displacing John Hume as nationalism’s leader.

With that, the moderate center-ground of Northern Irish politics—in as much as it had ever existed—disappeared and potential executive power was transferred, as one acerbic critic put it, into the hands of a theocrat and an autocrat who might or might not agree to exercise it in the best interests of all the people of Northern Ireland.

That this state of affairs was brought about by the handling of IRA decommissioning is beyond dispute. It is a central thesis of the second edition of this book that the IRA could have decommissioned all its weapons much sooner, but chose instead to prevaricate and thereby to inject an even more virulent strain of sectarianism into Northern Irish politics. In such a way were Trimble and the SDLP destabilized and Sinn Fein catapulted to political and electoral success. It is surely no coincidence that once the process of destroying Trimble and sidelining the SDLP was completed, all the IRA’s objections to final decommissioning suddenly vanished.

The plaudits that came the way of the British prime minister, Tony Blair, and his Irish counterpart, Bertie Ahern, in the wake of the IRA finally completing the decommissioning process are understandable. The two leaders had worked for years for this day to dawn, but any credit that they are due should, however, be balanced by the knowledge that it was they who facilitated this Sinn Fein strategy and it is they, ultimately, who bear responsibility for this triumph of the extremes in Northern Ireland.

At the time of writing, the DUP and Sinn Fein were still squabbling about the one issue remaining in the way of their sharing government: Sinn Fein’s acceptance of the policing and criminal justice system. It remains to be seen whether their differences, which concern the practicalities rather than the principle of the matter, can be overcome but of a number of things there can be little doubt. One is that the Provisional IRA’s war is over for good. Another is that the politics of Northern Ireland have been changed forever and in a way no one could have predicted. Whether this will be for good or ill is a verdict only the passage of time can deliver, but the democratic record of Northern Ireland’s new potential political leadership gives little cause for optimism.

Penguin’s decision to publish the second edition of this book has enabled me to tell that tortuous if significant tale in two new chapters—Nineteen and Twenty—as well as in an Epilogue which takes the story up to the Assembly election of March 2007. I have also been able to correct some errors and compensate for some omissions in the first edition and add new information about the betrayal of the gun-running ship the Eksund and the attitudes of different Army Council members, including Gerry Adams and Martin McGuinness, towards it. Fresh information has come to light about Adams’s IRA career after his release from Long Kesh in 1977 and that section of the first edition had been amended accordingly. The sad death of former Fianna Fail taoiseach Charles Haughey in June 2006 also releases me from the pledge of confidentiality I gave him while researching the first edition. Haughey played a crucial and often undervalued part in the genesis and evolution of the peace process and I believe that without him the birth of the peace process would have been much more difficult. At this point I should put on record that Haughey, who was in the early stages of the illness that claimed his life when we first spoke, did not search me out but rather was a reluctant and often grudging source. “The stage,” he once grumbled in the face of my persistence, “is already overcrowded with people attempting to claim credit.” It was I who deduced that he had played such an important role and it was I who sought him out. Only after numerous and lengthy visits to Kinsealy was I able to persuade him to tell his part in the story. I am now very happy to be able to acknowledge all this.

One key part of his story, and in its own way an extremely valuable historical document, is the letter of May 1987 written to Haughey by Father Alec Reid, seeking a secret dialogue between him and Gerry Adams and setting out the terms for an IRA cessation that Adams would find acceptable. I was able to read and fully transcribe this letter in Kinsealy. It is compelling and extraordinary evidence that, many years before the Irish public and the bulk of IRA members and leaders learned of the peace process, the Sinn Fein president was seeking to end the IRA’s war on terms little different from those implicit in the Good Friday Agreement. I am now able to reproduce this in full along with other documents pertaining to the early part of the peace process.

Father Reid himself has also started to acknowledge that the peace process began much earlier than any of the participants were previously prepared to admit. In a BBC Radio Four interview in November 2006 with Olivia O’Leary, the Redemptorist priest conceded that he began his discussions with Gerry Adams not long after the abduction and killing of a UDR soldier in South Armagh, an event that took place in the autumn of 1982 and was described in detail in the first edition of this book.[1] This is a welcome and overdue sign that the information permafrost surrounding the peace process is beginning to melt. It is of exceptional historical importance that those who were involved in the IRA during this time also begin to put their accounts and memories on record. History should not always be written by the victors.

By the time the second edition of A Secret History of the IRA is available on the bookstands, Gerry Adams’s prediction that there will be a united Ireland by 2016, the centenary of the Easter Rising, will be only nine years away. We will not have to wait long to discover whether that statement was the product of irrational exuberance, political expediency or an accurate assessment of constitutional possibilities.

Ed Moloney
New York
December 2006

Preface to the first edition

There were never any strategic considerations at stake, like those in the Middle East, for example, nor did the killing ever approach the carnage or savagery of the Balkans or Rwanda. There were no oil fields or gold mines to be captured, or any ideology to be overthrown or vindicated. But the Troubles in Northern Ireland had one quality that marked the violence there as special, and that was the sheer length of the conflict—that and the fact that no one could really see an end to it.

By the time of the first IRA cease-fire of 1994, the Troubles had lasted for a quarter of a century, so long that the violence had become an almost permanent feature of the world’s political landscape. Other conflicts would erupt, climax, and then fade away, their names soon forgotten, but the bombing and shooting in Belfast, Derry, and South Armagh seemed to last forever.

The Troubles were at or near the bottom of the list of significant global conflicts, a low-intensity war that occasionally exploded into spectacular bursts of violence but more often was characterized by a killing or two a week, deaths that by the end had become so routine that they scarcely merited a headline outside of Ireland. But the violence devastated a whole society, scarring two generations of Irish people, the baby boomers who came of age when the Troubles began in the late 1960s and their children, who grew up knowing only instability and bloodshed.

More than 3,700 people were killed in the violence, an average of just over two a week for the thirty years that the conflict lasted. Almost as many people died within a couple of hours in lower Manhattan on September 11, 2001, but to conclude therefore that the Troubles were a petty affair would be a mistake. Had a similar conflict consumed the United States, the equivalent death toll would have been over 600,000; in Britain, 150,000. Nearly 1 in every 50 of Northern Ireland’s 1.5 million people, some 30,000, were injured in the violence. The comparable figure in the United States would be 5 million; in Britain, just over 1 million. Very few people in Northern Ireland did not personally know someone who had been killed in the Troubles, and many knew several. There are many definitions of a civil war, but that is surely one of the most compelling.

Thousands were caught up in the Troubles in a more intimate fashion, becoming members of groups like the IRA and, on the loyalist side, the Ulster Defence Association and the Ulster Volunteer Force, for whom they killed, wounded, maimed, bombed, robbed, and went to jail or early graves. Putting a figure to such numbers is by definition difficult, for these were not organizations that kept detailed personnel records. But Martin McGuinness, who has held virtually all the senior ranks in the IRA, once told the author that he reckoned 10,000 people had been through the ranks of the IRA over the years—and who could know better?—while at the height of its power the UDA could, given an hour or two of notice, put 20,000 men on the streets of Belfast. In all the important ways the Troubles pervaded Northern Ireland.

The conflict was not confined to the geographical boundaries of the state. It regularly spilled over into the Irish Republic, injecting an unwelcome instability into the body politic there and warping the institutions of the young state almost beyond repair, the media and the legal system in particular. Its effects were also felt farther afield, wherever the Irish diaspora had scattered its unfortunate people—in the United States, Australia, and Canada—and the violence was repeatedly exported to Britain, where scores lost their lives over the years in bombings and shootings, mostly carried out by the IRA.

The Troubles were, above all else, the latest and the most protracted phase in an Anglo-Irish conflict that had properly begun some four hundred years earlier, with the Tudor wars and plantations of the sixteenth century, although there had been resistance of some sort to the English presence ever since the Normans invaded in the twelfth century. Ireland was Britain’s first colony and one of its last. Resistance to occupation went through alternating phases of violence and politics, and each stage in the conflict brought Ireland a little nearer to complete separation from Britain. The culmination of all this was the Treaty of 1921, a settlement that paved the way for twenty-six of Ireland’s thirty-two counties first to govern themselves and then, in 1949, to declare themselves a republic while the remnant of the island in the North remained British.

Had the architects of the 1921 settlement set out to create an inherently unstable entity, they could scarcely have done better than to design Northern Ireland in the way they did. The state contained within its boundaries the seeds of its own devastation. Packed into its narrow confines were two troubled communities. One was the uneasy majority, the Protestants whose ancestors had planted or colonized native lands so as to make Ireland a safer place for Britain. But they paid a terrible price for their service to the motherland. They and their forebears lived in constant fear of retribution from the substantial minority of Irish Catholics in their midst, the native Gaels they had supplanted and whose land they had taken, while they came to distrust the British almost as much as they feared the Irish. The Protestants knew nothing but insecurity and learned to reject anything else as a distortion of the natural order.

Trapped in a state not of their choosing, the Catholics were bitter, resentful, and full of foreboding. Abandoned by their Southern co-religionists after 1921, confronted by arrogant, superior-seeming rulers, and subjected to intermittent salvos of pogrom-like violence, they knew they could look only to themselves for protection, and trust only their own. Each community feared and distrusted the other, and in such circumstances it would have been odd had bigotry and discrimination not shaped the politics of Northern Ireland or had this not, in the end, been the cause of a tremendous conflagration.

Thirty years after they exploded, the Troubles have ended in what is arguably a most definitive fashion, an ending that marks not merely the closing of a war but rather the conclusion of the historic conflict between Ireland and Britain.

They have ended with the leadership of the Provisional IRA accepting Britain’s neutrality in Northern Ireland. No longer do its leaders preach that London is a colonial, occupying power, usurping the right of the Irish people to decide their own future. Instead, the republicans have accepted a political process whose foundation stone is the principle of consent, an acknowledgment that unionists cannot be forced into a united Ireland against their wishes. In their turn, unionists, with varying degrees of commitment and enthusiasm, have accepted that Northern Ireland must become a warmer and more welcoming place for Catholics. In a sense the two communities have struck a bargain—the Catholics agreed to abandon the goal of Irish unity in return for a secure place within the state, while the Protestants consented to behave toward their neighbors in a more civilized way.

None of this would have been possible without the proactive cooperation of the Provisional IRA leadership. Indeed it is the central thesis of this book—and its principal revelation—that it was that organization’s dominating figure, Gerry Adams, who launched, shaped, nurtured, and eventually guided the peace process to a successful conclusion. Many excellent accounts have been given, in written and televised form, about the peace process; without singling any out for mention, it would be fair to say that they have all dwelt mostly on the negotiations and high-level talks that characterized the latter stages of the process. None have examined, or been able satisfactorily to explain, the events that took place much earlier within the IRA and Sinn Fein which made all this possible, or to tell how they happened.

This book attempts to redress that deficiency, tries to delve deep within the belly of the beast—or as deep as any outsider can go—to expose the entrails for examination. The conclusion is unavoidable. The Irish peace process was not a spontaneous phenomenon, tossed around by forces outside its control, nor was it forced upon its architects by the fortunes of war. The process was a little like a precooked dinner whose basic menu had largely been decided long before most of the diners knew the meal was planned, even if the table settings, the guest list, the size and shape of the crockery, cutlery, and condiments, and so on were not. The peace process was, in other words, an exercise in management toward an already decided outcome, as much as it was anything else. There were many delays and threats to the event, but at the end there was little doubt that people would sit down to eat, and eat well.

After twenty years of reporting on, writing about, mixing with, and observing the IRA at close quarters, I had come to a number of conclusions about the Belfast-based leadership that came to power in the 1970s. Principal among these was that the people guiding the organization were longsighted, bright, talented, dedicated, determined, pragmatic, cunning, and all too often duplicitous. They were also utterly ruthless in their mission, which above all else was to survive and prosper, and were devoted to their leader and inspiration, Gerry Adams. The idea that he or the people around him would allow any but one of their own to control and direct their journey was so absurd that it was not even worthy of consideration. The facts about the peace process revealed in this book substantiate that assumption.

A Secret History of the IRA is as much about Gerry Adams as about the organization he dominated for so long. He is, indisputably, one of the largest figures in Ireland’s long and sad history, a revolutionary leader who deservedly ranks alongside those competing founders of Irish independence Michael Collins and Eamon de Valera, and a man whose qualities, both negative and positive, are fit to be measured alongside theirs. Collins, de Valera, and Adams all left indelible marks on the Ireland of their day, but whereas the veterans of the 1916 Rising and the subsequent Anglo-Irish war could allow much of their story to be told while they were still alive, Adams has not been able to. In December 1998 the Nobel Peace Prize was awarded jointly to David Trimble and John Hume, the leaders respectively of Ulster unionism and the Social Democratic and Labour Party (SDLP), for their contribution to the success of the peace process. Standing alongside them, sharing in the glory, should have been Gerry Adams and—arguably— Father Alec Reid, who between them kick-started and sustained the process. But while others have collected plaudits and the glittering prizes, Adams has been forced to stay silent, biting his lip lest by accepting the praise of the establishment he undermine the peace process in the eyes of his supporters. The truth was withheld to sustain the project, a fact to which Adams’s own less than enlightening autobiography pays painful testament. This book is also an attempt to rectify that, although not many of his friends will thank me for it.

I do not claim this work to be an exhaustive account of this extraordinary period in Irish history. Many blank spaces remain to be filled in, but perhaps now that task will look a little less onerous. The job of a correspondent, after all, is to inform and increase understanding. If this book has helped make the Irish peace process more intelligible to the outside world, then surely that can only be for the good.

Ed Moloney
New York
February 2002

Acknowledgments

This book has taken about four years to research and write, mostly in secret. I have little doubt that, had it been more widely known that I was involved in this enterprise, the research would have been made much more difficult to conduct, perhaps impossible, and the book would probably have been stillborn. I must, therefore, first of all thank those people who knew what I was doing but kept silent, and, equally, I must apologize to others I was obliged to mislead or behave evasively toward. This was done not out of any malice or lack of trust but in the knowledge that to ask a person to carry someone else’s secret is often to ask them to shoulder the most onerous of burdens.

Officially, the leaders of the Provisional republican movement did not cooperate in the research for this book, nor was their cooperation sought. They had made it abundantly clear to me, and to other journalists, on several occasions in the past that they had no interest in talking about, much less in assisting anyone in writing a book on, the genesis of the peace process. Having completed the research on this period, I can understand why. This work, however, is about more than the peace process. It is also a history of the Provisional IRA and as such reflects some twenty years of reporting the organization in Hibernia, Magill, the Irish Times, and the Sunday Tribune. For much of that time I kept notes of conversations and exchanges with many of the IRA’s current leaders, and these have proved to be an invaluable aid in this undertaking, not least in illuminating the explanation for the peace process which was deemed fit for public and internal IRA consumption. Those among them unhappy to see this book appear can at least console themselves with the knowledge that I have not betrayed their confidence after all these years.

Many republicans, past and present members of both the IRA and Sinn Fein, were, however, happy to speak to me, and it is no exaggeration to say that without the information and insights they gave me, this book would have been impossible to write. They all know who they are. They spoke to me in the greatest secrecy, and I pledged never to compromise them. For that reason they must be nameless here, but nevertheless I must acknowledge the debt I owe them. A number of past and present officials in, and members of, the British and Irish governments added hugely to this story, and their contribution was likewise invaluable. Tom King and Peter Brooke were both generous with their time and information, but others asked to stay unnamed and I must respect their wish. So did members of the British army, community workers, and clerical figures of both faiths in Northern Ireland who also helped me enormously. I thank them all. The day may arrive, perhaps, when such people will be able to speak freely and openly about the part they played in the historic events chronicled in this book. But I fear that day is still a long way off.

Thankfully there are some to whom I can express gratitude by name. Many of my former colleagues at the Sunday Tribune offices in Dublin gave support and encouragement to me both before publication and afterwards. To them and to Seamus Dooley of the NUJ, I express my gratitude. In particular I would like to thank the paper’s former news editor, Helen Callanan, whose decision to seek a more fulfilling life at the Bar has been a great loss to Irish journalism, who urged me to persevere at an early and critical stage, while Harry McGee, also of the Tribune, was generous with advice and morale-boosting comments. Dr. Anthony McIntyre gave me access to a number of embargoed IRA interviews he conducted for his doctoral thesis, which filled important gaps in the early history of the Provisionals. Frank Millar read the early drafts in London, and his positive response encouraged me to seek a larger publisher. Patrick Farrelly, Kate O’Callaghan, Sandy Boyer, and Terry Golway, all in New York, gave me encouragement, also at an early stage when it might have been easier to choose another path. Bob White and Kevin Mickey, both of Indiana University, helped obtain some beautiful maps, and while all the Irish photographers whose work is featured in the book were enormously helpful, I must thank Kelvin Boyes for going that extra mile. It is said that no book on the Troubles is complete without an acknowledgment of the part played in its production by the staff and management of the Linen Hall Library in Belfast. This book is no exception. In particular I owe much to the generous assistance of Yvonne Murphy and her staff in the library’s priceless political collection. Joan McKiernan worked hard and long to compile a most comprehensive index, and for that I thank her.

I have been extremely fortunate to have been published by two most distinguished houses—Penguin, in the UK and Canada, and W. W. Norton, in the United States. I would like to single out Simon Winder at Penguin in London for particular thanks. He was the first to recognize the importance of this book, and his support for it has been wholehearted. He is lucky to have such an impressive team to draw upon, and I would like to thank Ruth Killick, Rosie Glaisher, Jennifer Todd, Mark Scholes, Pippa Wright, Louise Wilder, Clare Needham, and Andrew Stephenson for their sterling work. My gratitude also to Cynthia Good at Penguin (Canada) and to my editor in Toronto, Michael Schellenberg. At W. W. Norton in New York, I am indebted to Bob Weil and Jason Baskin, who had the unenviable, day-to-day chore of dealing with the author. Fate and geography decreed that the daunting task of editing this book would fall to Bob Weil, and I wish to pay him a special tribute. His efforts, characterized by patience and good humor, transformed a passable work into something to which I am pleased to attach my name. His is a unique talent. W. W. Norton also provided an enormously impressive team and to each of them—to Otto Sonntag, Nancy Palmquist, Rene Schwartz, Andrew Marasia, Louise Brockett, and Dan Deitch—I give my thanks. To my agent, Jonathan Williams, in Dublin, must also go a special appreciation for tirelessly and so successfully generating interest in the book. It goes without saying that without him, none of this would have been possible.

Finally, I must thank members of my family who helped and encouraged me. My sister, Michelle, and her husband, Tom Bray—not to mention young Lawrence and Liam—were generous with their hospitality during my trips to England, while, as always, Joan and Ciaran were the lights in my life.

A <br>SECRET <br>HISTORY <br>OF THE <br>IRA

Prologue

There was only one thought in Gabriel Cleary’s mind, and it chilled him. As he checked the firing unit linked to the twelve explosive charges placed beneath the Eksund’s waterline, the signs of sabotage were unmistakable. With a growing sense of horror the IRA’s director of engineering realized that the most ambitious gunrunning plot ever in the IRA’s long war with Britain had been betrayed.

Cleary’s fears had been growing ever since the Eksund had left the Libyan capital, Tripoli, some two weeks earlier, as he later told an IRA inquiry in messages smuggled from a French prison.[1] The Panamanian-registered vessel had been loaded with some 150 tons of modern, sophisticated weaponry at Tripoli dockside by sailors from Colonel Qaddafi’s small naval service on October 13 and 14, 1987. Although that part of the operation had gone smoothly, Cleary was uneasy. This was the fifth trip since August 1985, but the four earlier cargoes, amounting in total to another 150 tons of weapons and explosives, had been safely and secretly transferred to IRA boats from a Libyan vessel off Malta, well out at sea and far from the sight of hostile, prying eyes.

This operation had to be handled differently. The Eksund’s cargo was as large as the four other shipments put together. The sheer size and bulk of weaponry involved meant that the loading process would be lengthy, and that made an operation at sea simply out of the question. With the CIA and other Western intelligence agencies taking an ever-greater interest in Libyan affairs, the chances of being spotted by satellite surveillance were too great. The Eksund’s manifest was breathtaking: 1,000 Romanian-made AK-47 automatic rifles, a million rounds of ammunition, 430 grenades, 12 rocket-propelled grenade launchers with ample supplies of grenades and rockets, 12 heavy Russian DHSK machine guns, over 50 SAM-7 ground-to-air missiles capable of downing British army helicopters, 2,000 electric detonators and 4,700 fuses, 106 millimeter cannons, general-purpose machine guns, anti-tank missile launchers, flame throwers, and two tons of the powerful Czech-made explosive Semtex. With a cargo like that to load there was no option; the work had to be done in Tripoli itself.

The Libyan harbor was a dangerous place for IRA men on a mission to smuggle weapons. The Tripoli docks were regularly jammed with ships unloading consumer goods, as a result of a massive oil-financed consumer boom; the bustling labor force was a mixture of Arabs drawn from nearly every country in North Africa and European expatriates lured by the high salaries offered in this former Italian colony.

Although the nearby British embassy was closed, emptied of its staff following a major diplomatic row with Qaddafi, everyone, IRA and Libyans alike, assumed that the British had left their spies behind. Some could easily be mingling with the crowds down at the harbor or at the nearby souk where traders bought and sold gold and silver and exchanged gossip.

The Libyans took precautions. The Eksund was loaded at nighttime to reduce the chances of being spotted, and the boat was moored in the military section of the dockside for added security. But even so, Cleary was glad when the Eksund finally weighed anchor.

Within hours of setting sail, however, the IRA commander’s doubts returned. A plane flew directly over the Eksund, and Cleary suspected it was an RAF spotter aircraft. Every day of the voyage thereafter a similar aircraft would perform the same maneuver. There seemed little doubt that someone was keeping a very close eye on the Eksund’s progress. Off Gibraltar the plane swooped down so low that the pilot was visible. Cleary grew more and more nervous.

As the Eksund passed the Brittany coast and veered left for Ireland, the boat ran into a different sort of trouble. The fifty-year-old vessel, which had shipped grain most of her life, had endured a difficult journey out to Malta. At one stage the vessel had to dock in England for engine repairs, and at another point the steering failed.

The steering problem struck again on October 27. The crew tried to make repairs but with no success, and the Eksund drifted closer and closer to the French coast. The next day Cleary realized the mission was doomed and took the fateful decision to scuttle the ship and sink its precious freight before it ran aground. His orders had been precise: on no account must the British learn of the IRA’s arms-smuggling operation; the very outcome of the war depended on secrecy being preserved.

As he assembled the crew on the top deck to prepare the inflatable dinghy that would take them ashore, Cleary started the process of triggering the timing device that would set off the bombs and slowly sink the Eksund. This was the job Cleary had been chosen for.

The colorful Dublin businessman Adrian Hopkins, who had found and purchased the Eksund, captained the vessel as he had the two other ships used by the IRA to facilitate the Libyan venture. The IRA had provided two sailors to assist him, James Coll and James Doherty, both of them County Donegal trawlermen. Hopkins’s friend and sometime business partner Henry Cairns, the man suspected of having introduced Hopkins to the IRA, was along for the ride.

Cleary had spent most of his adult life in the Provisional IRA and had become one of the organization’s most skilled bomb-makers. From the Tallaght area of Dublin, a vast sprawling working-class housing estate on the southwest edge of the city, he rose in the IRA engineering department, that part of the IRA which had the job of manufacturing homemade explosives and devising the organization’s impressive range of improvised and homemade weaponry. Although well known to the Irish Special Branch, he had managed to avoid imprisonment. Only once had the authorities come near to pinning him, and that was eight years earlier, in 1977, when he beat a charge of making bombs in Kildare. By the time he was appointed to oversee the Eksund voyage, Cleary had advanced to the top of the IRA’s military elite and was in charge of its vital engineering department. He was a natural choice to head the Eksund operation.

The Eksund’s ballast tanks had already been filled with water in preparation for scuttling. Cleary had crafted Semtex bombs that were just large enough to make holes in the vessel’s skin but not so large that the noise of the explosions would attract attention. French forensic experts later calculated that the Eksund would have sunk within seconds.

The IRA man had chosen a hole known as Deep Hurd in which to scuttle the Eksund. The plan was to sink the vessel and then head in the dinghy for the Brittany coast, after which the crew would catch a ferry back to Ireland without the authorities’ ever knowing about the IRA’s audacious plan. That was when he discovered that a traitor had wrecked his plan.

A cursory glance at the bomb mechanism told Cleary that the plan would have to be scrapped; the firing unit for the explosives had been sabotaged, its wiring damaged beyond repair. The device, known as a timing power unit (TPU), was simple to operate and safe enough for a child to use, but it was just as easy to put out of commission. Whoever had neutralized Cleary’s bombs would not have needed much training.

Cleary never got as far as even connecting the device. Instead the realization of treachery forced a number of thoughts to flash through his head, as he later told IRA colleagues. The British must have known about their plans all along, and soon the media would know as well. But the question that brought a cold sweat to his brow concerned the identity of the traitor. There was certainly a collaborator on board, but was there another one, someone back in Ireland who had betrayed the Eksund and its precious cargo?

Cleary knew that the TPU must have been tampered with after the Eksund had left Tripoli harbor and not before. The IRA man had made up the mechanism himself before sailing and had linked it to detonators fixed into slabs of Semtex not long after leaving the dock. The TPU had been in perfect working order. He had double-checked to make sure.

There was no time to repair the timing unit. The spotter plane had again flown overhead, and in the distance the crew could see motor launches speeding toward them. The net was obviously closing. Cleary watched the scene with a sense of grim satisfaction; his instincts had been right.

Within minutes the Eksund was surrounded and boarded by armed French customs men. Within hours a shocked and disbelieving public in Ireland and Britain would hear the news of the failure of this extraordinary smuggling venture. A few weeks later and the full scope of the IRA’s operations would be made public; everyone would know that the organization now had some 150 tons of explosives and modern weaponry, delivered earlier and safely stored away in secret dumps throughout Ireland. But the greatest secret of them all, that the Eksund had been betrayed, was to remain sealed.

THE STORY OF the IRA’s long relationship with the Libyan regime of Colonel Qaddafi is one of the most extraordinary and colorful tales to come out of the Irish “Troubles,” and the history of their partnership is almost as old as the events that brought both into existence. Now for the first time the story of the links between the oil-rich Arab country and one of Western Europe’s bloodiest conflicts can be told in full.

Both came into being in the summer of 1969. In Ireland the events that spawned the Provisional IRA began in August that year in the small streets that crouched beneath the towers of the Catholic Clonard Monastery in West Belfast, not far from where its future leader Gerry Adams had been born and raised some twenty years earlier.

In Tripoli, a few weeks after those violent events but some 1,500 miles away, Muammar Qaddafi, a young Libyan army officer only a few years older than Adams, led a group of soldiers to power in a bloodless coup that ended centuries of colonial rule in this large North African country. Neither man knew it at the time, but fate and a shared hatred of the British were to throw them together within a few short years.

By the summer of 1969 Northern Ireland had been simmering for months as Protestant resentment at a hugely successful civil rights campaign by Catholics, fueled by years of political and economic discrimination, threatened to spill onto the streets in violence. In August the cauldron boiled over, and the death and destruction began.

It was not long before the trouble spread to Belfast, the scene for over a hundred years of often vicious and regular anti-Catholic violence. The IRA had always been there to protect its communities, but in 1969 the IRA, by this stage a small, select, and secret body that was a shadow of the army that had fought the British to a standstill in 1921, stood by almost powerless to stop the shootings and burning.

By 1969 the IRA had come under the control of a group of orthodox pro-Moscow intellectuals and activists who had spent years patiently steering the organization away from the hallowed goal of driving the British out of Ireland in favor of a campaign to reform and modernize the Northern Ireland state.

In a process that uncannily echoes modern developments, the IRA of that day was slowly putting away the gun in favor of political methods, just as Gerry Adams was to do many years later. The ambition of the IRA’s then leaders in Northern Ireland was to replace national struggle with class struggle; they had no time for Belfast’s narrow sectarianism, no sympathy for armed struggle, no need for guns.

So when mobs of pro-British loyalists surged into the streets around Clonard, the local IRA had virtually no weapons with which to beat them off. An entire row of houses known as Bombay Street was burned to the ground, and a young boy was shot dead. When the IRA split acrimoniously later that year, the Belfast men who led the breakaway Provisional IRA swore they would never leave their streets defenseless again. For their icon they chose the phoenix, the mythical firebird rising in vengeance from the ashes of Bombay Street.

Within a few years those Belfast founders of the Provisional IRA, soon to be dubbed the Provos, were to turn to that small group of Libyan army officers and their leader, Colonel Muammar Qaddafi, for supplies of the guns and explosives they needed not just to defend their own streets but to use against the British army and Royal Ulster Constabulary in a twenty-five-year campaign of violence.

Libya had a history of colonial rule that surpassed Ireland’s. It had been first seized as far back as the earliest days of the Roman Empire, which had cultivated the country’s fertile coastal strip to feed Rome’s growing population, and had been occupied by outsiders more or less continuously ever since. In the twentieth century Libya was ruled first by the Turks and then by Mussolini’s Italy, whose grip on the country was maintained with brutal, bloody force. After the Second World War and Italy’s capitulation, the allies agreed to grant Libya independence, and it was ruled by the Idris family, which led one of the country’s largest and oldest tribal clans. But real power lay in the hands of the British, Americans, and French, who divided the country into spheres of influence and maintained large military bases to protect their political and economic interests in the region, especially vast oil fields that had been discovered deep in the southern desert. Oil wealth enriched the Idris family and the Western oil corporations, but little of the windfall made its way down to the ordinary population. Resentment at Libya’s festering poverty and the revolutionary gospel of Arab nationalism then sweeping North Africa and the Middle East inspired Qaddafi and his army colleagues to overthrow the Idris family and expel the Western powers.

Qaddafi and the leaders of the Provisional IRA were natural allies. Both blamed colonialism for their country’s miserable histories, and both believed that their occupiers must be ejected from their countries, by force if necessary. One of the first acts of the Qaddafi government was to order the British, Americans, and French to close their bases and leave. It would not be long before the new Libyan leader turned to the IRA as a way of doing one of Libya’s former rulers, Britain, even more harm.

In the wake of the split in the IRA, events in Ireland moved inexorably to conflict. In 1970 the Provisionals began organizing for war and within two years had launched a shooting and bombing offensive against the British army and the Northern Ireland state. As the conflict intensified, the need for modern weaponry grew.

It was, curiously, the Breton nationalist movement in France that was responsible for putting the Provisionals and the Libyans in touch with each other. In early 1972 a meeting was held between the Bretons and the Provisionals, one of a regular series of contacts with like-minded revolutionary groups opened up by the Sinn Fein president Ruairi O Bradaigh, who was particularly keen on finding common cause with some of Europe’s small and often exotic nationalist movements, such as those of the Basques, Corsicans, and Bretons. The contacts increased political cooperation and mutual understanding, but they also facilitated the mutual acquisition of weaponry and military expertise. During the 1972 meeting the Bretons suggested that it might be worth the IRA’s while to get in touch with Qaddafi’s government in Tripoli, a regime that had let it be known that it was ready and able to assist revolutionary movements willing to foment trouble for the old imperial powers that had once ruled the Arab Middle East. That duly happened, and in August of that year the contact was formalized when two members of the IRA’s ruling body, the seven-person Army Council, Joe Cahill and Quartermaster General Denis McInerney, flew out to Warsaw to meet agents from the Libyan Intelligence Service (LIS) in the offices of the Libyan trade mission to Poland. The IRA men knew they were plowing a fertile field as two months earlier Qaddafi, eager to cause trouble for Britain, had publicly thrown his weight behind the Provisional IRA.

The Libyan leader had chosen his own state radio service to tell the world of the new alliance. “At present,” he said in an address broadcast in June of that year, “we support the revolutionaries of Ireland who oppose Britain and are motivated by Nationalism and religion. The Libyan Arab Republic has stood by the revolutionaries of Ireland, their aims and their support for the revolutionaries of Ireland.”[2]

The Libyans agreed to supply money and weapons to the IRA, as long as a suitably secure smuggling route, or “line,” as the IRA called it, could be devised. The Libyans also offered to give the IRA semi-ambassadorial status in Tripoli. The IRA leadership readily agreed and set about selecting an envoy to send to the Libyan capital, where his task would be to liaise with Libyan intelligence and help set up the arms routes.

The man chosen to represent the IRA was, at the time, not a member of the organization but a strong sympathizer. A schoolteacher from the Border town of Ballybay in County Monaghan, he had already spent a year teaching English to Libyans and was back in Ireland on vacation when he was approached by a member of the IRA’s Army Council.

The Army Council member, a veteran activist from Coalisland, County Tyrone, won the teacher over and, using his background in teaching English language as a cover, this time supposedly tutoring the sons and daughters of the Libyan army officer cadre at the prestigious Tripoli College, the IRA’s newly recruited emissary flew out to the Libyan capital.

The man, known to his Libyan handlers as Mister Eddie, found himself housed in a splendid Italianate villa in the middle of Tripoli’s embassy district. Ambassadors from other European countries would soon be aware that they had an intriguing new neighbor. The Libyans, it appeared, wanted the world to know all about Colonel Qaddafi’s new friends.

Mister Eddie was put on a generous weekly wage and found that his villa had been sumptuously furnished. No luxury was spared; the crockery, stamped with King Idris’s crest, had come from the former royal palace. Every week he would meet with officers from the LIS to discuss the IRA’s needs and drink tea from an ornate china service.[3]

What the IRA got out of its first fling with Colonel Qaddafi is a matter of dispute. What appears certain is that in the three years of that IRA-Libyan liaison over $3.5 million ($10 million in current prices) was funneled via City of London banks to the IRA’s coffers, an invaluable supplement to the IRA’s income at a time when its campaign was at its height and its most expensive.

At least one shipment of guns and explosives was intercepted, but there are strong indications that perhaps three other shipments also got through to the IRA.[4] One smuggling attempt was intercepted in March 1973 when the Irish navy captured the vessel Claudia not far from the County Wexford coast, packed with five tons of Russian-made rifles, pistols, ammunition, and explosives.

On board was a man who from then onward personified the IRA-Libyan links: the IRA’s then chief of staff and overall military commander, a West Belfast veteran, Joe Cahill. With him was the IRA’s quartermaster general, Denis McInerney, the figure who had charge of the IRA’s weapons department.

The first liaison between Libya and the Provisional IRA was, however, doomed and destined to break up in acrimony. The weak link in the relationship was to be Mister Eddie, who turned out to have more independence of mind than the IRA leadership would have liked.

In 1974 he organized a conference designed to bring the Provisionals together with the largest loyalist paramilitary group in Northern Ireland, the Ulster Defence Association (UDA), ostensibly an enemy of the republican group. The UDA had helped organize a general strike of Protestants earlier that year, which brought down a political agreement that had established a government in Belfast in which nationalists and unionists shared power and a Council of Ireland was set up, which was aimed at bringing the two parts of Ireland closer together.

According to those who knew him, Mister Eddie was intrigued by the loyalist leadership’s ideas for an independent Northern Ireland state and later became a close friend of the UDA supreme commander, Andy Tyrie. In the weeks after the general strike, Mister Eddie hit upon the idea of inviting delegations from the UDA and the IRA’s political wing, Sinn Fein, out to Tripoli along with businessmen from his native County Monaghan to discuss an economic aid package to facilitate the execution of UDA’s independence ideas.

He also had new ideas about the IRA’s relationship with the Libyans and proposed that, instead of receiving cash and guns from Qaddafi, the IRA be given exclusive rights to handle trade between Ireland and Libya. In view of Libya’s insatiable appetite for beef and Ireland’s thriving cattle industry, it was an arrangement he believed would swell the IRA’s coffers.

The problem was that Mister Eddie had arranged all this behind the backs of his IRA commanders in Ireland. When they learned of his plans, especially the invitation to the UDA leadership to visit Tripoli, they were furious.

The conference went ahead in November 1974 and was promptly leaked to the media back in Ireland. The Sinn Fein delegation refused to attend, and the UDA returned home to Belfast, triumphant and claiming that they had driven a damaging wedge between the Qaddafi regime and the IRA. The Libyans, they said, had come to see that the Northern Ireland problem was not as simple as the IRA maintained, that there was a majority in favor of the union with Britain in Northern Ireland. As result, the UDA claimed, the Libyans had said they were rethinking their relationship with the Provos.

The view that the UDA was responsible for terminating the IRA-Libya alliance has become widely accepted as the reason why Qaddafi ended his dealings with the IRA at that time. And there is no doubt that the relationship with Mister Eddie did end in dramatic fashion, when in May 1975 Libyan police picked him up from his luxurious villa and, with sirens wailing and lights flashing, drove him to the Tripoli airport, where they deposited him on a plane bound for Rome. The message to Tripoli’s diplomatic community was again clear.

But the truth about what happened was more complicated, as one IRA source explained: “The Libyans were always doubtful about [Mister Eddie] and wanted assurances that they were dealing with real Provos, which was why Cahill was on the Claudia. After the [UDA] conference the Libyans were furious with him and suspected him of being a British agent and even of having compromised our operations. They actually offered to bury him in the desert, but the Provos said leave him alone, he’s harmless but just foolish.”[5] Even so, Mister Eddie could not relax. The last words he heard from his Libyan handler as he headed for the airport were ominous: “Beware the Irish, Mister Eddie. Beware the Irish!”[6]

Mister Eddie was eventually replaced as the IRA ambassador by a more loyal and reliable figure, but the experience had soured the Libyans. The public statements of senior Libyan figures thereafter painted a confusing and erratic picture of their links with the IRA, sometimes disowning the IRA, sometimes supporting it.

The truth was that the Libyans had put the relationship on the back burner, as the same IRA source explained: “The link with Libya that went back to Joe Cahill’s day… had never been broken, although there had been a period of quiet for a long time.”[7] Libyan intelligence had been alarmed at Mister Eddie’s freelance antics, and the Qaddafi government had lost its trust in the IRA. They decided to keep the IRA at arm’s length.

The event that revived the Libya-IRA axis was the death on hunger strike in 1981 of ten prisoners from the IRA and a related republican splinter group, the Irish National Liberation Army, in the Maze prison outside Belfast. The death fast marked the culmination of a protest against the withdrawal of special category or political status, which had been granted to IRA and other paramilitary prisoners in the 1970s. The prisoners’ leader and the first to die was Bobby Sands, who had been elected a British MP before his death.

The 1981 hunger strikes received widespread international publicity and generated headlines around the world for much of the year. The Libyans watched with fascination and mounting interest. According to one senior IRA figure, “The Libyans saw the numbers coming out on the streets, the tens of thousands who came out for Bobby’s [Sands] funeral, for instance, and they thought the revolution was starting. They were also very impressed with the way the hunger strikers handled themselves. I think it had something to do with the martyrdom thing in Islam. Anyway that’s when the money flow restarted.”[8]

The two IRA men responsible for reactivating the Libyan link were both former senior figures in the organization—Joe Cahill, who had been the contact on the Claudia, and Daithi O Conaill (David O’Connell), a twenty-six-year veteran and strategist whose republican credentials were impeccable. Both men had known the Libyans since the early 1970s and were trusted by them. In late 1981, following a renewal of contacts with Libyan intelligence, Qaddafi agreed to help the IRA once again. But there were to be no arms shipments, only modest amounts of money. Over the next three years or so they sent some $1.5 million to the IRA, a fraction of their contributions in the 1970s. Qaddafi was being cautious.

The event that changed all that took place in London in April 1984 just yards from the Libyan embassy, in St. James’s Street off Hyde Park. A crowd of anti-Qaddafi dissidents from the National Front for the Salvation of Libya (NFSL) were staging a protest against the Qaddafi regime when all of a sudden a burst of automatic gunfire split the air. When the noise subsided, a twenty-five-year-old London policewoman, Yvonne Fletcher, lay dying.

The shots had been fired from within the embassy; of that there could be little doubt. British police then placed the embassy under a state of siege for eleven days, but because of its diplomatic immunity no policeman or soldier could enter the building. Eventually a deal was made, and as the siege ended thirty Libyans were driven to Heathrow Airport and a flight home to Tripoli.

Behind the shooting incident was a story of worsening relations between Libya and Western governments, especially those of Ronald Reagan’s United States and Margaret Thatcher’s Britain. The CIA, under the aggressive leadership of its director, William Casey, believed Qaddafi was sponsoring anti-American terrorism throughout the world—including hijackings of aircraft and ships and massacres in the Vienna and Rome airports. He and Reagan set out to undermine and overthrow the Libyan leader.

To Qaddafi the evidence that the West was plotting against him was obvious. A month after the killing of Constable Fletcher, anti-Qaddafi elements inside and outside Libya linked up in Tripoli and launched a fierce, but unsuccessful, military assault on the army barracks where the Libyan leader had his headquarters. The Libyans suspected that the dissident groups were being sponsored by CIA allies in Sudan, Egypt, Iraq, and Saudi Arabia and that Britain was giving them a secure base.

The evidence was strong. Four months before Fletcher’s death the Libyan ambassador to Rome had been assassinated by anti-Qaddafi elements, and the Libyan consulate in West Germany was bombed. When the Libyan embassy in London was besieged, the staff inside suspected it might be a cover for another attack and, according to Qaddafi’s own account, decided to strike first. Qaddafi had already demonstrated his willingness to hit back ruthlessly at his enemies; perhaps fifteen dissidents died in shootings ordered by the Libyan leader at this time.

Qaddafi’s war with the Western powers intensified in the mid-1980s. In 1985 the United States concluded that Qaddafi was subsidizing some thirty “insurgent, radical or terrorist groups” around the world and in June of that year decided to support exile Libyan groups in executing a strategy of sabotage, violence, and propaganda.[9] Despite an earlier White House ruling forbidding political assassination, the various U.S. intelligence agencies hoped that “disaffected elements in the military could be spurred to assassination attempts… against Gaddafi.”[10]

U.S. warplanes twice attacked Libya, once in 1981, when the United States asserted its right to send military vessels into the Gulf of Sirte, and again in March and April 1986, when Tripoli itself was targeted. Nearly eighty people were killed, one of them Qaddafi’s adopted baby daughter, Hana, who died when the planes bombed the family home. On that occasion the U.S. jets took off from British air bases with the full approval of Margaret Thatcher.

The impact of all this on Qaddafi and his view of Thatcher’s Britain need not be guessed at. “Thatcher is a murderer,” Qaddafi protested. “She allowed planes to be sent from her country knowing that they intended to attack me, to attack my home and family…. Thatcher is a prostitute. She sold herself to Reagan and now she has sold her country too.”[11]

It was strong language, soon to be matched by deeds. The IRA and Qaddafi now had a common enemy in the British prime minister. Qaddafi accused her of complicity in the death of his baby daughter. The IRA loathed Thatcher too, blaming her for callously allowing the ten hunger strikers to starve to death.

In late 1984 and early 1985 the Libyan Intelligence Service moved to put the relationship with the IRA on an entirely different and much deadlier footing, this time offering the organization much more than cash to wage its campaign.

At this stage the IRA’s ambassador to Libya was Ivor Bell, a former chief of staff and a veteran West Belfast activist who for years had been a close political and military ally of Gerry Adams. An important IRA leader in his own right, Bell, along with Adams and Martin McGuinness, had been flown to London by the Royal Air Force for secret cease-fire talks with British ministers in 1972. He was an experienced and well-respected member of the organization.

Bell was working for the IRA’s GHQ and his job was to scour Europe for weapons, but when it came to the dealings with Libya, he reported directly to the chief of staff. The story of Bell’s relationship with the Libyans is a complex one that is still a matter of argument and debate within the IRA. Bell eventually fell out with Adams, and so the account given by the IRA leaders is tainted by their need to paint him in dark colors. The version they have circulated has Bell failing to get on with his Libyan contacts and, possibly for reasons having to do with his own leadership ambitions, refusing offers of weapons and explosives, much to the surprise and anger of Libyan intelligence.

The authorized version of events credits Joe Cahill with rectifying this state of affairs and for laying the foundations for the Libyan arms shipments. Whatever the truth of that, his links with Libya were long-lasting and deep; he would claim to colleagues that over the years he had become a personal friend of the Libyan leader, and according to this account, it was Cahill who asked the Libyans to come to Ireland to discuss the most ambitious arms-smuggling enterprise in the history of Ireland’s centuries-long conflict with Britain.[12]

The person Cahill invited to Ireland for face-to-face meetings with the IRA’s ruling Army Council was Nasser Ashour, the number three man in the Libyan Intelligence Service. Ashour was well known to British intelligence from the 1984 siege of the London embassy. He was the Libyan official who negotiated the safe departure of diplomats, including the gunman who had shot Constable Fletcher.

Ashour took a risk coming to Ireland, but he traveled on a false passport and under various pretexts, once to meet Libyan students studying in Dublin; on another occasion he pretended to be a trade official on a mission to negotiate a cattle deal with the Irish government. The ruses worked, and he was able to meet the Army Council without being detected.

The deal Ashour offered was extraordinary. The Libyans would give the IRA $10 million and three hundred tons of modern sophisticated weaponry, as long as both were used against Margaret Thatcher’s government. If the arrangement worked out, Nashour hinted, there could be more money and weapons and Libya might introduce the IRA to other sympathetic governments. Nashour’s offer was immediately and gratefully accepted. Not only would the money help relieve strains on the IRA’s always overstretched budget, especially as the Royal Ulster Constabulary Special Branch was then targeting some of the organization’s more dubious ways of fundraising, but the weaponry would put the armed struggle on an altogether higher military plane.[13]

At the moment Ashour turned up knocking at its door, the IRA was in desperate need of assistance. By the mid-1980s it was experiencing great difficulties getting its hands on the sort of weapons that could put the British under serious pressure. During these years the organization had two agents roaming Europe in the search for guns from sympathetic groups and often treacherous arms dealers, but their efforts met with mixed results. Weapons came in dribs and drabs, and frequently the arms dealers double-crossed the IRA, taking the organization’s money for weapons and ammunition that was either dud or mismatched and sometimes betraying entire consignments to the Irish or British authorities.

The only reliable source for guns was, as it always has been for Irish insurgents, the United States, where large Irish-American communities, especially on the East Coast, began providing generous amounts of sympathy, money, and guns once the “Troubles” broke out in 1969.

In those early days guns were easy to come by. The IRA had a network in place in the United States in the 1950s and it was a simple task to reactivate it when violence erupted in 1969. Headed by George Harrison, an IRA veteran from County Mayo, the network was headquartered in New York, where Harrison’s links to a Mafia-associated arms dealer introduced the IRA to a steady supply of guns and ammunition.[14]

Harrison was the single most important source of weapons in these years. His consignments, many of which had been sold to his contact in the arms business by soldiers stationed at the massive marine base Camp Le Jeune, in North Carolina were running at up to 300 guns a year. One estimate suggests that he sent a million rounds of ammunition to the IRA during his career, spent over $1 million of the organization’s money, and dispatched up to 2,500 weapons across the Atlantic. A key link man to Harrison was the ubiquitous Joe Cahill, part of whose job was to collect funds raised for the IRA in the United States and hand it over to the gunrunner.

In those early days the American security agencies, especially the Federal Bureau of Investigation (FBI), paid little heed to the Irish-American communities and their political activities. It was not until the late 1970s and early 1980s that pressure from Margaret Thatcher on the Reagan White House persuaded the FBI to turn its attentions to the IRA.

Harrison’s network was the first casualty. It was rolled up in the summer of 1981 and some of its key members arrested and indicted. Although Harrison was acquitted, along with Michael Flannery, the eighty-year-old head of Irish Northern Aid (Noraid), the body that ostensibly raised money for IRA prisoners’ families, the network was lost for good. It was the first of a number of paralyzing blows. A year later the IRA’s commander in New York, Gabriel Megahey, and four other men were arrested as they tried to buy a Red-eye missile from an FBI agent posing as an arms dealer. Two years later an ambitious plan to smuggle weapons from Boston by sea foundered when a high-level informer in Southern Command, Sean O’Callaghan, betrayed it to the Garda Siochana, the Irish police.

After the destruction of the Harrison network, arms supplies to the IRA from the United States were infrequent and erratic. “There was very little stuff coming in,” recalled one veteran.[15] All too often weapons, sometime purchased over the counter in gun shops, would make their way to Ireland in twos and threes, only to be intercepted or captured by the authorities, who would then be able to trace them back and arrest and charge the sympathizers responsible. The IRA was never again able to construct a network in the United States as productive as Harrison’s.

So when the Libyan Intelligence Service offered the IRA three hundred tons of the most up-to-date military equipment with no strings attached, it was the paramilitary equivalent of a lottery win. There could be no question of the IRA’s refusing the offer, but its leaders were troubled by other issues. The smuggling operation was bound to be highly dangerous, and questions surrounded the IRA’s ability to handle such a difficult enterprise. Even if the weapons were landed safely, the IRA then faced the problem of hiding vast amounts of equipment from the prying eyes of the security forces. But the biggest uncertainty confronting the IRA leaders was how to use such an enormous windfall to the best military and political advantage.

These issues raised a range of awkward questions, and the very first one was whether or not the quartermaster’s department was up to such an ambitious task. The weapons would have to be transported hundreds of miles by sea, across the Mediterranean, along the western coast of Europe, landed in Ireland, and then hidden. It was a complex, dangerous, and time-consuming operation and was going to require careful, detailed planning.

At the time the IRA’s Quarter Master General, the man who would have the job of organizing the enterprise, was Kevin Hannaway, Gerry Adams’s cousin, but there were doubts about whether he could handle such a big challenge. After Nasser Ashour’s negotiations with the Army Council, Hannaway had, along with Joe Cahill, traveled to Tripoli to view and select the weaponry that was on offer. But he was a poor traveler, and it soon became evident he was in bad health, the legacy of sensory deprivation during interrogations by police and army carried out when internment was introduced.

The Army Council decided to replace him and instead made him adjutant general. Not long afterward Hannaway quit the IRA altogether. For the new quartermaster general the Army Council turned to Micky McKevitt, a County Louth–born activist who by 1985 was the quartermaster for Northern Command, charged with ensuring that Northern IRA units were well armed. “McKevitt was younger [than Hannaway] and was full of fire,” recalled a contemporary, “and he had a big countrywide operation going. The [Army] Council believed he had what was necessary to do the job.”

McKevitt, who was in his mid-thirties at the time, agreed to take on the job and was soon traveling to Tripoli to assess the scale of the operation and his own needs. The same IRA source recollects, “When McKevitt returned, he was staggered at what was involved. But he cut a deal with the leadership and got what he wanted. If it hadn’t been for that deal, the IRA would have lost a great deal more than the Eksund.”[16]

The Army Council agreed to a simple plan. The entire operation would be run solely by the QMG’s department, and only it would be privy to the details. The Army Council would know in general what was happening, but precise arrangements—such as the landing spots, the modus operandi, means of distributing the weapons, and so on—would be known only to McKevitt and handpicked members of the QMG’s department.

The QMG’s department was split into two. One section would continue with its routine work, arranging other arms shipments, finding hiding places for guns and explosives, and transporting weaponry northward to the Active Service Units (ASUs). Another section, hidden from the rest of the IRA, would concentrate on the Libyan operation. Chief of Staff Kevin McKenna, a taciturn, guarded figure whose base was in County Monaghan, readily agreed to McKevitt’s scheme.

Through Henry Cairns, a forty-seven-year-old bookstore owner from Bray, County Dublin, and a republican sympathizer, the IRA made contact with the man who could supply them with vessels large enough to transport the Libyan weaponry. Adrian Hopkins, then fifty-two, was a man whose business had gone bankrupt, a man who was in great need of money.

In the mid-1960s Hopkins had started a travel agency, Bray Travel, which for a year was a great success, but in 1980 it went bankrupt, stranding 6,000 Irish vacationers in the Canary Islands and costing a further 1,500 their holiday deposits. Creditors were owed some $1.5 million. The collapse hit Hopkins hard. Bray Travel had supported a comfortable middle-class South Dublin lifestyle. He was a trustee of Wicklow tennis club, refereed rugby matches, and was active in the social life of one of Ireland’s most affluent areas. He needed money badly to sustain all this.

The IRA gave Hopkins the chance to make a lot of money quickly. The deal he cut with the IRA earned him £50,000 for each shipment he organized, half paid up front, the rest when the mission had been successfully completed. He was also allowed to keep the vessels afterward and to pocket the proceeds of their sale. The operation began in a modest way. The IRA decided the wisest thing was to test its systems by sending a small consignment of weapons in the first shipment. If all went well, the next shipment would be larger, the one after that even larger, and so on until it was time for the Eksund’s cargo to arrive.[17]

Hopkins bought a sixty-five-foot Irish fishing boat, the Casamara, which he sailed to Malta in July 1985. The next three smuggling trips took more or less the same form. Each journey involved Hopkins’s sailing the vessel to Malta, where he picked up the IRA crew members. They normally made their way to Malta by circuitous routes, taking in several European capitals, notably Paris, Athens, and Belgrade, in an effort to throw off any surveillance.

Together they then sailed the boat out to sea, where they met a Libyan mother ship that transferred its weaponry to the IRA. On board the Libyan vessel was Nasser Ashour from Libyan intelligence to make sure everything went smoothly. Only the Eksund was loaded in Tripoli docks.

Each IRA ship then made its way back to Ireland, where off the east coast, at a beach called Clogga Strand, the weapons were transferred to inflatable dinghies powered by outboard engines. These ferried the Libyan arsenal to shore, where it was loaded onto trucks and then taken to dumps throughout Ireland.

The Casamara brought seven tons of arms, including Taurus automatic pistols and AK-47 rifles, on that first trip. The trip went smoothly, and so two months later the next shipment was dispatched, consisting of some ten tons of arms, including one hundred AK-47s, general-purpose machine guns, Webley revolvers, and several tons of ammunition.

Bad winter weather conditions, the small size of vessels at its disposal, and the need to maintain security meant that the IRA had a relatively small opportunity each year, between February–March and October, to bring over Qaddafi’s weapons. For that reason it was not until nine months later that the next consignment came.

Hopkins gave the Casamara the new name Kula for its July 1986 trip. Fourteen tons of weapons were shipped on that journey, including the first consignment of SAM-7 missiles. For years the IRA had sought the means to bring down the British army helicopters used to ferry patrols and supplies throughout rural areas of Northern Ireland, and now it was within its grasp.

For the next, much larger shipment, Hopkins found a bigger vessel, a former oil rig standby vessel called the Sjarmar, which Hopkins renamed the Villa. The Villa sailed in October 1986 with the largest shipment yet, a massive 105 tons of weaponry. On board were 40 general-purpose machine guns, 1,200 AK-47s, 130 Webley revolvers, over a million rounds of ammunition, 26 heavy Russian-made DHSK machine guns, RPG-7 rocket launchers with grenades, and more SAM-7 missiles. The most important item in the shipment, however, was five tons of Semtex, the highly destructive plastic explosives developed by the Czech arms industry.[18]

The stage was set to bring in the last and largest consignment, the 150 tons scheduled to be the cargo on the ill-fated Eksund.

As a result of detailed and careful planning and the failure of British or Irish security to detect the operation, the IRA had been able to get four shiploads of Libyan weapons safely into Ireland. The vital question, though, was what the IRA was going to do with them all.

British security sources estimated the value of the shipments at the time at nearly $40 million, equivalent to five times the IRA’s total annual budget. Allowing for attrition through accidental losses, usage, and security forces successes, the Villa’s cargo meant that the organization had enough weapons to keep going for a further twenty years.

The IRA now had the wherewithal to fight a really long war, as it had been committed to doing since the late 1970s. But the evidence was that even though it was not losing that struggle, it certainly wasn’t winning it. The truth was that though IRA violence by the mid-1980s was a major problem for the British it was on nowhere near the scale needed to force Britain into rethinking its presence in Northern Ireland.

The IRA leadership, represented by the seven-person Army Council, had a choice. The leadership could continue at the current, more or less containable level of violence for another two decades and hope that an unexpected event or piece of luck would transform the IRA’s fortunes. Or it could opt for something much more dramatic, a daring strike that would compel British public opinion to demand a radical change in their government’s Irish policy. It was this second option that the IRA chose.

So it was that, as the Libyan venture was being organized, the IRA set about planning a major escalation of violence, something that would jolt Britain into reconsidering its options. The plan was modeled on the Tet offensive launched by the Vietcong in January 1968, the lunar New Year in Vietnam, when guerrilla forces mounted a widespread and unexpected assault on U.S. forces throughout the country. The Tet offensive is credited with beginning the end of American involvement in that part of Southeast Asia by convincing a decisive section of U.S. public opinion that the war against North Vietnam was unwinnable. The IRA hoped to do the same with the British public.

The Army Council made a number of adjustments in its battle order to facilitate the strategy. In February 1985 the South Armagh IRA leader and cross-Border smuggler Tom “Slab” Murphy, a fixture on the Army Council for many years, was promoted to the post of director of operations and given the task of drawing up the detailed plans for the offensive. Over the next few months he traveled regularly to Libya to inspect the Libyan army’s arsenals and to assess which weapons were best suited for his plans. There was another key change. Martin McGuinness was promoted from adjutant of Northern Command to Northern commander, replacing Murphy. Although it would be his job to put Murphy’s plans into operation, McGuinness’s new job meant that he would have considerable influence over how and where the weapons were used.

The IRA Southern Command meanwhile set about making preparations for training IRA volunteers in the use of the Libyan weaponry. Sophisticated underground firing ranges, some lined with concrete and soundproofed, were constructed all over Ireland. When the Eksund was intercepted and the Irish security forces launched a nationwide search for the other shipments, a number of apparent “bunkers” were unearthed, and the assumption was that they had been constructed to store the Eksund’s cargo. In fact they were these meticulously prepared practice ranges.

Later, selected IRA operatives, perhaps as many as thirty, traveled secretly to Libya for training in specialized weapons like the SAM-7s, Ireland being judged too small to perform such training undetected. The arrival of the Libyan shipments was also staggered to dovetail into the plan. The first four consignments were dispatched to dumps in the most southerly, westerly, and northerly parts of the country, to be held in reserve for later use. “The principle was to start outwards and work in,” said one IRA source.[19] None of the Libyan weaponry was to be used, the leadership decided, until all the Libyan shipments had safely arrived.

The first shipments were to be stored in reserve and the “Tet” was to be fought with the Eksund’s cargo. Once unloaded, it was to be stored temporarily in a single bunker in Arklow, County Wicklow, and then distributed to the ASUs who would lead the offensive north of the Border and elsewhere in Britain and Europe.

The thinking behind the offensive was to cause so much damage in the first two weeks that the momentum would keep the IRA going for another eighteen months. “The idea,” said one IRA Volunteer privy to the plan, “was to take and hold areas in Armagh, Tyrone, and Fermanagh and to force the British either to use maximum force or to hold off.”[20] Four areas on the Border, security bases and posts, had been earmarked for IRA units to hold and, in theory at least, defend for days.

The SAM-7s were to be used against British helicopters, ideally cutting off South Armagh and leaving it under the effective control of the IRA. The threat against helicopters would force the British to ground their aircraft throughout Northern Ireland and to use armored ground transport, which, in rural areas especially, would be vulnerable to the heavy Russian machine guns and rockets now in the IRA’s hands.

Spectaculars were also planned: 106-millimeter canons, whose presence on the Eksund has never been acknowledged by the Irish or British authorities, were to be mounted on motorboats and used to bombard and sink the British naval patrol boat that policed the waters of Carlingford Lough dividing Northern Ireland from the Irish Republic. There were plans, too, to blow up a ship in Belfast harbor, thus blocking access to the city from the sea.

Later the campaign would be carried to military and high-prestige government targets in Britain and Europe. At one stage the IRA even contemplated launching rocket attacks on British embassies throughout Europe. But the real offensive was to be a short and very sharp affair in Northern Ireland.

The IRA leaders had calculated that the British would probably respond to their “Tet” by introducing internment without trial and would press the Republic’s authorities to follow suit. It was not the first time that the IRA had attempted to provoke the British into draconian and potentially counterproductive security measures. The organization had long believed that it thrived on British repression.

The offensive was daring and ambitious, but it suffered from a single flaw. Its success hinged on the IRA’s preserving the element of surprise—if the British ever got to hear of the IRA’s plans, all would be lost. For that reason the Army Council resisted the temptation to dip into its Libyan hoard until the Eksund had safely delivered its cargo. Not even Semtex, a relatively common explosive, was to be used, so eager was the IRA leadership to capitalize on the surprise factor.

“You were all supposed to wake up one morning, switch on the radio, and discover that mayhem had broken out everywhere,” recalled one IRA activist. “The impact was supposed to have been earth-shattering.”[21]

But whoever betrayed the Eksund robbed the IRA of a priceless asset; the surprise factor vanished the moment the French customs police boarded the vessel. Afterward Hopkins talked freely to his captors, giving them precious detail about the contents of the earlier shipments. The British soon knew exactly what weapons had been brought in, and they were able quickly to put countermeasures in place.

From there it was all downhill for the IRA. Not only had the Eksund’s precious cargo been captured and the weapons destined for the “Tet” campaign lost, but the Libyans reacted angrily to the discovery. They were particularly annoyed when they discovered that Hopkins and Cairns had not been members of the IRA. The agreement Libyan intelligence had reached with the Army Council was that only IRA members would take part in the shipments. Almost immediately after their arrest, Hopkins and Cairns confirmed Libyan involvement to French and Irish intelligence officers, much to the embarrassment of the Qaddafi regime.

The Libyan leader immediately canceled the promised cash payments to the IRA; half the promised $10 million had been paid, and the rest would have been sent once the Eksund safely made it to Ireland. Now the IRA’s coffers were suddenly empty, its ability to intensify its campaign severely curtailed by a shortage of cash.

When the IRA attempted to use some of the Libyan weaponry, it found out what the loss of the surprise factor meant. The SAM-7s had been rendered useless when the British installed electronic countermeasures on helicopters. Two SAM-7s test-fired in South Armagh whistled harmlessly past their targets. The IRA then attempted to compensate by deploying the Russian-made DHSK machine guns against helicopters, but this too was a failure. The weapons were far too heavy to be lugged around the countryside, robbing the IRA ASUs of vital speed and mobility. The British also took to flying their helicopters in groups of up to five, so that if one was attacked the others could respond. Moreover, they reinforced the armor on their vehicles to withstand the IRA’s new capabilities. The British knew the IRA was coming, and they were ready.

The only part of the shipment still of major concern to the British was the tons of Semtex explosives that the IRA’s engineering department proceeded to deploy in a series of improvised weapons. Infinitely malleable and virtually impossible to detect, Semtex was used in a variety of inventive ways ranging from coffee jar bombs and deadly mercury-tilt booby-trap devices that were attached to the underside of vehicles, to drogue bombs—small, hand-thrown parachute-guided devices that burned through armor plating.

While the security forces in both parts of Ireland were naturally concerned about the IRA’s new and staggering military strength, the IRA leadership was, as it is to this day, consumed with the search for the identity of the informer who had betrayed the Eksund and sabotaged the “Tet offensive.”

The task of unmasking the traitor was made all the more difficult by the variety of available explanations. It was possible, for example, that the British had just been good at their detective work. The Libyans left enough public clues lying around suggesting that they were once again dallying with the IRA, and it would have been seriously deficient of the various British and Irish intelligence agencies to have ignored them.

In June 1986, for instance, Qaddafi’s deputy Major Ahmed Jalloud told a group of German Euro MPs that Libya planned to resume aid to the IRA.[22] In March 1987 Qaddafi informed the Observer newspaper in London that he had increased arms supplies to the IRA in retaliation for U.S. bombing raids the preceding year.[23] The following month Qaddafi’s teenage son, Sadi, made it known at an international conference in Tripoli, to which Sinn Fein had sent two representatives, that Libya would open an office for the IRA in its capital.[24]

There was also a strong possibility that the authorities had simply guessed what was going on. A briefing paper prepared by the State Department in Washington in 1986, declassified and obtained under the Freedom of Information Act, shows Assistant Secretary J. Edward Fox noting with some perspicacity that Constable Yvonne Fletcher’s killing in 1984 “had increased [Libya’s] political motivation for supporting the IRA.”[25] Fox also pointed to another piece of evidence suggesting that Libya had resumed giving weapons to the IRA. That was the discovery by Irish police in November 1985 of arms and ammunition on a Libyan-leased airliner at the Dublin airport.

If the authorities hadn’t guessed, they should have. In January 1986 Irish police raided a farmhouse in County Sligo and discovered a large haul of weapons, sixteen semiautomatic AK-47 rifles made in East Germany and fourteen made in Romania, along with bayonets, magazines, cleaning kits, and 7,560 rounds of ammunition, that had originated in Yugoslavia. The haul was packed into six boxes stored in the attic of the farmhouse and was marked with the words “Libyan Armed Forces” and “Destination Tripoli.” In two other related raids at separate locations in Counties Roscommon and Sligo, police unearthed another eighty rifles and handguns and 12,000 bullets.

The weapons were part of the August 1985 consignment smuggled on the Kula, and were on their way to hiding places in County Donegal when they were intercepted. In an incident that was later to cast a baleful light on Martin McGuinness, a member of the quartermaster’s department in McGuinness’s hometown of Derry had betrayed the guns. The British and Irish intelligence agencies should have explored the possibility that the weapons had come from Libya, but they were apparently thrown off the scent because among the weapons were ten West German–made Heckler and Koch rifles stolen in Norway. This and the fact that their informer, Frank Hegarty, who worked for the QM’s department in Derry, had been excluded from the Libyan operation and been given a false cover story to explain the origin of the weapons meant that the British missed the Libyan connection altogether.

Although there were compelling reasons to think the British or Irish authorities had chanced on the Libyan enterprise, the IRA automatically assumed that its secret had been betrayed, and the obvious suspects were the two non-IRA men on board the Eksund, Adrian Hopkins and Henry Cairns.

Both had reasons to betray the IRA. Cairns was penniless and may have been tempted to sell his precious information for cash. But an internal IRA inquiry concluded that Hopkins was the culprit. The Eksund’s skipper had two counts against him, not least that he had stolen IRA money. According to sources familiar with the details of his arrangements with the IRA, he had overcharged the Army Council for the Eksund, swindling the organization out of tens of thousands of pounds.[26] It also became clear that he had talked freely to French and Irish police, telling them all about the previous shipments and pinpointing the Arklow bunker that had been built to hold the Eksund’s arsenal.

There were also signs that he had made a separate deal with the authorities when suddenly the French granted him bail, enabling him to flee to Ireland, where there were no extradition arrangements with France. The suspicions that a deal had been cut hardened when Hopkins came to trial in Ireland. Eleven of twelve charges against him were dropped, and he received a relatively lenient sentence, three years, compared with the five-to seven-year terms handed down in Paris to the Eksund’s other crew members.

Some in the IRA wanted to shoot Hopkins, but friends spoke up for him. The Villa had nearly sunk during a terrifying storm in the Bay of Biscay during its 1986 voyage, and but for Hopkins’s seamanship its cargo of 105 tons of arms and the IRA crew would have sunk without a trace. Doubtless conscious of the bad publicity that would attach to any effort to kill Hopkins, the IRA leadership let him off with a warning to abandon rumored plans to write a book about his exploits.

Nonetheless, the belief took hold among IRA leaders that Hopkins was the man who had put Gabriel Cleary’s TPU out of action, and that at some point in the affair he had been turned by either the Irish or the British authorities. But suspicions that the Eksund had been betrayed by another informer, a much more important figure at a high level in the IRA, persisted elsewhere in the IRA’s highest reaches long afterward, overshadowing and souring the movement’s tortuous trek on the peace process. The IRA leadership would later split and divide over the peace process, but the fault line can be traced back to the doubts and distrust generated by the betrayal of the Eksund.

In the immediate wake of the Eksund’s loss and long afterward both the IRA and the British and Irish security authorities behaved as if the informer was still around. The IRA leadership has never admitted that the vessel was betrayed and has done everything to encourage the view that its capture was a piece of bad luck. “It was kept very, very quiet,” explained one IRA source. “The Volunteers were just not told.”[27]

The IRA’s stance was understandable. To admit that an informer had gotten so close to the heart of such a vital operation would cause enormous embarrassment to the leadership, as well as the doubts at grassroots level that would inevitably grow, fester, and eventually sap morale. Saying nothing about the betrayal left the rank and file still thinking that, the loss of the Eksund aside, the Libyan venture had been a success. After all, the IRA had imported unimaginable quantities of heavy weaponry, and it was clear that the leadership had successfully managed to outwit the British for most of the Libyan enterprise. Only a very few activists would know that the loss of the Eksund had scuppered the “Tet offensive.” In the eyes of ordinary republicans and IRA activists, the Libyan operation had been a success. The 150 tons of guns and explosives meant that the war could be fought almost indefinitely.

Equally, the British and Irish authorities went to great lengths, and did so for years to come, to pretend that the capture of the Eksund was an accidental event caused by the vessel’s faulty rudder and a vigilant French customs service. At the time British army and RUC sources encouraged the media to take the view that this is exactly what happened, and they made no attempt to disguise their apparent surprise at the scale of the IRA operation. Some security force spin doctors even went so far as to suggest that the Eksund’s cargo was far too big for the IRA to handle and that it had to be sharing the consignment with another terrorist group.

A consensus emerged that was reflected in media coverage: the Eksund episode was a piece of pure luck, and the failure to detect the Libyan shipments to the IRA was a major disaster for British intelligence. One respected observer, reflecting a common media view, later wrote, “The shipments… revealed an international intelligence lapse of mammoth proportions.”[28]

Within the IRA suspicions of a high-level leak persisted, however. The problem was that Hopkins was too small a fry to deserve such an impressive level of official protection. In any case the ruse was clumsy; Hopkins was bound to be at the top of the IRA’s list of possible culprits, so why would the authorities go to such extreme lengths to protect him? Some IRA members asked another question: Would British or Irish intelligence really forgo the public credit for the Eksund’s capture, keep silent when accused by the media of incompetence, just to shield a relatively low-level agent who was the IRA’s number one suspect anyway?

These were not the only reasons for suspecting that the real informer was not Hopkins. In late 1986, not long after the IRA leadership had decided to try to ship the last 150 tons of weapons, there were two serious lapses in the security surrounding the Libyan venture. They happened when Martin McGuinness, chairman of the Army Council and the IRA’s Northern commander, sought and obtained permission from the Council to give two crucial briefings and by so doing broke the strict rule of secrecy that had been agreed and imposed on the operation.

An IRA source explained, “The agreement was that people outside the Army Council would not be told anything about the Libyan shipments until the Eksund was in, but when Slab [Murphy] and Micky [McKevitt] were out of the country on one of their mysterious trips, the Council gave the go-ahead. When they got back and heard what had happened, they were livid.”[29]

McGuinness gave the first briefing to the IRA Executive, the thirteen-member body elected by the rank and file at IRA Conventions, the occasional conferences that determine both the IRA’s politics and its military policy. The Executive chooses the members of the Army Council and selects replacements when vacancies occur; but its more important role is to act as the voice and conscience of ordinary IRA Volunteers. Most of its members were new to the job; the bulk of the previous Executive had resigned from the IRA in protest against a decision to recognize the Irish parliament in elections, and McGuinness was addressing a group whose loyalty to the Adams leadership was still untested. The split had sapped morale and McGuinness had cited the need to lift the Executive’s spirits when he sought the Army Council’s permission to tell it about the Libyan weaponry.

McGuinness was given the go-ahead, but in the circumstances the Army Council’s decision was extraordinary. By the time the briefing was arranged, the Army council had learned to be circumspect in its dealings with the Executive. There were strong suspicions that there was an informer in its ranks, and the Council had decided to carefully control what its members were told about IRA policy and decisions. The suspicions were well-founded, for in 1994 an MI5 agent on the Executive was exposed. A Sinn Fein councillor and adjutant of Southern Command, he was spared execution because of the public embarrassment to the republican leadership that would follow his exposure. In that context the decision to brief the Executive about the Libyan shipments was an astonishing lapse.

McGuinness’s briefing included the revelation not just that shipments had arrived safely but that the biggest prize of all was on its way. “He told the Executive that a lot of gear [weapons] had come in, but that the cream on the cake was still to come,” said the source.[30]

The next briefing was given in County Donegal to the small group of IRA commanders who had been chosen to lead the IRA’s Active Service Units (ASUs) into the “Tet offensive” after the Eksund had safely delivered its cargo. They were from all over the North, and their job would have been to liaise with the ASUs and outline the operations to be carried out. The briefing, also given by McGuinness, covered the quantities and type of weaponry that would be available and detailed the cargo that was supposed to come on the Eksund, the weapons that would give the “Tet offensive” its cutting edge.

Once again McGuinness was addressing a group with a link to an informer. Among those briefed was an IRA veteran from the St. James district of West Belfast called Harry Burns. His career in the IRA went way back to the start of the Troubles when he was interned on the Maidstone, a ramshackle prison ship docked in Belfast harbor which had been pressed into service when the numbers of IRA suspects being arrested far exceeded the ability of the Northern Ireland prison system to house them. Burns was trusted, but not so one of his associates. A close friend of “Big Harry,” as Burns was known to his IRA colleagues and friends, was Joe Fenton, a West Belfast real estate agent and wheeler-dealer, who turned out to be one of the most important agents ever recruited by the RUC Special Branch. Fenton was so close to Burns and trusted by him that, contrary to the organization’s strict security regulations, he would regularly drive him to supposedly secret IRA meetings.

Fenton came under suspicion when the leadership first began shipping the Libyan weapons to the Northern ASUs. A number of consignments headed for Belfast were mysteriously intercepted by the RUC, and it was obvious that an informer was at his or her work. The common link in all the losses was Fenton, who turned out to have been directly or indirectly involved in the purchase or acquisition of the vehicles used to transport weapons from Southern dumps.[31]

Eventually, in February 1989, and only after he had fled to England but then inexplicably returned, Fenton was arrested by the IRA’s security unit, the specialized team whose job it was to ferret out informers from the ranks of the IRA, interrogate them, and, if it found them guilty, shoot them dead. What happened after Fenton’s arrest has taken its place in republican legend as one of the most far-reaching and squalid scandals in the history of the Belfast IRA.

Fenton was abducted by the security unit and held for interrogation in a house in Andersonstown, West Belfast, but after only two days he was taken out and shot dead. His body, with a single bullet wound to the head, the IRA’s customary punishment for those caught informing, was dumped in an alley in nearby Lenadoon.

The speed with which Fenton was killed caused a major row within the IRA. An informer as important as Fenton should have been taken away for lengthy interrogation and debriefing so that the damage he had done could be assessed. Fenton had been working for the RUC Special Branch since at least September 1984, when in an attempt to ingratiate himself with the IRA’s Belfast Brigade (BB) and apparently on the instruction of his Special Branch handlers, he betrayed two other informers, a young married couple, Catherine and Gerard Mahon, who were shot dead by the IRA for betraying arms dumps.

Fenton had a great deal to tell his interrogators, but he was never given the opportunity. Said one angry Belfast IRA member, “Fenton was a huge fish, and the BB squandered a great opportunity to uncover a network of agents. Through his estate agency he was getting homes for people and arranging fraudulent mortgages, helping people to defraud Building Societies by manufacturing income statements for people who were registered as unemployed. The [Special] Branch would have had a field day blackmailing his clients into becoming informers.”[32]

It also emerged that Fenton had been supplying allegedly “clean” cars to members of the IRA’s England department, that section of the IRA which was responsible for carrying out bombing missions in England. He had also provided safe houses for the IRA in Belfast where meetings were held and presumably bugged by the Special Branch. Between one activity and another Fenton provided the police with vast amounts of priceless intelligence.

When questioned as to why Fenton was killed so quickly, the Belfast Brigade claimed that a huge manhunt mounted by the RUC and British army was closing in on the house where Fenton was held and that there was no time to arrange his transfer to South Armagh, where his interrogation could be carried out in a more leisurely manner. But there were suspicions that Fenton was killed for other reasons, that senior figures in the Belfast Brigade and elsewhere did not want the full story of his dealings with them to be revealed to the rest of the organization.

Fenton had performed various private favors for selected Belfast Brigade staff, one of which was to provide meeting places, usually vacant houses he was trying to sell, where secret love affairs could be consummated. One such tryst was between a former Belfast commander and the wife of one of his senior staff officers. The estate agent cum informer would give the senior IRA man the keys of a house he had on his books, and the pair would meet. It can only be presumed that the RUC Special Branch recorded their exertions and used the tapes productively. There was every possibility, in other words, that Fenton had helped the RUC to “turn” very senior Belfast Brigade figures.

There are also strong suggestions that Fenton and the IRA commander took part in a freelance, unauthorized jewelry robbery in County Fermanagh that also involved loyalists and criminal elements in Drogheda, County Louth, and Dublin. If IRA weapons had been used in such an unauthorized operation, the consequences, under the IRA’s rulebook, could have been severe for the former commander. Although rumors and allegations about the commander’s activities were rife in the Belfast IRA at this time, nothing was ever done. The former IRA commander had powerful family connections in the republican movement. That his brother sat on the Army Council and was a senior member of Sinn Fein may have saved his life.

Whatever the reason, Fenton was condemned to an early death, and the IRA’s security unit never got the chance to properly question the informer about his long and eventful career as a Special Branch agent. Nor was it able to quiz him about his knowledge of the Libyan shipments and in particular whether or not “Big Harry” Burns had let slip any of the contents of the briefing he and the other IRA specialists had been given by the Army Council figure. Burns always denied he had said anything to Fenton. Burns, who had been severely injured some years beforehand when a mortar bomb he was attempting to fire exploded prematurely, died of cancer in February 1999. Ironically he died a fervent supporter of Gerry Adams’s peace strategy.

But afterward, when the dust thrown up by Joe Fenton’s killing had settled, the IRA’s security unit conducted a more thorough investigation of the informer’s activities and found a sophisticated bug hidden in a light fitting in the front room of Burns’s house. Someone in the world of British intelligence was very interested in what Harry Burns was up to.

The net effect of the two briefings was that when the Eksund was betrayed, the suspicions about the identity of the traitor were more generalized. Since the circle of knowledge about the Libyan shipments had been widened beyond the Army Council, so too had the range of candidates for the role of informer. In fact, by mid-1987 the circle of knowledge had expanded to include a group of activists sent to Libya for training in the weapons as well as a small number of key personnel whose support for major political shifts was needed by the Adams camp.

The special briefings nonetheless sowed distrust in the IRA’s higher reaches. But they were as nothing compared with the subsequent suspicion that the real culprit responsible for giving up the Eksund and with it the IRA’s war plans may have been at the very top of the organization.

“The October [1987] trip was not the first time the IRA tried to bring in the Eksund’s cargo,” said one knowledgeable IRA source. “The original plan had been to bring it in all in one go sometime in April or May, in the spring of ’87. But in February that year the IRA got information that Free State army units had been put on standby from Cork to Carlingford in anticipation of an arms ship coming in sometime in the coming weeks. Someone had leaked the operation. The Free Staters didn’t know where it would land, but they knew something was up. The IRA had no choice but to postpone the operation.”[33]

When the Eksund was captured, it therefore came as no surprise to the IRA leadership. As a result of the February leak, it was half expecting the venture to fail.

The question then became this. If the Irish government had a rough date for the spring 1987 plan but no precise intelligence about where the cargo would be landed, who in the IRA was in possession of just that level of information? The answer appeared to rule out Adrian Hopkins as the main suspect. He knew exactly where the planned shipment was to be off-loaded—at Clogga Strand on the Wicklow coast, where the other four shipments had been unloaded and where the Eksund was headed some five or six months later. If Hopkins had been the informer in February 1987, the Irish authorities would have had no need to put their forces on alert along the entire eastern coast; they would have known exactly where to go.

These claims appear to exonerate Hopkins, although he was still suspected of being the man who tampered with Gabriel Cleary’s timing power unit. “Some believe that [it was only] after the February leak [that] Hopkins was turned by the Brits,” said one IRA source.[34]

The finger of suspicion for the source of the February leak, the main informer, appears to point elsewhere, to someone whose knowledge of the February plan was much less specific, whose information, although priceless, could not pinpoint the cargo’s destination.

Among those who knew that the last cargo was on its way but were unaware of the details of the operation were those members of the Army Council not involved directly in organizing the Libyan shipments. The Council members’ state of knowledge was this: they were aware that a boat was moored in Malta awaiting final arrangements to bring the last shipment to Ireland and that it was due sometime in the late spring. They needed to know that in order to make plans for the “Tet offensive.” But they were unaware of vital details, such as the precise destination of the cargo, the spot where IRA members would unload it.

Ironically the Eksund’s fate was sealed by an uncharacteristic act of recklessness urged on their Army Council colleagues by Gerry Adams and Martin McGuinness. Against the advice of even those most intimately involved in planning the Libyan adventure, both men insisted that the Eksund’s deadly cargo be transported to Ireland as quickly as possible, and in one shipment, so that the “Tet offensive” could go ahead on schedule. Others on the Council counselled that a wiser course would be to transport only a small part of the Eksund’s payload to Ireland. If it got through safely, then the IRA could be confident that there had been no serious infiltration and the Eksund could set sail as planned. If it was intercepted then they would know for certain there was a traitor in their midst, while the loss of weaponry to the IRA would be minimal. Alternative plans could then be drawn up to bring the rest of Eksund’s cargo to Ireland at another time and by a different way. But both Adams, whose customary caution in all matters was legendary, and McGuinness argued successfully that the IRA’s urgent need to launch the big military push demanded otherwise and that the Eksund’s voyage should proceed as planned. They got their way, but had they not prevailed then the story of the IRA’s war against the British— and with it the peace process—might have been very different indeed.

The question of who betrayed the Eksund has never been satisfactorily settled, but suspicions that there was—and possibly still is—a high-ranking traitor in the IRA have nevertheless festered for years, poisoning relations and fueling distrust as the peace process gathered pace.

THE YEAR 1987 ended on a disastrous note for the IRA. In November, only days after the capture of the Eksund, an IRA bomb exploded at the cenotaph in Enniskillen, County Fermanagh, as local Protestants gathered to remember their war dead. Eleven were killed and over sixty injured. The bomb helped ratchet up public hostility to the organization in the Republic to unprecedented levels.

In the wake of the Eksund’s capture, the Irish government ordered a nationwide search, code-named Operation Mallard, for the four other shipments. Some sixty thousand homes and farms were raided and searched. Although none of the weaponry was discovered, the IRA nevertheless suffered several other serious setbacks. The officer in charge of training ASUs in the use of the Libyan arms was captured, and documents unearthed during the search led the FBI to a key IRA operative in the United States, Richard Johnson, a skilled Boston-based electronics engineer with federal security clearance who was helping the IRA develop homemade surface-to-air missiles.

The mood of the IRA as the year turned was mixed. Recalled one activist:

Despite the Eksund and [the] Enniskillen [bombing] the mood of the rank and file was quite good. They knew, thanks to Hopkins, what had come in—they were especially delighted to get the AK-47s—and they were upbeat, even jubilant. Don’t forget, for years the cry had been to get heavy gear in, and here it had happened. From mixing with the leadership, however, you could see things were very different there. The mood was much more somber. They knew they had lost the vital element of surprise, and it was back to the drawing board.[35]

The betrayal of the Eksund condemned the IRA to military stalemate with the British. The successful Libyan shipments certainly made the IRA a more dangerous enemy than it had been for years, dangerous enough eventually to persuade the British that talking to the IRA might be more productive. But the chance of securing a decisive military advantage over the British—the aim and purpose of the “Tet offensive”—had been lost forever.

It was in such an atmosphere that the idea that politics might be an acceptable, even unavoidable, alternative to armed struggle took hold and was nurtured. When Gabriel Cleary inspected the sabotaged firing unit on the bridge of the Eksund and realized that its precious cargo was doomed, he was not to know that the spy who had betrayed his mission had also boosted another secret operation then under way, an operation that not even the Army Council knew about but which the world would soon know as the Irish peace process.

PART ONE
The Dogs of War


British troops take cover from an IRA bomb attack in Belfast in the early 1970s. (Pacemaker Press, Belfast)


1. Roots

The Northern Ireland state created by the Anglo-Irish war of 1919– 21 and the subsequent settlement agreed by the British prime minister, Lloyd George, and Irish republican leaders was not much more than a quarter of a century old when, on October 6, 1948, Annie Adams gave birth to her first son and christened him Gerry after her husband. The centuries-old struggle for Irish independence had burst into violent guerrilla warfare between the IRA and the British just as the First World War ended, and although the rebellion was widespread and popular in a way unmatched in Irish history, it was only partially successful. When the two sides, exhausted by their bloody efforts, finally agreed to sit down and negotiate a settlement, the deal that emerged, the Treaty, as it would forever be known, eventually gave most of Ireland—twenty-six of its thirty-two counties—political freedom from Britain. But six counties, Northern Ireland, stayed British at the insistence of their large Protestant and unionist majority. By October 1948 Northern Ireland was enjoying peace, albeit an uneasy one. But trapped inside the state into which Gerry Adams was born was a significant Catholic and nationalist minority, a third of Northern Ireland’s one and a half million people, whose oppressive treatment at the hands of the unionists ensured that there would always be a role for the IRA and an audience for its seditious gospel.

To say, however, that violent republicanism was the predominant sentiment among Northern Ireland’s Catholics would be wrong. By far the bulk of them supported constitutionalist politicians, principally the conservative and strongly pro–Catholic Church Nationalist Party. Support for the IRA was a minority activity; membership was even more so. Nevertheless there was a republican tradition in Northern Ireland, and by the standards of early postwar Belfast the Adams family were in its blue-blood line. It was this political lineage that would ordain what the newborn Gerry Adams would do in life. As the infant Gerry grew up, he was surrounded by relatives who had fought, been jailed, and, in the case of his father, even shot for the cause of Irish freedom. The IRA—its traditions, history, and values—was imbibed with his mother’s milk.

His paternal grandfather, also called Gerry, had been in the highly secretive Irish Republican Brotherhood (IRB), the forerunner of the IRA commanded by the legendary guerrilla leader Michael Collins, during the Anglo-Irish war. His father, Gerry Adams Sr., joined the IRA as a sixteen-year-old and in 1942 was sentenced to an eight-year jail term after an IRA ambush of members of the predominantly Protestant police force, the Royal Ulster Constabulary (RUC), went badly wrong and he was felled by three bullets. Two paternal uncles, Dominic and Patrick, had been interned without trial because of their IRA sympathies, one by the government in Dublin, the other by its Unionist counterpart in Belfast. For a short period prior to the Second World War, Dominic Adams had held the highest rank in the IRA, that of chief of staff, a position his nephew was later to occupy, also briefly but with much more effect.

His mother’s family, the Hannaways, had a similar history. His maternal great-grandfather, Michael Hannaway, had been a member of the Fenian movement, which bombed England in the 1860s and 1870s. His grandfather, Billy Hannaway, was Eamon de Valera’s election agent when the hero of the 1916 Easter Rising ran for election in West Belfast in 1918. Later he broke with him when de Valera became a constitutional politician.

Adams’s mother was a member of the women’s branch of the IRA, the Cumann na mBan;[1] she was a “staunch republican,” her son was to write many years later.[2] Three of her brothers—Tommy, Liam, and Alfie—were IRA stalwarts in the city. Uncle Liam Hannaway was to play a crucial part in steering the young Gerry Adams toward the Provisional IRA, while Uncle Alfie, a leading light all his life in the IRA’s boy scouts movement, the na Fianna Eireann, was a daily communicant at Clonard Monastery, in the heart of West Belfast, helping to establish a relationship between the Adams family and Clonard’s Redemptorist priests which proved pivotal many years later when the Irish peace process began.

The Adams and Hannaway families may have been seen as republican aristocrats, but that was only among their own small rebellious circle. In the wider world they were members of a beleaguered minority that had more often tasted isolation and defeat for their dedication to Ireland’s cause. In the Northern Ireland of the 1940s and 1950s very few Catholics were in the IRA or indeed knew much about it. Many were deterred from having anything to do with it by the knowledge that any display of IRA sympathies would be sure to invite the hostile attention of the fledgling unionist government’s security agencies, in particular the RUC Special Branch, a detective force that specialized in monitoring political opponents of the Protestant-dominated state.

A draconian law, called the Special Powers Act (SPA), gave the authorities exceptional powers to arrest, detain without trial, and suppress political dissent. So severe were its penalties, which included the death penalty for some firearms offenses, flogging, and the confiscation and destruction of property, that a South African prime minister during the apartheid era once famously remarked that he would swap all his emergency laws for one clause of the SPA.[3] On top of conventional forces, the first unionist government had created an armed paramilitary force, known as the B Specials, manned by thousands of pro-British Protestant supporters, which could be mobilized in emergencies to put down armed revolt. In peacetimes they kept a wary eye on their Catholic neighbors.

Unionism was an ideology that thrived on a sense of siege. Even the creation of a state whose gerrymandered border guaranteed unionists virtually permanent majority rule could not overcome a deep political psychosis. Fear of retribution from their downtrodden and disenfranchised Irish Catholic neighbors was possibly the most potent single factor in their political makeup. It had haunted the Protestants since the sixteenth and seventeenth centuries when English monarchs like Elizabeth I, anxious to create a loyal buffer in rebellious Catholic Ireland, confiscated native-held land in the northeastern part of the island nearest to Britain, in the province of Ulster, and gave it to trustworthy, loyal Protestant planters imported from Scotland and England. Deep inside, many of the planters were terrified at the thought that one day the native Irish would take their revenge and their land back, a communal dread that has survived the centuries.

The idea behind the plantation of Ulster was to make invasion of Protestant England by Catholic France or Spain via Ireland that much more difficult. It was the first expression by England that the occupation and colonization of Ireland served to advance its wider strategic and military interests. But serving England’s interests created a poisonous mix of sectarian and political division that was to shape and deform Irish politics for centuries.

Even in the 1940s and 1950s, when Gerry Adams was growing up, many unionists still looked upon their Catholic fellow countrymen in much the same way as white settlers in the nineteenth-and early twentieth-century American West had viewed the Sioux or Apache or the South African Boer his Bantu servants—that is, with a mixture of fear, guilt, ignorance, and hatred.

The actual threat that the IRA posed to the new state in Northern Ireland was more debatable, especially when the appetite for militant policies south of the Border waned and with it sympathy for the Northern nationalists’ plight. Once the 1921 partition settlement had taken root, and especially as Catholic Church influence on the infant Irish Free State tightened, the commitment of Southern republicans to dismantling the Border and freeing Ireland became increasingly rhetorical. As the years passed and they developed their own set of economic and social priorities, the two states in Ireland gradually drifted apart and there was a consequent decline in Southern enthusiasm for the IRA’s aims.

A key event in this journey was de Valera’s decision in 1926, just five years after the Treaty had been signed, to end the IRA’s armed struggle and embrace parliamentary politics. De Valera’s supporters had opposed the Treaty and fought a bitter civil war against Michael Collins’s forces but had been soundly beaten. Eventually de Valera persuaded most of the IRA to abandon violence—as Gerry Adams was later to do in his day—brought them into his new Fianna Fail party, and eventually entered government. A die-hard rump, from which the modern IRA sprang, refused to compromise and, accusing de Valera of betraying the freedom struggle, vowed to continue the fight. In the North, nationalists and republicans increasingly felt abandoned.

Although Fianna Fail discarded objectionable aspects of the 1921 Treaty settlement, such as the oath of allegiance to the British monarch, de Valera eventually took as tough a line against the recalcitrant IRA as any of his predecessors. The IRA was often divided and at times spectacularly incompetent, but Northern unionists liked to imagine that the IRA was a much more potent threat, that it was an organization swollen with thousands of members, armed to the teeth, just waiting for the word to attack. The truth was much more prosaic. At the time of the arrest of Gerry Adams’s father, for instance, the IRA in Belfast could barely muster three hundred members and was poorly armed.[4] Even so, the unionist government did not hesitate to use its formidable powers at any sign or hint of a threat. Sometimes, such as during the “Hungry Thirties,” when unemployment rates soared as high as 25 percent, the unionist government found the IRA specter a successful way of scaring its supporters away from cross-community, left-wing politics. Either way, life was made uncomfortable for those Catholics tempted to dabble in the IRA. Republican suspects were regularly arrested and occasionally interned, and detailed records kept of likely sympathizers. Parades, marches, and other expressions of support for the IRA were often banned and the organizers pursued.

Gerry Adams’s father became a victim of this system in 1942 after a ban was slapped on the republican parade to Milltown cemetery in West Belfast, an annual event to remember the IRA dead of the 1916 Easter Rising in Dublin. An effort by the IRA to divert police away from the cemetery so that a brief procession could take place failed, and there was a short, sharp exchange of gunfire between the IRA men and the RUC during which one policeman was killed. The six-man IRA unit was captured and sentenced to death, but after pleas and protests, including an intervention from Pope Pius XII, only one of their number, Tom Williams, the officer commanding the unit, was hanged. When the IRA decided to protest against Williams’s execution in September that year with further attacks, Gerry Adams Sr. was shot and arrested.

Those who joined the IRA faced other pressures. If they were caught and convicted, a record of their imprisonment would be stamped on their employment records, with the result that republicans often found it difficult to get work even after they had severed all links with the organization. Gerry Adams Sr. was denied entry to Australia with his family because of his prison record. Others discovered they had been declared “politically suspect” even though they had done nothing wrong.[5] Many were forced to emigrate to England or the United States or to move south to Dublin in search of a living.

Not surprisingly, involvement in the IRA was confined to a few. Nationalists in Northern Ireland may have secretly supported the aims for which the IRA fought even if they had qualms about its methods, but very few would go so far as to join. The risks and potential burdens were too great.

The result was that republican involvement tended to be an inherited rather than an acquired activity. Gerry Adams’s background was a classic of its type. His parents, like those of many other republicans of this time, would pass on to their children their political views as well as a special, exclusive sense of shared suffering. The IRA in places like West Belfast, where the Adamses and Hannaways came from, grew heavily dependent on a small, often interrelated network of extended families.

West Belfast republicanism was dominated by three families: the Adamses, the Hannaways, and the Burnses. They were all intermarried, the consequence of the imprisonment of their male members. When figures like Gerry Adams Sr. emerged after having served their jail terms, they found girls of a marriageable age either already spoken for or reluctant to marry into the IRA. Inevitably they drifted into relationships with the sisters of their IRA comrades.

The family was always an iconic and powerful image for the Provisional IRA. Years later when the IRA’s war with Britain raged, the organization’s leaders would routinely refer to their supporters and members as “the Republican family.” Later during the peace process, when Sinn Fein organized gatherings to brief supporters on the latest developments, these were called “family meetings.” The image that was evoked, one of benevolent parents nurturing loving, obedient, but united children, was deliberate. And like children, the IRA’s supporters were not expected to ask too many awkward questions or disobey their elders; the leadership treated those who did in much the same way as a wayward son or daughter who had offended his or her parents.

One characteristic that the young Gerry Adams did share with his non-republican Catholic contemporaries was poverty, deprivation, and the consequences of state-sponsored anti-Catholic bigotry. In the one-party state that was Northern Ireland, Catholics routinely found themselves the victims of economic and social discrimination.

Anti-Catholicism was built into the state ideology and promoted by its leaders. On occasions unionist government ministers would urge their supporters to employ, wherever possible, only “good Protestant lads and lassies.”[6] One prime minister, James Craig, famously described the parliament at Stormont, an extravagant neoclassical pile set on a hillside in East Belfast, from where he ruled Northern Ireland, as “a Protestant Parliament for a Protestant state.”[7] There was little or no room in the new Northern Ireland state for Catholics.

The message was reinforced by occasional sectarian violence. Riots, burning, shootings, and bombings—carried out mostly by Protestant mobs—had been a regular feature of political life in the north of Ireland since the mid-nineteenth century, when Irish nationalists first began to agitate for Home Rule and a degree of separation from Britain.

In 1912 the unionists rebelled against the British when a Liberal government at Westminster threatened to grant Ireland Home Rule. Protestant leaders recruited and, with the support of British Conservative leaders in London, organized a private army that they called the Ulster Volunteer Force and threatened to resist Home Rule by armed force facilitated by thousands of rifles smuggled into the Ulster port of Larne from Germany. The first Irish paramilitary group, the first effort to import weaponry from Britain’s enemies, came not from Irish republicans but from people who loudly proclaimed their loyalty to Britain.

The Anglo-Irish war of 1919–21 brought fresh communal violence, as did the Treaty settlement when unionists faced the problem of stabilizing their new state. Previous manifestations of nationalist rebellion, such as the bids for Irish Home Rule in the 1880s, had been met with often terrible violence, and now that Northern Ireland’s political leaders were presented with the problem of constructing a political order that was opposed by up to a third of its citizens, they and their supporters in organizations such as the Orange Order turned to old, reliable methods. The early 1920s saw scores killed in riots, gun battles, and burnings; in the early 1930s violence erupted again. Catholics made up a disproportionate number of the fatalities.

Faced on the one hand by official state forces that regarded them as hostile and on the other by irregular Protestant mobs that often went on the rampage while the RUC and B Specials turned a collective blind eye, Catholics inevitably came to look on the IRA as a defensive force first and foremost.

The Northern Ireland that Gerry Adams was born into was a society in which most Catholics were at the bottom of the heap, at best tolerated, at worst regarded as a fifth column intent on undermining the state. The best-paid and most skilled jobs, such as those in the Belfast shipyards where the Titanic was built or in engineering factories like Shorts, went mostly to Protestants.

Years later, in the 1970s and 1980s, when official statistics were first recorded, this pattern of discrimination was confirmed. Catholics were found to be at least three times more likely to be unemployed than Protestants and disproportionately represented in the poorest-paid, least-skilled, and most insecure jobs.

To buttress their economic domination, Protestants banded together in a semisecret society known as the Orange Order. Founded in the eighteenth century by the Anglo-Irish middle classes, the Orange Order saw its primary role as resisting Catholic and radical Protestant demands for independence that were modeled on the French and American revolutions.

From the mid-nineteenth century onward industrialization transformed the north of Ireland and politics on the island as a whole. The desire to retain access to British markets fueled Protestant resistance to Irish independence, while the flood of rural Catholics into Belfast attracted by the new work opportunities brought competition with Protestants, and sectarian tensions rose.

As unionism developed into a coherent political ideology, the Orange Order, whose ranks were open only to those who had been born Protestant, acted as an umbrella under whose generous frame factory boss and factory worker could both find shelter. The order’s ranks swelled. Orangeism became an instrument of sectarian division and privilege. By the end of the 1940s, when Gerry Adams was taking his first, faltering steps, no unionist politician could aspire to elected office if he or she was not a member. The prime minister and all his cabinet were usually Orangemen. Huge parades of Orangemen were held annually on the Twelfth of July. These would see tens of thousands of men wearing bowler hats and orange sashes marching in formation in Belfast and elsewhere behind military-style bands to celebrate the victory at the Battle of the Boyne by the Protestant King William over the Catholic King James in 1690. The Williamite triumph was a major event in British and European history, but by the middle of the twentieth century these annual celebrations had become archaic and incomprehensible to the outside world. They were, though, demonstrations of Protestant domination, designed to remind Catholics of their subordinate place in the political, social, and economic order.

Gerry Adams’s parents were not untypical of many Catholics of the day. His mother, Annie, was a doffer, replacing spools of thread in one of the dozens of linen mills that dotted West Belfast, while his father, when he was not unemployed, was an unskilled building laborer. Although the Catholic working class was large, for the first thirty to forty years of partition the Catholic middle class was inconsequential, confined to a few thousand schoolteachers, bar owners, and lawyers who serviced mostly their own communities.

The Catholic clergy held a disproportionate sway over their flock. Once the unionist government handed over control of non-state schools to the Irish hierarchy, the Catholic Church’s interest in the status quo became almost as strong as that of the unionist cabinet. Most Catholics, including the young Gerry Adams and most of his contemporaries, had little chance of rising much above their appointed place in life.

What prospect Gerry Adams did have of upward economic mobility came courtesy not of the Northern Ireland government or the Catholic Church but was due to huge political changes across the Irish Sea in postwar Britain. At the end of the war with Nazi Germany, Britain’s voters turned against Winston Churchill and the Conservative Party with such determination and numbers that what happened approached a social revolution.

The Labour Party was swept into power on a platform of social equality that included the nationalization of railways, utilities, heavy industry, and coal and the provision of a free national health service. Arguably the most radical social measure was in the field of education, where college education was thrown open to working-class children. Those who were bright enough and could pass an intelligence test when they were eleven years old were streamed into an academic education. For the first time ability dictated how far children could go.

No other factor was more responsible for causing the Troubles. The Eleven Plus, as the exam became known, opened the door to the middle classes and introduced a significant element of social mobility into British society. This was also the case in Northern Ireland, but the state’s Catholic population encountered an extra obstacle, the built-in systems of discrimination that had been constructed to preserve unionist privilege. As more Catholics obtained a college education, their economic, their social, and ultimately their political expectations soared. It was unionism’s refusal and inability to satisfy these expectations that finally unplugged the Northern Ireland volcano.

Gerry Adams’s early life, like that of his peers, was a tough one. After a brief period living with his maternal grandmother in the Falls Road in West Belfast, the family moved to the northern outskirts of the city, to Greencastle on the picturesque shores of Belfast Lough, where they rented a one-room flat. The family grew, each year bringing a new addition and added strain on living space and financial resources.

The family, which eventually numbered five boys and five girls, hankered to be back in West Belfast, where the rest of the extended Adams and Hannaway clans lived, and so in the early 1950s, and largely thanks to his mother’s efforts, Gerry Adams made a move that was to have significant consequences for Irish history.

At the end of the Second World War, Belfast had a huge homeless problem. German bombers had twice raided the city but missed their economic and military targets and devastated inner-city housing estates instead. The unionist government was forced to build public housing projects, and it was into one of these, a sprawling estate known as Ballymurphy on the slopes of the hills overlooking West Belfast, that the Adams family moved.

Home at 11 Divismore Park, Ballymurphy, gave the growing Gerry Adams much-needed social stability, comfort, and welcome contact with family as well as a new circle of friends. But it was also to provide him less than two decades later with a base that he would use first to dominate the IRA in West Belfast, next the city, and then the entire organization. If Annie Adams had not insisted on making the move to Ballymurphy, the IRA might never have been led by Gerry Adams, and Irish history would now look very different.

His parents were conscious of the opportunities offered by the Eleven Plus, and they encouraged their eldest son to sit for the exam. At the second attempt he passed and was granted a place at the boys-only St. Mary’s Christian Brothers Grammar School, situated on what was then the more affluent fringe of West Belfast. Most St. Mary’s boys were expected to stay at school until the age of eighteen and would be encouraged to go on to college or enter a profession. The Adams family’s hopes for Gerry Jr. were naturally high. Family photographs of the day show a gangly, if not awkward, teenager whose thick spectacle frames reinforced a bookish image.

But the family’s hopes were to be dashed. By 1964 he was flirting with republican politics and had helped campaign for the Sinn Fein candidate during an election in West Belfast, the highlight of which was two days of stone-throwing between RUC riot squads and local nationalists and republicans, including the adolescent Gerry Adams.

Adams’s studies suffered, and by Easter 1964 he was at the bottom of the class.[8] With many contemporaries in Ballymurphy already working and a growing number of mouths to feed at home, he left St. Mary’s in early 1965 and took a job as a bartender, first in a Catholic-owned pub on the Loyalist Shankill Road and then in one of Belfast’s most famous hostelries, the Duke of York, home at the time to what passed for the city’s artistic and left-wing intelligentsia.

A year later, when he reached eighteen, Adams’s career in the IRA began. He was sworn in to D Company (D Coy) of the Belfast Brigade, the unit his father had belonged to and whose members came from the Falls Road and West Belfast.[9] Within six years D Coy was to be known as one of the most ferocious and active units in the emerging Provisional IRA, earning its members the nickname “the Dogs,” after “the dogs of war.”

Adams had chosen a period of tumult within the Republican movement to join the IRA. Always prone to ideological, military, and personality disputes, the IRA was again on the verge of a bitter and bloody split when the Belfast officer commanding took Gerry Adams through the IRA oath and welcomed him into D Coy.

The history of the IRA both before and after it tasted defeat in the 1921– 23 Irish civil war is essentially the story of military failure followed by retreat and introspection. After each reverse the remnants of the IRA would divide into two camps, those who retreated to their firesides and dreamed of better days to come and a second chance to take up the gun and bomb, and those who advocated a new and radical change of direction. Invariably that would involve advocating heresies, usually that the IRA should ditch its almost mystical distaste for parliamentary politics, and an internal row, possibly a split, would follow.

Virtually every twentieth-century republican leader had trodden this path. Michael Collins walked down it, and Eamon de Valera did as well. So too would Gerry Adams, although in circumstances that were to make the efforts of Collins and de Valera appear amateurish and clumsy.

After the 1921 Treaty the IRA divided into those who stayed loyal to Michael Collins and those who supported Eamon de Valera, and a bitter civil war followed. The split was not caused by the partition of Ireland. The Treaty had set up the Boundary Commission to draw the borders of the new Northern Ireland state, and even anti-Treatyites firmly believed that when nationalist areas were removed from the six partitioned counties, as the British had implied during the Treaty negotiations, the truncated remnant would not be viable and the new state would collapse into their hands.

The Treaty had imposed on members of the new Irish parliament and the “Free State” government an obligation to swear an oath of allegiance to the British crown, and it was this that divided the IRA. Collins’s men argued that the oath did not matter. Ireland had secured the freedom to achieve freedom, bit by bit, county by county, and that was all that mattered. Predictably his supporters became known as stepping-stoners. De Valera eventually agreed to sign the oath but claimed this was not the same as swearing it, an elasticity of attitude to such matters that Gerry Adams was to imitate and incorporate wholesale into his own peace process strategy.

The resulting Irish civil war was an unequal battle. Armed by Britain and facing an opposition that hesitated to strike the first decisive blow, Collins’s “Staters” put the IRA on the defensive almost from the beginning. The war was over by 1923. The IRA leadership, on de Valera’s urging, ordered its members to dump arms.

Three years later “Dev” abandoned military methods. He resigned from Sinn Fein and announced the setting up of a constitutional republican party known as Fianna Fail, or Soldiers of Destiny, a name that was chosen to appeal to the militarist tradition from which the party sprang.

The vast bulk of the IRA followed de Valera, and IRA units were transformed almost overnight into Fianna Fail branches, or cumainn. Those who rejected Fianna Fail did not like what they saw but were confused about what to do. Some were content to wait and see whether or not de Valera did deliver on the republican rhetoric, especially after election victories brought the party to power. Others, like Peadar O’Donnell, argued that republicans should eschew establishment politics, move to the left, and take up radical social and economic policies. Many IRA men who thought as O’Donnell did later went to Spain to fight against Franco.

At first de Valera welcomed IRA support, not least because the civil war had left deep divisions in Irish society. Trust was hard to find, old civil war enmities simmered just beneath the surface, and Fianna Fail needed friends wherever it could get them. But the alliance was to be short-lived. In 1932 de Valera won enough seats to form a coalition government, and within four years he moved against the IRA and declared it an illegal organization.

By this stage the IRA was once again turning its attention to the older enemy. The British had reneged on promises made in 1921 when the Boundary Commission was set up. In a majority report the commission brushed aside nationalist concerns and recommended that all six partitioned counties be incorporated in the new Northern Ireland state, and the fledgling administration in Dublin had little option but to acquiesce. The decision made the new entity a viable one but at the cost of sowing the seeds of future conflict. Nearly half a million Catholics and nationalists, a third of the population, had been forced against their will into a state with which they did not identify and whose leaders were openly hostile to them.

Nationalist Ireland was unsure about what tactics to adopt. De Valera’s answer was to turn up the rhetorical heat. He drafted a new constitution in 1937 which set the goal of reuniting Ireland in legal stone. Physical-force republicans advocated a more traditional approach and urged renewed war against the British but not against de Valera. The IRA ended its conflict with the Southern state. It was a seminal development because it started a process that eventually led to the IRA’s formally recognizing the Southern state and then participating in its institutions. From then on, the main goal of the IRA was to get the British out of the North rather than to eject the impostors in Dublin.

By 1939 the IRA felt confident enough to declare war against Britain, and under the leadership of Sean Russell, a veteran of the 1916 Rising and an opponent of O’Donnell’s socialism, a bombing campaign was launched. As republicans had done before the 1916 Rising, lines were opened with Germany, whose Nazi leaders were themselves at war with Britain. Although the IRA hoped for all sorts of assistance, little came of the relationship.

The Forties Campaign, as the IRA’s war came to be called, forced its own split. Sean MacBride, son of the legendary Maude Gonne MacBride and himself a former chief of staff, broke with Russell. When the Second World War ended, the future Nobel laureate quit the IRA and formed his own political party, Clann na Poblachta, which enjoyed considerable, but brief, electoral and political success.

The IRA was the author of its own defeat in the Forties Campaign. In the sort of botched operation that would play such a crucial role in the modern peace process, a bombing in Coventry in the English Midlands at the start of the campaign went badly wrong, and five civilians were killed and another sixty wounded. At around the same time the IRA in Dublin raided the Irish army’s weapons reserves and, much to its own surprise, netted a dozen truckloads of guns and one million rounds of ammunition, most of which the IRA promptly lost when the police discovered their hiding place.

The Coventry debacle roused the English police, which used harsh methods against IRA suspects, while the Dublin arms raid permitted de Valera the political space to seek emergency powers, which he used to intern IRA leaders. De Valera was concerned that the IRA’s attacks on Britain and its overtures to Hitler’s Nazis could give the British the excuse to force him to take Britain’s side in the “Emergency,” as the Irish government termed the Second World War. Chief of Staff Russell had journeyed to Germany in a bid to get arms and other assistance from the Nazis, while the Germans hoped to use the IRA network to facilitate espionage operations against the British. It was just the sort of activity that gave de Valera nightmares, but luckily for him the Germans had grossly overestimated the IRA’s capabilities and virtually all the spies they sent to Ireland were exposed and arrested. Russell meanwhile died during a journey on board a German submarine not long after the Coventry bombing, and the IRA campaign soon petered out. By 1945 the IRA had effectively ceased to exist. Its structures and leadership had evaporated. Not even a membership list had survived the defeat. For a short while it seemed as if the long history of violent Irish republicanism had come to an end.

But the movement was not quite extinguished. By 1947, after a slow, painful reconstruction effort, the IRA was showing signs of revival. Structures were rebuilt and a leadership of veterans, headed by the new chief of staff, Tony Magan, put in place. Recruitment was under way, as was training in the hills outside Dublin, and a monthly newspaper was being published. A year later, in 1948, the year of Gerry Adams’s birth, the IRA was large enough to hold a Convention, the gathering of IRA representatives that the organization’s constitution decrees exerts supreme authority over its policies, ideas, and military direction.

The Convention decided that Oglaigh na hEireann, as its own members called the IRA, was to make plans for a new military campaign to end the British occupation of Northern Ireland. There would be no repeat of the mistakes of the Forties Campaign, its leaders determined; this time the campaign would be in the North itself, not in Britain.

It was at this point that the IRA leadership took another small but significant step toward accepting the existence and legitimacy of the southern Irish state, whose creation it had once declared illegal. It forbade units from making any attacks on the Irish police, the Garda Siochana, or any other military forces of the state, for fear that Dublin government reprisals would undermine the offensive against Northern Ireland. The IRA’s ruling body, the seven-man Army Council, issued General Army Order no. 8, forbidding such military action. From then on the South was to be the IRA’s logistical base, while the North would be the war zone.

Preparations for the campaign began with a series of arms raids in Northern Ireland and Britain. In 1955 came a sign that Northern nationalists might be receptive to an IRA campaign when, in the midst of agitation over the arrest of suspected IRA activists, two republicans, one of them an IRA man imprisoned for his part in an unsuccessful arms raid, were elected to the Westminster parliament. The Nationalist Party had stood aside to give Sinn Fein a free run, and overall the republicans won 152,000 votes, an unprecedented level of support, which the politicized Adams movement took twenty years of political work and a major ideological U-turn to better.

Thus boosted, the IRA campaign, code-named Operation Harvest but known popularly ever since as the Border Campaign, began in earnest. In December 1956 a series of cross-Border raids on security and government installations signaled the start. The campaign was strictly limited to areas outside Belfast. The IRA leadership feared that the city’s Catholic population was vulnerable to Protestant attack and might be held hostage by the unionist government for the IRA’s good behavior elsewhere. In practice IRA actions were confined to the Border counties of Fermanagh, Tyrone, and Armagh.

The Northern authorities introduced internment within days of the first acts of violence, while the Gardai harassed the IRA leadership unmercifully. By mid-1957 de Valera was back in power, and when an RUC officer was killed in an IRA booby trap bomb in Tyrone, he introduced internment and later set up military tribunals that handed out draconian sentences. More seriously for the IRA, the public support evident in 1955 failed to materialize on the ground. The bulk of Northern Ireland Catholics simply ignored the IRA’s call to arms.

The Border Campaign limped on for a further five years but was effectively over at that point. In February 1962 the IRA leadership finally acknowledged defeat, ordered its units to dump arms, and admitted the great part played by Northern nationalist indifference to the campaign in the decision to end hostilities.

Twelve people had been killed in the 1956–62 campaign; six were RUC members and six IRA men. Another thirty-eight people—civilians, IRA men, and Northern security personnel—were wounded.[10] At the time unionists were alarmed at this level of IRA activity, but by the standards of the coming conflict it was a tame affair. In 1972, for instance, the worst single year of the Troubles, the entire casualty list for the five-year Border Campaign could be compressed into an average ten-day period.

Exhausted and demoralized, republicans retreated once more. Most quit and took up normal lives, but others returned to the fray determined to rescue something out of the wreckage of defeat. Among them was a forty-six-year-old Dublin painter and decorator called Cathal Goulding, who had several qualifications for leadership. He was one of the small number of IRA men who in 1945 had met in Dublin and agreed to start the slow process of rebuilding the IRA from the ashes of the Forties Campaign, and he was possibly the most enthusiastic of the group. A good friend of the playwright Brendan Behan, with whose widow he was later to father a child, he had a family background that was impeccable from a republican standpoint. His father had “been out” in the 1916 Rising, while his grandfather had been a Fenian revolutionary.

Goulding had also proved his mettle. In 1953 he led a high-profile raid on a British army base in Felsted, Essex, in southeast England, along with the Derry IRA man Manus Canning and a London-based ex-RAF member, Sean MacStiofain, who was later to become the first chief of staff of the Provisional IRA. The raid netted a huge haul of weaponry but so loaded down the men’s getaway van that a routine police patrol became suspicious and stopped them. They were later sentenced to eight years’ imprisonment each. Goulding was released in 1959 and returned to rejoin the IRA’s Army Council and become its quartermaster. But because he had been in an English jail in the early years of the Border Campaign when the most serious setbacks were suffered, he escaped blame for its dismal failure. As a neutral figure amid factions fighting in the ruins of a failed war, Goulding became, in 1962, the IRA’s new chief of staff.

The collapse of the Border Campaign provided a major punctuation mark in the history and development of the IRA. In the space of forty years the organization’s fortunes had ebbed and flowed. At its peak the IRA enjoyed the backing of a majority of the Irish people, and it had fought a long and bloody campaign, which had brought the British to the negotiating table. But that brief success was followed by division, civil war, and disillusionment while the numbers remaining true to the faith of Pearse and Connolly dwindled, as did public support for and even tolerance of the IRA’s activities. The two partitionist states continued to glower at each other across a heavily militarized border, but they also grew roots and developed their own legitimacy. As they did, the IRA was forced to slowly soften and even abandon its hostility to the Southern state and shift the emphasis of its conflict with the Treaty settlement to the existence of Northern Ireland. The Forties Campaign had failed miserably to shift Britain’s support for the unionist state, and then the Border Campaign had disintegrated in the face of Northern nationalist indifference. Smarting from failure and lacking any clear sense of future direction, the IRA in 1962 was at a crossroads.

REPUBLICAN ACTIVISTS surveying the scene at the dawn of the sixties would have seen little to be cheerful about. As the prisoners were released from the internment camp at the Curragh just west of Dublin and from Crumlin Road jail in Belfast, they returned to communities whose indifference to their fate was as pronounced as it ever had been. The internees and sentenced men had hoped for and some even expected popular demonstrations and protests when they were arrested, but there were none, just as there were no crowds to stage welcome-home celebrations upon their release. The depth of demoralization could be measured by the numbers who had agreed to “sign out” from prison as the Border Campaign petered out, men who had given their hated jailers written promises never to take up arms in the IRA’s cause again in return for their freedom. And of those who did not become “signees,” as they were termed, a similar number voted with their feet and refused to report back for duty. The hearths to which many of the IRA retreated after 1962 were cold and lonely places to dream of what yet might be. Few of the released IRA men, surely, could have imagined that in just seven years Ireland would be plunged into the most violent cataclysm in its history.

The 1960s were a period of change for Ireland as for the rest of the Western world. Although the Cold War still raged, other and older enmities were fading. In the United States, John F. Kennedy conquered what many feared was an overwhelming prejudice to become the first ever Catholic president. In Rome fundamental reform was under way. Under the radical leadership of Pope John XXIII the Catholic Church had opened a dialogue with the Church of England, and in 1961 the pontiff met its head, Queen Elizabeth II, whose ancestors had led the English Reformation. Religious ecumenism began to flourish.

In Ireland ancient enmities appeared to be softening too. In 1963 Lord Basil Brookeborough, the conservative, diehard defender of unionism, finally retired and was succeeded by a young Anglo-Irish aristocrat, Captain Terence O’Neill, who could trace his lineage back to sixteenth-century landowners. Despite this conventional, establishment background, modernizing reform was also on his agenda.

In both parts of Ireland the new postwar realities were beginning to make their impact. In the North old traditional industries like linen and shipbuilding were in decline, and their replacement was a matter of urgency. The Northern Ireland government was forced to turn to the outside world for new investment. The need for political stability and the requirement to present a less distasteful image to foreign investors—which in practice meant building bridges to the nationalist community—acquired a new if unfamiliar importance.

O’Neill was also aware that there was now a Labour government in London that was more likely to listen sympathetically to nationalist complaints of discrimination and human rights abuses. Slowly, gingerly, and with frequent glances over their shoulders at their own hard-line grassroots, some unionists began to reach out to Catholics.

O’Neill’s approach infuriated hard-line Protestants, but it was resented by some nationalists, who called it cosmetic, patronizing, and at times insulting. On one famous occasion he explained his approach in almost racist terms:

It is frightfully hard to explain to Protestants that if you give Roman Catholics a good job and a good house they will live like Protestants because they will see neighbours with cars and television sets. They will refuse to have eighteen children but if a Roman Catholic is jobless and lives in the most ghastly hovel he will rear eighteen children on national assistance…. If you treat Roman Catholics with due consideration and kindness they will live like Protestants, in spite of the authoritarian nature of the Church.[11]

Nevertheless many Catholics welcomed O’Neill’s conciliatory policies, and hope of real change was in the air. The Southern state was also in transformation. The era of de Valera ended in 1959 with his retirement as taoiseach, and he was succeeded by another, albeit younger, veteran of the 1916 Rising, Sean Lemass, who quickly discarded Dev’s protectionist economics in favor of attracting foreign investment. The era of the technocrat, North and South, had arrived.

The two governments in Ireland were on a similar course, and it made sense for them to examine ways of improving cooperation. In 1965 the political ice cracked when O’Neill invited Lemass to Stormont and a month later made a return trip to Dublin. The journeys continued when Jack Lynch succeeded Lemass as Irish prime minister.

Throughout the continent of Europe, a wave of liberalism brought new power to the Left in a way that had not been seen since the 1930s. In Britain a dozen years of uninterrupted Conservative rule were brought to an end when Labour’s Harold Wilson swept into power on a ticket of economic modernization. In Ireland the Left made gains in both jurisdictions. The Northern Ireland Labour Party was winning seats to Stormont, while in the South the Labour Party was shedding its conservative Catholic image and beginning to talk more openly of the merits of socialism.

In Europe, as a whole, moves were afoot to dismantle economic borders and to bury old hatreds. Inspired by the American civil rights movement and the excitement brought on by an era of unprecedented sexual liberation, among other factors, the movement toward the left grew increasingly strong.

And that is certainly what Cathal Goulding discerned. Under his leadership the IRA made the most radical and determined move to the left in its history, embracing a doctrinaire Marxist analysis of Northern and Southern Irish politics that would eventually split the organization and spark the most violent upheaval in modern Irish history. The manner in which Goulding failed to take a united republican movement down his chosen path was to provide Gerry Adams with one invaluable lesson when he made his own very similar journey some twenty years later. Preventing the sort of split that destroyed Goulding’s hopes became a strategic imperative for Adams.

The greatest obstacle blocking Goulding was the immense conservatism of the Republican movement. Overwhelmingly rural and lower middle class in their makeup, most IRA members were people of a traditional Catholic outlook. This was reflected in the rituals and the ways in which the IRA was organized, many of which had changed little in forty years.

Men and women, for instance, were still segregated into separate military units, as they had been in 1916 when Patrick Pearse sent out a female Irish Volunteer from the GPO in Dublin with a white flag of surrender to present to the British army. The IRA of the 1960s, like Irish society at large, was slow to accept gender equality. The IRA had always been an exclusively male organization, and its members were the soldiers who did the actual fighting when there was any to be done. The women were organized into the Cumann na mBan (Women’s Group), with their own distinct structures and leadership. Their role was the IRA equivalent of being stuck in the kitchen and the bedroom; they carried messages and smuggled weapons and explosives and they nursed wounded IRA Volunteers. They could be useful for gathering intelligence and carrying weapons, but they did very little, if any, actual fighting. Nor did they play any part in the formulation of IRA policy or strategy. That was a male preserve, the privilege of the seven-man Army Council, to which Cumann na mBan was subservient.

Catholic Church ritual permeated IRA ceremonials. It was commonplace for a decade of the rosary to be said, often in Irish, at the start of IRA ceremonies, such as the annual Easter commemorations of those who were killed in the Rising of 1916 and other phases of the struggle. It would be difficult to imagine a more effective way of confirming unionist prejudices about the IRA.

Most IRA members had a simplistic set of motives for joining. They believed that only armed force could remove the British from Ireland and that people who advocated parliamentary methods had sold out the struggle. As one commentator put it, “To go into the Dail [the Irish parliament], to seize power, was not only an invitation to corruption, a tainted tactic already proven sterile, but also, and most important, outrageous immorality.”[12]

Their history was full of examples of IRA leaders who had abandoned physical force for parliamentarism, yet had failed to force the British out of Ireland, North or South. They viewed with suspicion, therefore, the motives of any who advocated such a course. As one account put it, “They believed no one ever went into politics except a failed revolutionary.”[13] Their opposition was not just practical but almost spiritual as well.

The only parliament to which they gave allegiance had by the 1960s long since disappeared. That was the Irish parliament of 1921, the so-called Second Dail, the last gathering of representatives chosen in a pre-Treaty, all-Ireland, thirty-two-county election. The Second Dail had a Sinn Fein majority elected on a platform of support for the 1916 Easter proclamation of independence, and republicans regarded the mandate as almost sacred. Although the Second Dail later voted narrowly for the Treaty, it became an article of faith among the recalcitrant IRA that no other parliament, no other government, could claim the legitimacy bestowed on that parliament. After the Treaty was endorsed, the Second Dail ceased to exist as far the IRA was concerned, and its partitionist successor, the Northern parliament at Stormont as well as the new parliament in Dublin, was regarded as illegal.

The rump of the Second Dail, those surviving TDs who had voted against the Treaty, were deemed to be the only legitimate government of Ireland, but they were dying off, year by year. In 1938 the dozen or so survivors passed on their authority to the IRA’s Army Council to safeguard until all the people of Ireland could again freely choose their own government. To the outside world it may have looked absurd, but it was on this basis that the IRA leadership framed its claim to be the sovereign government of Ireland.

Because of this, one of the most hallowed principles of traditional republicanism was the refusal to recognize or take seats in either Stormont or the Dail. It followed that any republican who betrayed this principle was implicitly recognizing the Treaty and by so doing betraying the Irish people and all those IRA men and women who had laid down their lives in the fight to free Ireland. The logic of all this was that republicans had to deny legitimacy to the other institutions of the state as well. Republicans refused, for example, to ask either police force in Ireland for permission to parade or raise funds. IRA standing orders forbade prisoners from entering a plea in criminal courts—that was the same as recognizing the state—although an exception was made when IRA men faced a capital charge. Abstentionism was the defining characteristic of Irish republicanism, and it was written in legal stone in the constitutions of both the IRA and Sinn Fein. Any IRA or Sinn Fein TD or MP who took his or her seat in a partitionist body or even suggested discussing the idea would be liable to automatic dismissal or expulsion.

These were the uncomfortable realities that faced Goulding and the small number of radical advisers and confidants he had gathered around him. And it was the left-wing background of this coterie that produced the first signs of internal unrest. With the IRA laid low after the Border Campaign, Goulding had turned to new men and new ideas for inspiration. There was already a sentiment in favor of moving leftward. IRA prisoners in the North’s largest jail, Crumlin Road in Belfast, had produced a journal called Saoirse (Freedom) under the editorship of a young County Cork militant called Daithi O Conaill, which had published an article, “Quo Vadis Hibernia?” (“Whither Ireland?”), advocating involvement in social and economic agitation.[14] Not for the first or last time prison was a crucible in the development of IRA politics.

The move to the left accelerated in 1963 when two radical intellectuals put their stamp on the movement. One was a young computer scientist called Roy Johnston, who had returned to Dublin from England, where he had been prominent in the Communist Party of Great Britain–linked Connolly Association. The other was Anthony Coughlan, a young lecturer at that most Anglo-Irish of bastions, Trinity College, Dublin, who had been national organizer of the Connolly Association in Britain.

The Connolly Association members had been urging on the IRA what they called “a new departure” prior to the 1962 cease-fire. Long before Johnston and Coughlan were able to exercise direct influence over Goulding, their program had been spelled out. One chronicler of the period wrote, “[The IRA was] advised to work through a broad alliance for the displacement of the Unionist regime and for political democracy in the North and for social progress… in the South. They were also advised to end their abstentionism and seek to work through parliamentary institutions.”[15]

Johnston, who was quickly appointed to the IRA’s Army Council by Goulding as a sort of political adviser cum commissar, was the inspiration for the establishment of a debating society named after the eighteenth-century Protestant founder of modern Irish republicanism, Wolfe Tone. With branches in the main universities and in Belfast and Dublin, the Wolfe Tone clubs discussed and advanced the Connolly Association’s agenda. Left-wing veterans like Peadar O’Donnell and George Gilmore re-emerged from obscurity to contribute to the debate.

The political program developed by Johnston and Coughlan was borrowed directly from a former Soviet leader, none other than Joseph Stalin, who had died a decade before. Called the “stages theory,” it mapped out a rigid, dogmatic path to Irish socialism in which Ireland would pass through three distinct phases before reaching the goal of a workers republic.

The first would be the creation of a normal liberal, parliamentary democracy in the North, which would be achieved through agitation on civil rights issues. The employer class in the North would cooperate with this, since it was in its economic interests to do so, and, Johnston predicted, there would be growing working-class Protestant and Catholic cooperation and unity as workers realized how much their interests coincided. In the second phase, revolutionary links would stretch across the Border as radicalized and increasingly united Northern workers would make common cause with their Southern counterparts, who themselves were being radicalized by Sinn Fein agitation. The third phase would be revolution and final victory.

It followed that clinging to abstentionism was an absurdity. How could Sinn Fein hope to relate to the social and economic problems of the Irish working class, Johnston and his allies asked, as long as it refused to address those problems in the only forums available to Irish workers, their parliaments and governments? The people accepted the partitionist arrangements, and so should the IRA, they urged.

The “new departure” had implications for the IRA and the direction of its armed struggle. Gradually the emphasis of its activities shifted to economic and social agitation and away from the traditional goal of waging war against Britain. The IRA and Sinn Fein became involved in rural cooperatives, and in Dublin they set up a housing action committee that staged sit-ins to highlight poor living conditions and overcrowding.

A campaign was launched against foreign ownership of mining and fishing rights. The IRA organized illegal “fish-ins” on exclusive salmon runs in the west of Ireland and offered manpower to help striking workers. Foreign landowners, mostly Germans, were targeted and buildings torched. In Limerick the IRA burned a bus transporting strikebreakers to an American-owned company. In Kerry a foreign-owned lobster boat was sunk.

Some twenty years later Roy Johnston explained the reasoning behind the leftward shift: “The idea was that if links could be cultivated between the movement and the people, the roots would be firmly in the ground and a principled, political stand would be made, even in ‘illegal assemblies’ such as Leinster House [the Dail] without automatic corruption.”[16]

While all this naturally alarmed the Fianna Fail governments of Sean Lemass and his successor Jack Lynch, the implication was that the need to wage armed struggle against Britain had moved farther and farther down the IRA’s list of priorities. That meant downgrading the IRA itself, and when the split finally came there were bitter accusations from his critics that the Goulding leadership had deliberately run down the organization, dismantling command structures, discouraging or diverting promising recruits, scaling down training, and—worst of all—diminishing the IRA’s stores of weapons.

The logic of the “stages theory” was that if Northern Protestants were going to cooperate in the democratization of their own state, then the idea that Catholics needed arms to defend themselves against the Protestants was nonsense. Weapons acquisition, once a priority for the IRA, was virtually abandoned. Money that would have been earmarked for this was devoted instead to political activities.

The last defining characteristic of the Johnston-Goulding strategy was the notion that all this could best be achieved if the IRA and Sinn Fein cooperated and perhaps even merged with similarly minded, progressive political parties. It was the classic broad-front strategy so beloved of left-wing groups: Johnston and his associates called it the National Liberation Front (NLF), a term with echoes of the Vietnamese resistance then beginning to radicalize American youth. The other members of the NLF, at least those mentioned most often as candidates for partnership, were the Communist Party of Northern Ireland, its Southern counterpart, the Irish Workers Party, the Connolly Association, and the Connolly Youth Movement. It was this proposal which fixed the notion in conservative, dissident minds that the IRA was being slowly taken over by godless Marxists.

And there was resistance to the Johnston agenda. The IRA had always distrusted outsiders, especially those preaching foreign ideas. Most of its members were God-fearing, Mass-attending Catholics who had been reared in an Ireland thoroughly dominated by a deeply conservative Catholic Church. It was, though, the manner in which Goulding pursued the Johnston-Coughlan strategy that finally fractured the IRA as much as the ideology he and his supporters espoused. He made two fatal errors. First, he tried to push his ideas onto the IRA at too early a stage. Not only did he not have enough support before he moved; he had failed to isolate his opponents first. Gerry Adams would never make this basic error. Goulding moved too early, and when he met failure his frustration grew and propelled him into confrontation with his opponents. That was his second mistake— and Adams learned from that too.

Having decided upon a strategy of open confrontation, Goulding set the stage for a series of damaging rows within the IRA during the mid-1960s that were destined to end in a damaging and, as it turned out, bloody split. The battles were fought at IRA Conventions and at Sinn Fein’s annual conferences as Goulding and his allies attempted to push through the Johnston-Coughlan agenda, only to find their way blocked by traditionalists and conservatives. Rebellious branches, particularly in rural areas, were expelled, as at one stage was the entire branch of the women’s IRA, the Cumann na mBan. Frustrated at failure, Goulding turned to tactics that sometimes verged on the ludicrous. At one point he expanded the Army Council from seven to twenty members in a bid to secure a majority for his policy. The normally highly secretive body, which went to great lengths to hide its meetings from the authorities, was now so large that it almost had to hire a hall for its gatherings.

Of the Council’s usual seven members, Goulding could count on the support of only three. On his side were Sean Garland, a fellow Dublin republican and a hard-line Johnston supporter, and the quixotic County Wicklow republican Seamus Costello. Opposing him were Ruairi O Bradaigh (Rory O’Brady), a County Roscommon schoolteacher who had been chief of staff when the Border Campaign ended in 1962; Daithi O Conaill, who had been badly wounded in County Tyrone during the campaign; and Sean MacStiofain, Goulding’s IRA comrade on the ill-fated Felsted arms raid in 1953 and the IRA’s then director of intelligence. Between them, holding the balance of power and uncertain as to his sympathies, was Tomas MacGiolla, the lofty president of Sinn Fein who doubled as chairman of the Army Council. He did not finally throw in his lot with Goulding until the summer of 1969.

Goulding and his allies resorted increasingly to the tactic of purging their enemies. Senior Northern IRA figures were forced out or left in disgust at the tactics being used against their friends. Among them were men who would play key roles in the formation of the rival Provisional IRA a few years later, characters like Jimmy Steele, who had spent twenty years in jail for IRA activity, and Sean Keenan from Derry, a republican stalwart for years. With other Belfast men who had left or were eased out during this period—people like Joe Cahill, Jimmy Drumm, Billy McKee, the Kellys of North Belfast, the Hannaways, and Seamus Twomey—they were to form the core of a bitter opposition just biding their time to strike back at Goulding. If Adams learned another lesson from all this, it was to divide and discredit opponents lest they unite against him.

It was into this maelstrom of conflicting political ideas and personalities that Gerry Adams plunged when he joined the Belfast IRA’s D Company, although for a while it did not impinge much on his life. As he was to say himself many years later, all this ferment was “going on, if you like, above my head.”[17] He was only sixteen when he joined the republican movement, and the debate within it was taking place far away, at leadership level and mostly in Dublin, where the IRA policymakers were based. But within four or five years events were to force Adams to take sides in their quarrel, and that decision would have unforeseeable consequences for Anglo-Irish politics.

The failure of the Border Campaign had set the scene for the tragedy that has always befallen Irish republicanism in the aftermath of defeat, when the ambitions of those who were convinced that fresh directions had to be taken clashed with the stubborn belief of others that change and compromise were indistinguishable and that deviating from principle meant devaluing the sacrifices of dead comrades. And so it was as the 1960s unfolded; the IRA was on the verge of another rupture.

IDEOLOGICAL, generational, and personality differences had set the scene for internal republican division, but they were not the only factors at work. Beyond the control of the IRA’s leaders, beyond the control even of its unionist governors, social and economic change was remolding Northern Ireland and in the process creating the circumstances that would bring political instability. Two factors in particular destabilized Northern Ireland in the 1960s and helped plunge it into the most violent and sustained conflict in Irish history.

The first and most damaging development was the furious reaction within unionism to the moderate reformism of Prime Minister Terence O’Neill. This took three forms. Some within O’Neill’s Ulster Unionist Party saw his modernizing policies as a threat to the union with Britain and argued that softening hostility to nationalists was a fatal weakness, while others saw the turbulence as providing the opportunity to snatch the prime minister’s crown from his head.

The second group coalesced around the powerful, hectoring frame of the Reverend Ian Paisley, a young Protestant street preacher with political ambitions who combined a fundamentalist biblical view of the world and a hatred for the Roman Catholic Church with a fiery opposition to all things ecumenical, religious, and political. Paisley’s oratorical skills made him a formidable opponent. He could sway a mob with a few well-chosen words. He also represented a virulent strain within unionism that could trace its roots back over a hundred years, to Belfast street preachers who would regularly incite crowds to riot, burn, and kill their Catholic neighbors.

Paisley’s violent rhetoric had a similar effect in the Belfast of the 1960s. On the Shankill Road in North Belfast, one of the city’s toughest loyalist areas, a group of men led by a former British soldier called Gusty Spence met to form a paramilitary group they called the Ulster Volunteer Force (UVF), after the huge private Protestant army that early unionist leaders like Edward Carson had mobilized in 1912 to resist Home Rule. Convinced that the IRA was planning a new violent campaign to mark the fiftieth anniversary of the Easter Rising, in 1966, and inspired by Paisley’s oratory, Spence’s UVF launched a pre-emptive strike.

They aimed to kill IRA leaders but, like their counterparts when the Troubles intensified, were content to accept any Catholic target. By the summer of 1966 they had killed three times. One victim was a seventy-seven-year-old Protestant widow who died when a gasoline bomb intended for a Catholic-owned pub engulfed her house instead. Another was a tipsy twenty-eight-year-old Catholic pedestrian shot dead on the street; like so many of the victims of the oncoming Troubles, he happened to be in the wrong place at the wrong time. The third was an eighteen-year-old Catholic barman, Peter Ward, who had the misfortune of choosing a Shankill Road bar frequented by Spence and his colleagues for an after-work drink with friends. The deaths were the peals of thunder that foretold the impending storm.

Unionist and Protestant reaction was mirrored by Catholic and nationalist assertiveness. By the early and mid-1960s that new mood, fueled by a growing educated and frustrated middle class, took two forms. On the one hand Catholics began to participate in a state that they had initially boycotted. The first sign of this came in 1965 when, in response to the O’Neill-Lemass meetings, the Northern Nationalist Party entered the Stormont parliament for the first time since 1930 and became the official opposition, a step that signified recognition to the Northern Ireland state.

The other sign of the new mood was an increasing willingness of Catholics to stand up and demand their rights. Although Catholic and nationalist politics had been characterized for many years by meekness in the face of unionist strength, some in that community were well aware that they were the victims of discrimination in jobs and public housing. In the 1960s they set about compiling the evidence. The Campaign for Social Justice was started by a County Tyrone doctor and his former teacher wife, Con and Patricia McCluskey. Their research showed that Catholics were hugely underrepresented in places like the central and local government civil service. Soon the idea was mooted of setting up a civil rights body based on the NAACP in the United States, which had fought for the rights of blacks.

In 1967 the Northern Ireland Civil Rights Association (NICRA) was formally launched at a public meeting in Belfast and drew up a list of reforms. Its most important demand was for the scrapping of a rule that restricted to property owners the right to vote in elections to the local councils. The rules gave some businessmen, most of them unionists, up to six votes, while thousands of working-class Catholics, who did not own their homes, were disenfranchised. The councils exercised considerable power. Not only did they give employment, but they were responsible for building public housing projects. That meant they could decide who, Protestant or Catholic, got decent public housing.

NICRA also sought the scrapping of gerrymandered electoral wards that, in places like Derry, Northern Ireland’s second city, gave unionists control over the local council even though the Protestants who voted for them were in a minority. The disbanding of the B Specials and the removal of the Special Powers Act, the law that gave the unionist government unprecedented powers and a points list for public housing, completed the roll call of reforms demanded. Significantly none of NICRA’s demands touched directly on the existence of the Border. Composed of a mixture of liberal and radical opinion, NICRA’s leadership had a simple attitude: if Nationalists were obliged to be British citizens, then they should have British rights.

These stirrings soon showed themselves on the streets. In June 1968 republicans in Caledon, County Tyrone, organized a sit-in at a house allocated by the local council to the unmarried secretary of a local unionist politician, and they demanded that poorly housed Catholic families take precedence. The protest drew wider publicity when a Nationalist MP at Stormont, Austin Currie, joined the sit-in. Two months later NICRA helped organize a protest march from the nearby nationalist village of Coalisland to Dungannon, Tyrone’s largest town, where protesters found their way blocked by a flag-waving crowd of Paisley supporters.

Standing up to demand community rights also meant that nationalists were more and more ready to resort to physical force to defend themselves. The early part of the 1960s had seen two examples of this. In October 1964 the lower part of the Falls Road, the main Catholic artery leading into Belfast city center, was convulsed by a two-day battle when, after threats from Paisley, the RUC invaded the area to remove an Irish tricolor from the offices in Divis Street of the Sinn Fein candidate in the Westminster general election. The trouble resulted in fifty civilian and over twenty RUC injuries. Eighteen months later, in June 1966, a month after Spence’s UVF started to kill, Catholics again fought hand-to-hand battles with the police when Paisley brought a mob through the nationalist Markets area en route to a city center protest against ecumenical Protestants. The tinder was in place and was dry; it needed but a spark to set it ablaze.

The spark came on October 5, 1968, when a poorly attended civil rights march in Derry was batoned off the streets by a force of RUC officers who appeared to have lost all control of their emotions. Derry had been chosen for protest because it was in nationalist eyes the capital city of injustice. Not only were the majority Catholics prevented from taking control of their city, but years of unionist rule had condemned them to appalling housing conditions and disproportionately high unemployment.

The unionist government reacted with alarm to the proposed civil rights march. In the eyes of its hard-line supporters, the walled city had been the symbol of Protestant supremacy over nationalist Ireland ever since its inhabitants had withstood a terrible siege in 1688 and helped the Protestant King William of Orange defeat the Catholic usurper, King James II. Its official name was Londonderry, after the City of London companies that had financed its reconstruction as a planter city. In celebration of its heroic resistance during the siege, unionists christened Derry “the Maiden City,” and any violation of its walls had to be resisted. The civil rights marchers wanted to parade right into the walled center, and unionists saw that as a bid to ravage the maiden. After an obscure and unheard-of loyalist group threatened a counterprotest, the unionist home affairs minister, William Craig, promptly banned the march.

The marchers assembled and found themselves blocked on all sides by riot police who laid into the crowd with batons flying. Eighty-eight demonstrators were injured and thirty-six arrested. Two British Labour MPs had turned up as observers, guests of West Belfast’s Republican Labour MP, Gerry Fitt, who was himself batoned around the head. Later the police turned a water cannon on dispersing protesters, in a violent scene reminiscent of police brutality in Alabama just five years earlier. The whole scene was captured by a camera crew from RTE, whose film was also screened on British TV and around the world. So was an interview with a blood-bespattered Fitt. These scenes seemed more appropriate in South Africa, but Northern Ireland was British and such things were not supposed to happen in Britain. Derry on October 5 sent a shock wave through both Ireland and Britain at the realization that “John Bull’s Other Island,” a headline borrowed from Shaw by one London newspaper to describe Northern Ireland, was a political slum.

It rapidly became an unstable slum as well, as events took on their own momentum. The nationalist pressure intensified in November after a repeat march in Derry brought fifteen thousand onto the streets and the RUC, hopelessly outnumbered, allowed the protest through. That in turn fueled more unionist anger at Terence O’Neill, and the crowds attending Paisley’s rallies grew. As NICRA held more marches and mobilized protests throughout the North, the British Labour government applied pressure on O’Neill to move faster and further to satisfy nationalist demands for change. The government in London led by Harold Wilson had been horrified and alarmed by the events of October 5, and the British prime minister demanded action from the unionists. At the end of the month O’Neill relented and announced a package of reforms and followed this up with a televised appeal for a break in the civil rights agitation.

NICRA obliged, but a small left-wing, radical student group, the People’s Democracy (PD) based at Queen’s University in Belfast, responded to O’Neill’s package with scorn and announced plans for a seventy-five-mile trek between Belfast and Derry to start on New Year’s Day 1969 to protest about the inadequacy and slowness of change. The route would go through some tough unionist areas, and trouble was inevitable. On January 4, as the march reached the outskirts of Derry, Paisleyite supporters ambushed the marchers with stones and cudgels. In response fierce bloody rioting between nationalists and the RUC broke out in the Bogside, the main Catholic area of Derry, in reality a slum that nestled at the foot of Derry’s walls.

As unionist discontent grew, O’Neill’s cabinet colleagues deserted him one by one, and in February he called a sudden and unexpected election to the Stormont parliament, hoping it would strengthen his hand—but if anything the result strengthened his opponents while bringing more able nationalists into political life. In March bombs exploded at an electricity substation and punctured a pipeline that supplied Belfast with most of its water. The bombs were really the work of loyalists, but at the time most people assumed a resurgent IRA had planted them. The subterfuge worked, and O’Neill was undermined even further.

In April the radicalization of nationalists continued apace when the student civil rights leader Bernadette Devlin was elected to Westminster in a by-election in Mid-Ulster, overcoming Catholic divisions to capture the seat. Unlike the republicans, she had no qualms about taking her House of Commons seat to highlight her community’s grievances, although her appetite for parliamentary politics quickly waned. In the same month O’Neill’s cousin Major James Chichester-Clark, a fellow Anglo-Irish aristocrat, quit his cabinet in protest of the prime minister’s policies, and an exhausted O’Neill surrendered and resigned. Chichester-Clark, whose ancestors had planted Ulster in the seventeenth century, promptly succeeded him as prime minister.

The fire was well alight, and soon the explosion came. During July so-called loyalist vigilante groups appeared on the streets in Belfast. They claimed that their role was to defend Protestant areas, but in reality they were there to attack Catholic districts. One group in particular, the Shankill Defence Association, led by an ally of Paisley, was involved in vicious clashes with Catholics in North Belfast, as were RUC riot squads. Reports of Catholics being intimidated into leaving their homes began to circulate.

The focus switched back to Derry when the Apprentice Boys, an offshoot of the Orange Order, made it clear they would go ahead with that year’s annual march on August 12 despite predictions of disaster. The march duly went ahead and calamity came in its wake. After skirmishes on the edge of the Catholic Bogside, ferocious rioting broke out between nationalists and the RUC. The rioters repulsed wave after wave of baton-wielding policemen with rocks and gasoline bombs, but soon the RUC turned to CS gas and flooded the narrow streets of the Bogside with acrid clouds that choked indiscriminately. Chichester-Clark mobilized the B Specials while the Irish prime minister, Jack Lynch, warned that his government would not “stand by” and watch as nationalists were attacked. Northern Ireland was moving rapidly toward the edge of civil war, and it looked as if the conflict might engulf the whole island. By August 14 the police were exhausted, and the Stormont government was forced to request military assistance from Britain. That day the first contingents of British soldiers took to the streets of Derry. Thirty years would pass before they left.

As the riots raged in Derry, civil rights leaders tried to ease the pressure on the Bogsiders by appealing to nationalists elsewhere to stage protests and demonstrations so as to stretch RUC numbers and resources. The ploy succeeded but with consequences that few could have foreseen. In Armagh a thirty-year-old Catholic man was shot dead by a party of B Specials in riots that had followed a civil rights rally. But the worst trouble was in Belfast, where that night five people died in gun battles between the RUC, loyalists, and republicans.

There were two trouble spots. In Ardoyne, a small nationalist enclave in North Belfast, two Catholics were shot dead by the RUC in clashes that were to follow a classic pattern. First rival crowds gathered to exchange insults and the occasional stone. The RUC arrived and almost immediately clashed with nationalists. As the clashes worsened, Catholics built barricades and threw gasoline bombs and stones at police, who invariably replied with baton charges. Behind the police came mobs of loyalists, many armed with guns and gasoline bombs, setting fire to homes and public houses as they swept through. Soon gunfire was echoing through the small streets, and mobs fought hand-to-hand battles.

The worst trouble was where it had always been worst, in the jungle of side streets that linked the nationalist Falls Road with the fiercely loyalist Shankill Road, the setting for regular outbreaks of rioting that dated back a hundred years. The two roads, with their crowded streets of small two-story homes, converged in a V-shape near the city center, and it was here, where they were closest, that the worst violence took place. The trouble began outside a police station on the Falls Road when a republican crowd protested and continued with clashes between the RUC, B Specials, and loyalist crowds on one side and nationalists on the other in and around Divis Street where the V of the Falls and Shankill narrowed. A Protestant member of a crowd trying to break through onto Divis Street was shot dead by one of two IRA gunmen firing from the roof of a nearby school. The RUC returned fire, and a gun battle ensued that left three policemen wounded. The RUC reaction to this was dramatic. Senior officers deployed armored cars fitted with Browning heavy machine guns. The weapons could fire high-velocity bullets over a range of 2.5 miles and were completely inappropriate for use in a heavily populated urban area. The results were again grimly predictable. A nine-year-old Catholic boy was killed when a bullet tore through the walls of his bedroom in an apartment block in Divis Street and removed the back of his head.

The next day the situation deteriorated dramatically. A local-born British soldier home on leave was shot dead by the RUC near his home in the Divis Flats complex. Earlier in the day serious rioting had broken out in the mid-Falls area when a Protestant mob invaded the Clonard district, burning homes as they came. Their target was Clonard Monastery, home of the Redemptorist order, whose priests later played such a significant role in the peace process. There was fierce street fighting and a lot of gunfire. One teenage boy, Gerald McAuley, was shot dead. A member of the republican youth wing, the Fianna, he was the first IRA member killed in the Troubles. The loyalists were eventually repulsed, but not before they had burned down much of Bombay Street under the shadow of Clonard Monastery, forcing its terrified inhabitants to flee. Earlier six hundred soldiers from the Third Battalion of the Light Infantry had taken up positions in Catholic and Protestant districts between the Falls and Shankill Roads but were seemingly unable to stop the destruction. Belfast’s infamous peace line had been drawn; soon it would take a more solid form and a twenty-foot-high wall would eventually be erected, making the division permanent. The Troubles had begun.

After two days of rioting and disturbances across Northern Ireland, eight people lay dead—two Protestant, the rest Catholics. Much of West and North Belfast resembled a war zone. Barricades blocked dozens of streets in nationalist areas, erected to prevent loyalist incursions and to keep out the RUC; scores of houses had been torched, bars lay wrecked and looted, and public transport had ground to a halt. Some seven hundred people had been wounded or injured and nearly three hundred buildings destroyed or badly damaged. Belfast was no stranger to such sectarian conflict. Throughout the second half of the nineteenth century and again during the 1920s and 1930s, there had been fierce rioting, burnings, and gun battles, but the scale and intensity of the violence of 1969 was beyond living memory. Everywhere anxious Catholics organized street defense committees and began patrolling their areas. Less visible but more traumatic was the intimidation and forced evictions that had occurred throughout the city. The British government later set up a commission of inquiry into the events, headed by a senior judge, Lord Scarman, which concluded that 1.6 percent of all households in Belfast had been forced to move between July and September 1969 and that Catholics had suffered most. Over fifteen hundred Catholic families, Scarman said, had been forced from their homes, five times the number of Protestants. That summer hundreds of Catholic families sought refuge in the safety of ghettos like West Belfast, as the biggest forced population movement in Europe since the Second World War began.

These were the events that finally forced the festering divisions within the IRA into the open. The riots and gunfire, the threat to their neighborhoods and communities, had brought back into circulation republicans who had quit in disgust at the Goulding-Johnston leadership, and now they had a serious charge to lay at their opponents’ door. The move to the left, they angrily protested, and the leadership’s obsession with politics had led to a military rundown of the IRA, and this had left Catholic areas defenseless. The emphasis given to politics over military matters by Goulding and his henchmen, they argued, meant that the IRA had failed in its primary role in the North, to protect its own people. The anger within this relatively small group of disgruntled former IRA men would simmer throughout the autumn and early winter of 1969 before bursting to the surface in the most serious split in the ranks of physical-force republicanism since the awful Irish civil war.

Throughout the years leading up to the tumultuous events of August 1969, Gerry Adams suffered, as he wrote years later, divided loyalties within the warring IRA: “I was in a strange position, one of a small cadre with contacts in both factions.”[18] His father, his uncle Liam Hannaway, and many family friends like the future Provisional IRA chief of staff Joe Cahill sided with the traditionalists. But Gerry Adams, a generation younger, clearly took with enthusiasm to some aspects of Goulding’s strategy, especially the emphasis on political activity, and was ambivalent about others like the dropping of abstentionism. Most important, he stayed in the Goulding-led IRA and refused to follow the example of his father’s friends, many of whom had quit the organization in protest against its policies.

In many respects his track record as a young republican was a model of what Roy Johnston had in mind; indeed in one interview he recalled attending lectures given by Johnston to the Wolfe Tone Society in Belfast, saying he had found them “excellent.”[19] Adams threw himself into the sort of agitational activity that the new leadership firmly believed would win them working-class support. He was to the fore in housing campaigns like that in 1966 which opposed the building of high-rise apartments in Divis Street in the Lower Falls Road section of Belfast. He advocated other forms of political work that traditional IRA members frowned upon, such as producing local news sheets to deliver door to door. This was not the sort of work of which soldiers approved. The young Adams also supported Goulding’s notion of the National Liberation Front, the left-wing alliance that his father’s generation suspected had opened the door to Communist influence. In 1967 Adams could be found attending the inaugural Belfast meeting of NICRA, whose activities gave substance and meaning to Goulding’s strategy.

On the defining question of dropping abstentionism, Adams adopted a position that in years to come would be the hallmark of his approach to such internal controversies. “Although I was opposed to dropping the traditional abstentionist policy, I had no objection in principle to developing a debate on electoral strategy or abstentionism….”[20] The distinction between the two was sometimes difficult to discern, since urging “debate” about abstentionism could be regarded merely as a way of creating the space within which the “debate” could be won. It was a clue that he was much more pragmatic about the issue than many of those who helped him found the Provisionals.

Adams’s sole disagreement with the Goulding-Johnston approach was over their idea that Northern Ireland could be transformed into a normal democratic society, one in which left-wing issues could unite Protestant and Catholic, Unionist and Nationalist, in common cause. His experience on the streets of Northern Ireland, he believed, flew in the face of this. That view was formed, according to his own account of events, early on in his political activity when along with other Sinn Feiners he had forged links with Protestants living on the New Barnsley housing estate, adjacent to his native Ballymurphy, in a campaign to demand a pedestrian crossing at a spot on the busy main road dividing the estates where a child had been killed by a car. The campaign was successful, but the Protestant-Catholic unity proved to be short-lived when a Paisleyite politician from the Shankill Road arrived to stir up fears of popery and republicanism among the Protestants of New Barnsley. Adams later commented acidly on the episode: “If the State would not allow Catholics and Protestants to get a pedestrian crossing built together, it would hardly sit back and watch them organising the revolution together.”[21]

As the civil rights campaign gathered pace and sectarian tensions were sharpened, Adams’s conviction that cross-community unity was a chimera hardened, as did his misgivings about the way the IRA leadership was handling the gathering crisis. “Riot situations were beginning to develop,” Adams wrote, “but neither the Belfast nor Dublin [IRA] leaderships were able to understand what was happening let alone to give proper direction.”[22] As 1969 wore on and street conflict loomed, some republicans, North and South, began to agitate about the need to prepare for defense but found the Dublin leaders unsympathetic and wholly absorbed in their coming internal battles with the traditionalists. That summer IRA dissidents demanded to know from Goulding how much weaponry existed to defend the Northern nationalists. “When we asked what gear there was the reply was enough for one job, i.e., one rifle, one Thompson submachine gun, two short-arms [handguns], training stuff in other words,” recalled one.[23]

Goulding and his allies ignored dangerous signals from Belfast. In late August a group of former IRA men, people who had long disagreed with the leadership and had either left or been expelled, met in the New Lodge Road district of the city to decide what they should do about the Belfast leader ship of the IRA then in the control of a Goulding loyalist, Billy McMillen. They were angry about McMillen’s failure to organize the August defense and they blamed Goulding for the Catholic deaths. Among them were Jimmy Steele, Joe Cahill, Billy McKee, John and Billy Kelly, Jimmy Drumm, Seamus Twomey, and a Southern ally, Daithi O Conaill. Gerry Adams was the only serving Belfast IRA man invited.

The old guard resolved to oust McMillen, and in September its opportunity came when McMillen called a meeting of IRA commanders in the Lower Falls Road area. Armed men led by Billy McKee were sent to demand his resignation. McMillen refused, and after heated arguments a compromise was hammered out. The Belfast IRA would withdraw from Dublin control for three months, McMillen would add some of McKee’s men to his operational staff, and weapons would be sought. If Goulding had not abandoned his strategy by the three-month deadline, McMillen would be replaced and Belfast would be run by a separate Northern Command. Meanwhile the Belfast units would play no part in the coming debate about abstentionism, effectively opting out of it.

The old guard clearly believed that Adams would be on its side not least because his father and all their family friends were. Adams had been deeply angered by the events of August. He had toured the city on August 16 to view the destruction and been appalled by what had happened in and around Bombay Street. He met his cousin Kevin Hannaway and others who had been in the thick of the violence. “The fighting here had been at very close quarters,” he later wrote, “yet the poorly armed defenders had repelled a large, much better armed group of attackers. What particularly incensed us was that Bombay Street was burned despite the presence of British troops who had been deployed around the area sometime beforehand.”[24]

Even so, Adams still had a foot in both republican camps. He had a soft spot for McMillen, liked much of Goulding’s political approach but deplored his failure to defend Belfast’s Catholics. As the August riots subsided, it was the fear among republicans that their people were still vulnerable and needed proper defense that was really stirring opinion in their circles. For once republican fears were shared in the city’s working-class nationalist areas, and it was this harmony that fueled support for the Provisionals and eventually persuaded Adams to throw in his lot with them.

Despite the violence and trauma of that summer and the divisions all this brought, the Goulding leadership proceeded with the plan to drop abstentionism. Goulding had set up an internal IRA commission under Sean Garland to examine the movement’s attitude toward the issue, and predictably it recommended that republicans take their seats in all three parliaments. The enlarged Army Council endorsed this by a vote of twelve to four, with four absentees. Ruairi O Bradaigh, John Joe McGirl, a friend of Adams from County Leitrim, Sean MacStiofain, and one other Northern republican were the dissenters, all of whom were to be to the fore in the impending split. An IRA Convention was called for December, and Goulding prepared well for it, ensuring that this time a majority of the delegates would be in his camp. They were, by a majority of thirty-nine to twelve. The anti-Goulding faction walked out and, with MacStiofain as its leader and new chief of staff, elected a caretaker IRA Executive and Army Council that would be loyal to the traditional principles and doctrines of republicanism.

The new dissident IRA bodies were interim and temporary; the Executive and Army Council could be ratified only by a General IRA Convention, which was not scheduled to meet until September of the following year. But until then they would represent the new group. News of the split and the formation of the rival IRA was leaked to the press by Goulding’s people in late December 1969, and in a statement confirming the story the new group claimed that already a majority of IRA units had sworn allegiance to the Provisional Army Council and Executive of the IRA. The tag “Provisional” stuck not least because it was a handy way for the media to distinguish the new group from those who had remained loyal to Goulding. His faction became known, inevitably, as the Official IRA. The two groups would later feud violently and bloodily about whose members had the right to call themselves republican, but there was little doubt that the new IRA, the Provos, as most activists would soon call the group, had arrived.

Adams’s criticism of the leadership’s handling of this episode, written admittedly after he himself had successfully and painlessly accomplished in 1986 what Goulding had failed to achieve in 1969, suggests that his reservations had more to do with their clumsy tactics than with any qualms he might have had that Goulding was traveling in the wrong political direction. “At the very least,” he wrote, “the leadership should have recognised the need for new priorities and suspended its pursuance of the new departure in republican strategy until a more settled time.” He added, “For many of the dissidents the issue was not abstentionism itself but what it had come to represent: a leadership with a wrong set of priorities which had led the IRA into ignominy in August.”[25]

Having secured the IRA Convention, Goulding pressed Sinn Fein to follow suit at its conference, or Ard Fheis, in Dublin in January 1970. A motion to drop abstentionism was carried but narrowly failed to secure the two-thirds majority needed to change the party’s constitution. A Goulding supporter then proposed a motion of confidence in the IRA leadership, a roundabout way of achieving the same thing. Realizing they were about to be defeated, the dissidents walked out. Later that evening they met to set up an Executive for their own version of Sinn Fein and elected Ruairi O Bradaigh the first Provisional Sinn Fein president.

Adams’s own account has him missing the crucial debate, barred from the venue, Dublin’s Intercontinental Hotel, by a Sinn Fein official because he didn’t have the proper accreditation. Instead he went off to join an anti-apartheid protest at a rugby match between Ireland and the South African Springboks. His version of events is controversial not least because the only witness, the official who allegedly barred him, is long since dead. At the very least Adams’s memory of events is hazy since the walkout took place on a Sunday and the Springboks match was played the day before. Jim Sullivan, McMillen’s adjutant and one of those who stayed loyal to the Goulding faction, has claimed not only that Adams was present during the Sunday session but that he was sitting beside him when the walkout happened.[26] Adams, he asserts, stayed in his seat and remained loyal to the Goulding leadership. Adams has denied this, but if his version of events is accurate and he was among those who did walk out, then he should have been at the meeting that set up the caretaker Sinn Fein Executive. But according to Ruairi O Bradaigh only two Northerners were present when the Provisional Sinn Fein came into being: Mary McGuigan from Ardoyne in Belfast and Liam Slevin from Beleek, County Fermanagh.[27] Adams, he says, was definitely not there.

Whatever the truth, Adams does concede that he did not immediately align himself with the Provisionals, and that admission is consistent with the Sullivan–O Bradaigh version of the 1970 Ard Fheis rather than his own. Before the December Convention, Ballymurphy republicans had asked for a meeting with Sullivan, which debated the IRA’s internal differences, and afterward Adams told Billy McKee, the leader of the Belfast dissidents, that the Ballymurphy unit had decided to sit on the sidelines “until we saw what way things were going to break.”[28] This led to accusations at the time that Adams was waiting to see which side came out on top before joining it. It was not until four months later, in April 1970, that Adams officially brought the Ballymurphy IRA into the Provisionals, by which time the vast bulk of IRA companies in the Belfast Battalion, fifteen out of sixteen, according to one estimate, had deserted the Goulding leadership for the new organization.[29] He had joined the winning team.

2. The Defenders

The new Provisional IRA was led by men who represented traditional, conservative, republican values, who believed that armed struggle was the only way to rid Ireland of the British presence and that parliamentary politics was an endeavor best viewed with suspicion and treated with disdain. A majority on the caretaker Army Council, and later the Army Council elected after the first IRA Convention, had seen active service during the 1956–62 campaign, and most of them came not just from the twenty-six counties that made up the Irish Republic but from predominantly rural counties as well. Only two of the seven were from the North, from where the impetus for the new IRA had come; there was no representative from Dublin, where Goulding’s influence had been strongest. They were all men, all approaching middle age, and all Catholics, some devoutly so.

No one exemplified these characteristics better than the new IRA’s chief of staff, Sean MacStiofain. The son of an Irish mother and English father who was raised in South London, MacStiofain was forty-one years old when he attained the highest rank in the IRA. He had been involved in radical Irish politics since his teenage years and became a fluent Irish speaker. Heavily influenced by his mother’s strongly nationalistic views, he joined the IRA’s English unit in 1949, at the age of twenty-one, but later moved to Ireland and eventually settled in County Meath, where he worked for the Irish language body, Conradh na Gaeilge.

Infused with the special zealotry of the convert, he angrily resisted British media attempts in the early years of the Troubles to anglicize his name to John Stephenson, the designation on his birth certificate. His arrest and imprisonment with Goulding for the Felsted raid meant that, like Goulding, he avoided the internal warring that had plagued the IRA at the end of the failed Border Campaign and escaped blame for the fiasco of 1956–62. His decision to remain inside the organization despite the differences with the Goulding leadership, meanwhile, put him at a great advantage when the split came.

The way MacStiofain chose to express his opposition to Goulding appealed to a very broad section of dissidents, not all of whom were as directly affected by the events of August 1969 as were the Belfast men. He had two celebrated clashes with Goulding, both of which endeared him to the Catholic, conservative center of the republican movement, even though each time Goulding got the better of him.

Not long after Roy Johnston had been installed as adviser to the Army Council, MacStiofain confronted Goulding with Johnston’s past membership in the Communist Party (CP) and its Irish offshoot, the Connolly Association. The IRA’s rule book, known as General Army Orders, specifically prohibited Volunteers from joining the CP; it was the product of an age in Ireland when Sunday Mass would end with a prayer for the conversion of Russia. MacStiofain demanded that Goulding dismiss Johnston, but the chief of staff stood up to him and said if Johnston went, then so would he. MacStiofain was forced to back down.[1] The new IRA commander later explained why he took such a hard line: “We opposed the extreme socialism of the revisionists [the Goulding faction] because we believed that its aim was a Marxist dictatorship which would be no more acceptable to us than British imperialism or Free State capitalism.”[2]

Anticommunism was to become a recurring obsession of the new IRA. The editorial in the very first edition of Republican News, the Provisionals’ Belfast weekly newspaper, set the tone. Republican News, or RN, as it became known, had been set up in February 1970 by two IRA veterans, the dissident leader Jimmy Steele and Hugh McAteer, a former chief of staff from Derry. IRA men thought of it as “the Belfast Brigade newspaper.”[3] Outlining the malign influence of the Goulding supporters, RN railed against them in language that would not have been out of place in a speech by Senator Joe McCarthy: “Gradually into executive posts both in the IRA and Sinn Fein, the Red agents infiltrated,” the paper complained, “and soon these men became the policy makers. Young men and girls were brainwashed with the teachings and propaganda of the new policy makers and well-trained organizers were sent into different areas to spread the teachings of the Red infiltrators.”[4] The same paper, edited by someone who many years later became a leading light in the Irish Tridentine Mass movement, later described the mission of the IRA’s youth wing, na Fianna Eireann, in simple terms: “Our allegiance is to God and Ireland….”[5]


These views persisted well into the 1970s when more and more IRA members found themselves in jail, where debates on left-wing politics became more frequent. Those who espoused such views would be hounded. One former prisoner can recall a member of the IRA jail staff sending a priest to see him because he had been spotted reading a book by James Connolly.[6] The 1916 leader, badly wounded in the Rising and then strapped into a chair and executed by a British firing squad, was in the pantheon of IRA martyrs, but he had been a Marxist and trade union leader as well. The Provisionals’ rivals in the Official IRA revered Connolly, and so any Provo Volunteer who showed an interest in his writings automatically came under suspicion. Even as late as 1975 the leadership had a knee-jerk attitude to left-wing politics. When some IRA inmates of Long Kesh internment camp started dabbling with socialism, an order was issued, according to the former Belfast IRA commander Brendan Hughes, to burn all Marxist books.[7]

MacStiofain regarded men like Roy Johnston as godless atheists who were more interested in undermining the influence of the Catholic Church in Ireland than in ending British occupation of the North. When Johnston wrote a letter to the Sinn Fein paper the United Irishman criticizing the practice of reciting the rosary at republican commemorations, calling the practice “sectarian,” MacStiofain, then living in County Kerry in the deep southwest of Ireland, stopped the offending issue from being circulated and sold, for fear that the letter would be used against the IRA by its enemies. Interfering in the distribution of the United Irishman cost MacStiofain his place on the Army Council, the automatic consequence of a six-month suspension from the IRA imposed by Goulding as punishment.

MacStiofain’s tilts against Goulding earned him plaudits south of the Border, but in the North the way to the hearts of the hard men in Belfast was by a different route. Throughout the eventful and threatening year of 1969 MacStiofain liaised closely with the dissident Northerners and shared their concerns that the increasingly demilitarized IRA would be unable to defend Belfast’s Catholic areas. At one point, according to his own account, he went behind Goulding’s back and secretly supplied guns to some Belfast units.[8] A shrewd if somewhat uncomplicated tactician, MacStiofain realized that the engine of the upcoming war would be in Belfast, and throughout his IRA career, until his fall in 1972, he strove to ensure that his relations with the Belfast IRA were always the best.

Two other figures dominated the new Army Council, both then in their thirties and both Southerners who had distinguished themselves during the Border Campaign. Ruairi O Bradaigh, a schoolteacher from County Roscommon in the west of Ireland, was chief of staff at the end of that campaign. He had written the famous IRA statement ending the campaign and ordering IRA units to dump arms, conceding that lack of popular support from Northern nationalists had dealt the IRA the final blow. O Bradaigh inherited his republicanism from his father, who had been badly shot up by the Royal Irish Constabulary in 1919 and who died when O Bradaigh was only ten. O Bradaigh’s military credentials were impeccable—he had led an arms raid in England that netted five tons of British army hardware—but he was not as opposed to political activity as others were. In 1957 he was one of four Sinn Fein members elected to the Irish parliament on an abstentionist ticket, and throughout his career with the Provisionals he favored selective and tactical electoral interventions. In 1970 he became the first president of Provisional Sinn Fein, and as the custom was that Sinn Fein always had one seat on the Army Council as of right, O Bradaigh automatically qualified.

The other major personality was Daithi O Conaill from County Cork, also a schoolteacher by profession. Unlike Sean MacStiofain, his new chief of staff, O Conaill did not mind if people used the English form of his name, and most Provisional associates and the media called him Dave O’Connell. Like O Bradaigh’s, his military record was beyond criticism. He was interned by the Dublin government early on in the Border Campaign and was incarcerated in the Curragh camp in County Kildare, once the British army’s headquarters in Ireland, and there he met O Bradaigh. In 1957 the pair devised a successful escape plan, and they went on the run together. When O Bradaigh became chief of staff, he chose O Conaill to be his director of operations. It was the start of a long and close relationship.

O Conaill was a hands-on commander and fought in Counties Fermanagh and Tyrone, where he was badly wounded by the RUC. Confined to Crumlin Road prison, Belfast, he appears at first glance to have been an unlikely candidate for leadership of the Provisionals. In jail he had edited the Saoirse magazine, which had first floated the idea of IRA involvement in social and economic agitation, and after the campaign he became interested in the cooperative movement, often a route into left-wing politics. But his time in Belfast jail had also convinced him that the removal of partition had to be the republican priority, and he believed strongly in the need for armed struggle. The job he undertook with the new IRA emphasized that; he was the Provisionals’ first quartermaster general, charged with rearming the organization and giving it the sinews to fight the impending war against the British.

The remaining four Army Council members were also veterans of the 1956–62 campaign: Sean Tracy from County Laois; Leo Martin from West Belfast, who was OC of the Northern IRA; Patrick Mulcahy from Limerick; and Joe Cahill from Belfast. At a second Convention, held in September 1970, Billy McKee joined the Council. The presence of three Northerners on the Council, all Belfast men, serves to dispel the myth that the Provisionals were Southern-dominated from the start and that it was not until Adams and his allies captured the Army Council in the late 1970s that the Northern voice was allowed to be heard at the IRA’s highest levels.

The first Provisional leaders were sure of the rightness of their cause and the reasons for breaking with the Officials. The initial statement from O Bradaigh’s breakaway Sinn Fein in January 1970 listed five reasons for splitting with Goulding: his recognition of the Irish and British parliaments; the move to embrace extreme socialism; illegal internal disciplinary methods; the failure to defend Belfast; and the policy of defending the Northern parliament at Stormont.[9] The list demonstrated that the Provisionals were essentially a coalition of differing grievances; for some, Marxism was the major problem with Goulding, and for others the military rundown of the IRA. One characteristic of the new IRA above all others that united the coalition—the glue that held it together—was a distrust of politics, parliamentary politics in particular, and an unshakable belief in the correctness of armed struggle.

The early Provisional leaders were determined that they would not stray down the path of parliamentary reformism trod by other nationalist and republican leaders. Each previous generation of freedom fighters had been betrayed, they believed, by leaders seduced by the siren call of parliamentary politics. They would be the exception. For this reason they defined the relationship between Sinn Fein and the IRA in simple and traditional terms. The military wing, the IRA, was in charge, and Sinn Fein would obey and be subservient to the Army Council. That was the case in the South and also in the North, where, according to one veteran party activist, Sinn Fein was secondary to the IRA from the outset: “Sinn Fein was the poor relation. It wasn’t worth bothering about. Sinn Fein in the 1970s was an organization without clout; it supported the ‘campaign’ and held lofty ideas of a united Ireland but nothing else. The IRA was boss.”[10]

As the war intensified and more and more Northerners joined up, the antipolitical nature of the Provisionals intensified, as one of the Provisionals’ founding members recalled:

When the resistance began, Northerners came in droves, and they were reacting to events for a number of years. The Northern guys were quite slow to be politicized. They looked down on Sinn Fein and dismissed it saying, “We’re Army men,” I shared a cell with them in Mountjoy, and that was their view. They were quite happy sitting in their cells reading the Sun or the Mirror boasting about operations. They were purely militaristic—hit, hit, keep on hitting.[11]

Whereas the first IRA commanders were Southerners, the foot soldiers in the war, the Volunteers, came overwhelmingly from the North and at first mostly from Belfast, where the attempted loyalist pogroms of August 1969 had taken place. Many IRA units elsewhere in the North, in republican heartlands like Tyrone, Armagh, and Derry, were slower to take sides in the republican split; in some cases months went by before they decided whether to follow Goulding or MacStiofain. The Provisionals were born in Belfast and sustained by the city’s bitter sectarian politics.

Some of those outside Belfast were repelled by the Provisionals’ simplistic politics. Typical of this category was the Derry republican Mitchel McLaughlin, who stayed with the Officials for several months before joining the Provisionals, later rising to become a key Adams aide and advocate of his peace strategy. “At the time of the split,” he once told an interviewer, “I actually stayed with the Official Republican Movement. Mainly because of their politics which undoubtedly were more progressive than the more kind of nationalistic rhetoric that I was hearing [from the early Provisional leaders].”[12] Gerry Adams and the Ballymurphy unit were not the only IRA members to hesitate before taking sides in the split. Not surprisingly, many were waiting to see who came out on top, and so what happened in Belfast was crucial. When Belfast republicanism went over to the Provos, as it did during the crucial year of 1970, many of the rural units followed, and soon if angry young Northern Catholics wanted to hit back at either the loyalists or the British army, they knew they would find a warm welcome in the Provisionals.

The IRA before August 1969 was an organization kept going by family tradition. Membership was passed from father to son, mother to daughter, but the recruits who flocked to the ranks of the Provisionals were a new breed, motivated by an atavistic fear of loyalist violence and an overwhelming need to strike back. Known as Sixty-niners, they joined the IRA literally to defend their own streets, were resolved that the near-pogroms of August 1969 would never be repeated, and were ready, if the opportunity arose, to retaliate. They joined the Provos because the Officials had failed to defend their communities in the way that was expected, and they automatically associated the Officials’ obsession with politics with military weakness and betrayal. From the outset abhorrence of politics and the requirement for defense and armed struggle were just different sides of the same coin.

Typical of the new Provisional IRA Volunteer was Bernard Fox, an apprentice coach-builder from the Falls Road who joined the IRA in 1969, when he was just eighteen years old. He is now a senior figure in the leadership and was named in 2001 in the British media as a senior figure in the Provisional IRA’s GHQ staff. In 2006 he was appointed to the Army Council. He spent nineteen years in prison, either jailed or interned, for IRA activity. His motive for signing up was straightforward, as he once explained in a newspaper interview after the peace process reforms had secured his release from prison: “I was almost shot in a gun attack at Norfolk Street. I came away wanting a gun. It was survival. You wanted to protect your own people… my family and myself. When the barricades went up I wanted a gun so I approached this fella who was in the IRA and asked for gun and he said: could I shoot a British soldier? At that time I hadn’t the idea that it was the British government’s fault….”[13]

Brendan Hughes from the Lower Falls Road district, a figure who later became an IRA legend, was similarly affected by the violence of August 1969. “At that time it was simply ‘Here we are being attacked by Loyalists, by B Specials, by the RUC, by the British army,’ and there was a need to hit back,” recalled the former Belfast commander. “I mean I was in Bombay Street the morning after it was burned out, helping people out, and I went to the bottom of the Falls Road and seen all the burnt-out homes. I had relatives in Bombay Street who were burnt out, and I felt the desire to get back at these people who were doing it.”[14] Micky McMullen, a former long-term IRA prisoner, came under similar pressure but managed to resist it: “Up to 1969 there was nothing, but August 1969 was the turning point. I became involved in community defence you know and stuff like that, helping families to move after they had been burned out. At that time a lot of my friends would have been trying to join the IRA and the rationale would be just to get stuck into the ‘Orangies’ you know. It was a defence thing but something stopped me from getting into that.”[15]

Fox, Hughes, and McMullen and the many hundreds who followed them into the Provisional IRA in the first years of its existence were part of a Northern Catholic tradition that went back nearly two hundred years, when another armed uprising had very nearly ended British rule in Ireland. The United Irishmen’s rebellion of 1798 is celebrated as the moment when modern, secular Irish republicanism was born, but it also coincided with the birth of sectarian politics in Ireland and left a scar that would mark Northern society for centuries to come.

The United Irishmen were mostly Presbyterians whose determination to sever the link with England had been partly inspired by the American and French revolutions but which was also forced upon them by the penal laws, harsh anti-Catholic measures that also penalized Protestants who were not members of the established Anglican Church in Ireland. The leader of the United Irishmen, the Dublin-born lawyer Theobald Wolfe Tone, defined his objectives in heady language that inspired future generations of Irish freedom fighters: “To break the connection with England… and to assert the independence of my country—these were my objects. To… substitute the common name of Irishman in place of the denominations of Protestant, Catholic, and Dissenter—these were my means.”[16]

These were noble sentiments, but the reality on the ground in the late eighteenth century was the reverse of what Tone wanted to happen. Ireland was beginning to divide very deeply on religious and sectarian grounds, nowhere more so than in the North. The cause was simple—land. Protestant planter tenant farmers in the North had enjoyed superior rights over native Catholics since the 1600s when the plantation settlement had accelerated. A practice known as the Ulster Custom gave them security of tenure as long as they kept up their rent payments, and it also entitled them, if they decided to quit the land, to receive compensation for improvements they had made to the land or property. They had an incentive to work and invest that was denied their Catholic countrymen, and the rewards made their sacrifice worthwhile. But in the early 1780s the English began to relax the penal laws, Catholics were allowed to hold leases on the same terms as Protestants, and the planters’ privileged lifestyle came under threat. Catholics, who were used to a lower standard of living, were also ready to pay higher rents, with the result that Protestant planters saw more and more of their farms falling into the hands of their native rivals.

Around the mid-1780s the worm began to turn. The first secret Protestant societies appeared and announced that they were dedicated to preserving privilege and driving the Catholics out. The largest was the Peep O’Day Boys, so called because its members would appear at dawn to burn or intimidate Catholic families. This Protestant violence stimulated a Catholic response, and rival secret societies devoted to defending Catholics were formed. Inevitably they became known as Defenders, and as they toured the countryside raiding for arms and skirmishing with troops, they adopted a strategic dictum that would have been familiar to the Provisional IRA leadership: the best means of defense, they preached, was attack. Although their roots lay in what are now the Border counties between the Irish Republic and Northern Ireland, the Defender phenomenon spread throughout Ireland. The objectives of the Defenders varied from place to place; in the South discontent over high rents and low wages combined with an insurrectionary and defiant politics animated by events in America and France to swell their ranks. But the Defenders were less inspired by the idea of freeing Ireland from British rule and more by the hope that life in Ireland could be improved.

In the North there was a sharp sectarian edge to the Defenders’ activities, and as Protestant anxiety over land losses intensified, clashes with the Peep O’Day Boys became more and more frequent and bloody. One confrontation was to have momentous consequences for Ireland’s history. In September 1795 just outside the city of Armagh, at a place called the Diamond, over twenty Defenders were killed in a pitched battle with a much smaller force of Peep O’Day Boys. The victorious Protestants then decided to re-form themselves into the Orange Society, later called the Orange Order, named after the Protestant hero King William of Orange, who had defeated the Catholic King James in the late 1690s and secured the Protestant crown and succession in England and Ireland.

As the thoughts of the leaders of the United Irishmen moved increasingly to violent rebellion, they turned to the Defenders for the muscle they would need to take on the English redcoats, and the two groups coalesced in 1796. The ensuing rising in 1798 was brutally put down by Lord Cornwallis, who had just lost the American colonies and was determined not to suffer another humiliating defeat at the hands of impudent rebels. Too late to help his associates, Wolfe Tone returned to Ireland with a force of French troops but was captured, arrested, and sentenced to death. Determined to cheat the English hangman, Tone took his own life in prison. The 1798 rebellion went down as another glorious defeat, and Wolfe Tone became known as the founding father of Irish republicanism. His grave at Bodenstown, County Kildare, has become a place of pilgrimage for all shades of Irish republicanism. The Provisionals’ commemoration takes place each June, and the keynote speech is invariably used by the Army Council to spell out current IRA policy.

Tone’s influence survived his death, and so too did that of the Defenders. The Defenders and the early recruits to the Provisional IRA two centuries later shared a number of important characteristics. In both organizations the prime motive for taking up arms was to defend their people from Protestant attack, not to free Ireland. The Defenders organized themselves to protect Catholic farms and land in rural Ulster; the early Provisionals did so to defend Catholic streets and ghettos in Belfast. Both believed that the best defense was attack, and both put the need for guns or other instruments of defense before political ideas.

Both were reactive responses to the iniquities of British rule. In 1795, for instance, a Kildare schoolteacher, Laurence O’Connor, sentenced to death for administering a Defender oath, told his judge, “[P]rosecutions were not the means of bringing peace in the country but if the rich would alleviate the sufferings of the poor, they would hear no more of risings or Defenders and the country would rest in peace and harmony.”[17] Two hundred and five years later Eamonn McDermott, an IRA Volunteer from Derry sentenced to life imprisonment in 1979 for killing a policeman, expressed not dissimilar sentiments: “Its a cliché now but the British Army created the IRA…. They brought the national question into it, before then it wasn’t really an issue. Republicans would argue that it was there from the start, but that’s a load of rubbish. Initially [we] were trying to reform the state; the national question and partition came later.”[18] In their different ways both men were saying the same thing: if the British had behaved sensibly and fairly, they and their communities would probably have been content.

AS CATHOLIC Belfast recovered from the shock and bloodshed of August 1969, the need for defense was uppermost in the minds of republicans in Ballymurphy, where in November 1969, after the death in a car accident of the local IRA commander, Liam McParland, Gerry Adams took charge. In those days the vast sprawling Ballymurphy housing project was divided by the Springfield Road, one of the city’s main arterial routes. The lower, southern side was Catholic, while the upper, northern side, New Barnsley, which dominated the rest of the district from the slopes of hills that overlook the city, was Protestant. Tension, suspicion, and fear were rife after the August riots, and barricades set up then by Catholics were still in place. Ballymurphy Catholics were worried that loyalists could direct sniper fire on them from the high ground of New Barnsley. Eager to acquire weapons of their own, locals turned to the as yet nonaligned IRA unit, as Adams later recalled: “Many people involved with the defence committees flocked to the IRA, which speedily mushroomed out of all proportion to its previous numbers.”[19] There was little doubt that in Ballymurphy defense against loyalist attack was the main motive for those who joined the local unit, as one Volunteer explained: “It was a gut reaction. The Loyalists were attacking our ones and we knew that sooner or later it would come to the ’Murph. The Brits didn’t really figure in those days. They just didn’t belong in the fight.”[20]

At its first meeting, in January 1970, the MacStiofain-led Army Council devised a three-stage strategy that initially placed the emphasis on the need to defend the Catholic areas of Belfast. “All our energies would be devoted to providing material, financial and training assistance for the Northern units,” MacStiofain later explained in his autobiography. “The objective was to ensure that if any area where such a unit existed came under attack, whether from Loyalist extremists or British forces, that unit would now be capable of adequate defensive action.”[21]

As soon as possible thereafter, according to the MacStiofain plan, the IRA would go into a second phase, a mixture of defense and retaliation, the latter designed to deter loyalists from mounting attacks. When the IRA was strong enough and political circumstances suitable, the IRA would move into the third phase, an offensive war designed to bring Britain to the negotiating table to agree to the final withdrawal of its forces from Ireland. Within weeks the Army Council also approved the design for the new IRA’s icon, one that would symbolize the determination that the killings and burnings of August 1969 would never happen again; a phoenix rising from the ashes—meant to be the ashes of Bombay Street—soon adorned lapel badges and ties worn in bars and drinking clubs in West Belfast.

The political circumstances quickly began to move in the Provisionals’ favor. The deployment of troops ended half a century of effective British non-engagement in Northern Ireland. The Harold Wilson cabinet in London realized it had no choice but to force the pace of political change. A reform program was announced. The B Specials, hated by the Catholics, would be disbanded and replaced by a militia to be drawn from both communities. It would be called the Ulster Defence Regiment. The RUC would be disarmed and reorganized in the hope of attracting more Catholic recruits. The allocation of public housing and other powers were taken away from Northern Ireland’s unionist-dominated local councils, while plans were laid to outlaw public expressions of sectarian hatred and to improve community relations. The British home secretary was given oversight of Northern Ireland, and that was a dark hint to Unionists that London could take away their powers entirely; if they didn’t clean up their act, their parliament at Stormont could be suspended and the days of exclusive Protestant rule brought to an end. Although the government of Major James Chichester-Clark had little choice but to accept Wilson’s diktat, the unionist grassroots were alarmed and restive. The disarming of the RUC had led to rioting on the Shankill Road, and three people were killed in gun battles between British troops and loyalist snipers. One of the fatalities was an RUC constable, the first police death of the Troubles. Early in 1970 the Reverend Ian Paisley, already a fairly reliable barometer of grassroots Protestant discontent, won Terence O’Neill’s old seat at Stormont in a by-election called after the unseated prime minister’s retirement from politics. Chichester-Clark was under constant attack from right-wingers on his backbenches, some motivated by hostility to the reform program, others by fear of losing their seats to Paisley supporters. Events on the streets also contributed to the sour unionist mood. In the wake of the August riots, the RUC had been expelled from nationalist areas in Belfast and Derry, and the British army had taken over control of law and order. The defense committees—in many cases controlled by either the Provisional or the Official IRA—banned the police from their areas and refused to dismantle the August barricades. Their behavior infuriated the unionists.

There is little doubt that most Northern nationalists welcomed Britain’s reinvolvement in Irish affairs, especially since it came from a British Labour government, some of whose leaders were known to be sympathetic to the nationalist cause. To the discomfort of many early Provisional leaders, relations between Catholics and British soldiers were, if not friendly, then certainly not hostile, mirroring this new, hopeful mood.

As unionist angst grew, the need to buttress Chichester-Clark against his right-wingers brought a change in British policy. The honeymoon between nationalists and the British began to end in Easter that year, when there were clashes between British troops and republicans at commemorations of the 1916 Rising in Armagh and Derry. But the worst trouble was in Ballymurphy in riots that finally forced republicans there to align with the Provisionals. As in other parts of West Belfast, day-to-day control of law and order in the estate was in the hands of the British army, at first the Royal Scots Regiment, which set up headquarters in a deserted school on the Springfield Road. To the astonishment of Ballymurphy Catholics, at the end of March the commander of the Royal Scots gave Orangemen from New Barnsley permission to parade down the Springfield Road past their estate. As outraged nationalists heard the flute-and-accordion bands striking up sectarian, anti-Catholic tunes, the first stones were thrown. This time the troops sided with the Orangemen. As Catholic rioters hurled rocks and gasoline bombs, the troops replied with CS gas, which seeped through the housing estate, disabling young and old, rioter and non-rioter indiscriminately. Military snatch squads charged the crowds to arrest and beat stone throwers. Ballymurphy nationalists contrasted this tough response with the more low-key and measured reply to loyalist rioters. Chichester-Clark’s critics had demanded that the British army remove its kid gloves in Catholic areas, and this had now happened.

The Easter 1970 Ballymurphy disturbances were significant not just because they pushed the local unaligned IRA unit into the arms of the Provisionals but because, for a sizable section of Belfast Catholic opinion, they began to put British soldiers on the same side as the unionists and Protestants. It was this development, repeated elsewhere with increasing frequency and ferocity in the coming weeks and months, that transformed matters to the advantage of the Provisionals. And no one was better able to take advantage of this than Gerry Adams, the new commander of the Bally-murphy IRA.

By the time of the Easter 1970 riots the Provisional IRA in the city had grown so quickly that it was obliged to reorganize. At that time the IRA operated at three levels. At the bottom were local defense committees, in effect vigilantes who patrolled areas at night to warn of loyalist incursions. Above them was the Auxiliary IRA, men whose job was to defend their own areas, should the loyalists attack. They were in practice part-time IRA members whose services could be called upon in an emergency and who, in the meantime, provided the IRA with ears and eyes. Above them was the full-time IRA unit or company, modeled on the British army’s structure. Each company was recruited from within its home area.

With their ranks swollen by new recruits, the IRA expanded from one Belfast battalion to three, each with its own complement of companies. The First Battalion was based in Andersonstown and the Upper Falls Road area. The Second Battalion encompassed Ballymurphy and the Lower Falls Road area. The Third Battalion took in the rest of Catholic Belfast, principally the isolated and often besieged ghettos of Ardoyne in North Belfast, the Short Strand in East Belfast, and the Markets district in South-central Belfast.

Nearly all Gerry Adams’s family went with the Provisionals. Only one sister, Margaret, whose husband was in the Official IRA, refused; the rest— father, mother, siblings—all gave their allegiance to the reborn phoenix. Adams was well placed to rise in the new movement. At the time of the split there were no more than forty to sixty IRA members in the whole of Belfast and even fewer of Gerry Adams’s youth and ability. The Provisional leadership in Belfast naturally turned to able republicans to show the way to the newcomers, among them Adams. Talent-spotted by the pre-split Belfast commander, Billy McMillen, the young Adams was already headed for leadership, but the Troubles gave him the opportunity to utilize his special skills with deadly effect. Other IRA leaders rose because of their record as “operators,” as gunmen and bombers. But not Adams. Although he was to dominate the IRA for the next thirty years, there is no evidence that he ever fired a shot in anger against the British or their local allies. “I have never met anyone who has ever been on an operation with him,” recalled one early colleague. “Usually you get to hear about people, that so-and-so is a nerveless operator or this one’s a wreck and so on, but never with Adams. He was never on a robbery, never on a gun crew, a bombing or anything.”[22]

The key to his rise lay in other qualities, prime among them the skills of a ruthless general. The first opportunity to display his strategic dexterity came during the Easter riots of 1970. The new Provisional Belfast commander was Billy McKee, a confirmed bachelor and devout Catholic who had devoted his life to the republican cause. Like other conservative critics, he had quit the Goulding IRA in protest against its policies but rejoined when the riots of August 1969 broke out. He became the first commander of the Second Battalion and then, after the split, took over command of the entire Belfast Brigade.

Deeply affected by the IRA’s failure to defend nationalist areas in August 1969, McKee was determined that the Provisionals would make up for that lapse. When the Ballymurphy riots erupted, he ordered an armed unit from D Coy in the Lower Falls Road to go to Ballymurphy and take on the British army. When Adams found out, he was furious and detained McKee’s men at gunpoint. “Adams put us all in this house and wouldn’t let us out,” recalled an IRA colleague who was impressed by this first encounter with the Ballymurphy leader, “and there we sat with all our guns stacked against the wall. McKee wanted a gunfight, but Adams didn’t. Adams wanted ordinary people involved in the rioting as a way of radicalizing them. That impressed me. He seemed to be very competent and capable; he knew what he was talking about.”[23] The rioting lasted for four days and affected thousands of West Belfast people. Had the IRA opened fire on the first day, the trouble would have possibly been over in a few hours and could have ended with the IRA’s defeat at the hands of superior British firepower and Ballymurphy’s substantial Catholic community rendered less angry, less willing to take up guns.

Three months later the chance came to repeat the exercise. The months of June and July 1970 were to witness a series of historic blunders by the British military and the unionist government which catapulted the Provisionals into an organization large and strong enough to embark on the second and third stages of MacStiofain’s strategy. The riots of that summer also made the Ballymurphy IRA the most militant in Belfast and helped ensure that Gerry Adams would rise through the ranks.

The spark that lit the fire came on June 18, when in Britain a general election ousted Harold Wilson’s Labour government and installed a Conservative administration led by Edward Heath. The result buoyed the unionists in Northern Ireland. Harold Wilson’s cabinet had been openly sympathetic to the civil rights agenda, whereas the British Tories had historical and cultural ties to unionism. Encouraged by the result, unionists began to press for harsher security measures, including internment, the measure that they believed had crushed the IRA during the 1956–62 campaign. But there were warning signals from the election as well. The Reverend Ian Paisley was elected to the Westminster parliament at the expense of a more moderate unionist, and there was a fear that mainstream unionism might lose control unless it was bolstered. It was in this atmosphere that Chichester-Clark secured the agreement of the British army commander in Northern Ireland, Lieutenant General Sir Ian Freeland, that a number of impending Orange parades would be allowed to go ahead. The unionist premier believed he could not survive much more grassroots antagonism, and the British, desperate to sustain Chichester-Clark, agreed.

Two controversial parades were scheduled for June 27. One was due to march past the Ardoyne district of North Belfast, where some of the worst loyalist violence of August 1969 had happened. Another would go through Ballymurphy, as the Easter march of that year had done with such disastrous consequences. Predictably both marches degenerated into violence. In Ballymurphy fierce rioting broke out, again between British troops and local Catholics, but in Ardoyne things took a more serious turn. The Ardoyne parade was attacked with stones and gasoline bombs, and the guns quickly came out on both sides. In August 1969 the loyalists and RUC had given local nationalists a bloody nose, but this time it was the Provisionals who meted out the punishment. Three loyalist snipers were shot dead by gunmen from the IRA’s Third Battalion and several more were wounded.

As news of the killing spread, angry loyalist mobs gathered in East Belfast and began moving in on the small Catholic ghetto of Short Strand, situated on the banks of the river Lagan. Inexplicably the British army made no effort to stop them, although it was clear that the loyalists were intent on death and destruction. It was a heaven-sent chance for the Provisionals to demonstrate they could defend their community, and a small group of IRA Volunteers took up position in the grounds of the local Catholic church, St. Matthew’s, and opened up on the Protestant crowds. Loyalist snipers returned fire, and soon a full-scale gun battle was raging. The IRA squad was led by the Belfast commander, Billy McKee, his presence testimony not just to his devotion to the Catholic church but to his determination to expunge the shame of August 1969. Although McKee was badly wounded in the five-hour-long exchange, his mission succeeded. The loyalist mobs retreated, carrying two of their number home dead. One local Catholic, who IRA leaders claimed was a member of the Auxiliaries, was also killed, but there was no doubt that the events in Ardoyne and the Short Strand— “the siege of St. Matthew’s” in Provisional folklore—had established the Provisionals as the Catholics’ only reliable defenders. The British army was discredited. The importance of the gun battles did not escape the IRA, as one of those involved in the Short Strand battle recalled: “It was very significant. On our way over to the [Short] Strand that night there was a lot of Brits and peelers [police] just sitting outside the area. Did they allow that battle to develop to sicken the Protestants and Catholics? A lot of people joined the Republican Movement after St Matthew’s. It finished the business of IRA equals ‘I Ran Away.’ If that trouble had not broken out the IRA was dead.”[24]

The IRA’s success cleared the way to implement the offensive stage of MacStiofain’s battle plan. In the past small Catholic areas like Ardoyne and the Short Strand had been held hostage by unionism for the good behavior of republicans elsewhere. This was one of the reasons the IRA had confined its 1956–62 campaign to rural Border areas. But now that the loyalists had been repulsed in East Belfast, the most vulnerable Catholic district in the city, the IRA could contemplate taking the war to the British, knowing areas like it could be adequately defended.

There then followed one of those events that historians may never be able to properly explain but whose consequences were plain for all to see even at the time. On Friday, July 3, British troops descended on a small, four-room house in the Lower Falls Road district and began tearing it apart in a search for arms. These they soon discovered, and it was obvious they had been tipped off. The puzzling question, though, was why the military had acted on the intelligence. The arms dump belonged to the Official IRA, the Provisionals’ rivals, and the area was at the time largely sympathetic to the Goulding faction. Although forced by events to adopt a more militant pose, the Official IRA’s leadership was not spoiling for a fight with British troops. The arms raid risked forcing the Officials into retaliation but did the Provisionals no harm at all. If the British were unable to distinguish between the two IRAs, as some observers have suggested, then they made a monumental blunder, for it was the Provisionals who gained most from the ensuing events.

As the troops prepared to leave the Falls Road, they came under heavy stone-throwing from a large and angry crowd. Some soldiers were trapped in the maze of narrow unfamiliar streets, and reinforcements were sent for. Soon three thousand troops were deployed in the area that was now cloaked in CS gas and littered with rubble and exploding gasoline bombs. Freeland ordered a curfew, and troops began house-to-house searches, in many cases wrecking homes, bars, and businesses. The soldiers looted virtually every public house in the area. Gun battles broke out between the Official IRA and the British, and then the Provisionals’ D Coy joined in. The curfew lasted until Sunday afternoon, when it was broken by over one thousand women, many with babies in carriages, who marched from Andersonstown with milk and bread to feed the besieged inhabitants. Four men, all civilians, were killed by the British army that weekend, three by bullets and one crushed by an armored car in circumstances that made the Provisionals’ violent strategy seem unavoidable, appealing, and even necessary to many West Belfast Catholics. The Falls curfew, as the events of that weekend became known, marked a victory for the tactic of armed struggle.

Provisional IRA gunmen had opened up in the Falls Road, Ardoyne, and East Belfast; in Ballymurphy, however, Gerry Adams’s constraining order on the local unit still applied. As the Protestants of New Barnsley evacuated their homes, the riots that had started on June 27 continued almost nightly and lasted for the next six months. Ballymurphy republicans were in the thick of the fighting, but not once did they break open their arms dumps. The Belfast Brigade was unhappy at the rioting and wanted to see some armed action, but Adams persisted, determined to radicalize his own people.

The strategy was effective, but it was also ruthless. During the six months of riots that followed, hundreds of people were injured or arrested and imprisoned; others, old and young alike, suffered from the effects of CS gas. Dozens of young people joined the IRA to seek revenge against the soldiers; some were to end up in jail, others in early graves. But the effect of the rioting was profound, as Adams later boasted: “Every man, woman and child was involved. They didn’t fire a shot but for months, the British army had the hell beaten out of them. The women were humiliating and demoralising them. The kids were hammering them. You had the whole community organised right down into street committees, so that you had a sort of spider’s web of regular coordination.”[25]

The result was that Ballymurphy became the most militant republican district in the city, and its IRA units made the Second Battalion area the most active and ferocious in the Belfast Brigade. One crucial consequence of this was that, as the IRA’s war intensified, the brigade leadership in Belfast relied heavily on figures drawn from the Second Battalion area. This was Gerry Adams’s route to the top of the IRA.

BY CHRISTMAS 1970 the Provisional IRA was just a year old, but it was a vigorous and rapidly growing infant. A unique combination of events had come together to create the Provos, and during its first year of existence another series of extraordinary, unpredictable incidents—the siege of St. Matthew’s, the Falls Road curfew, and the Ballymurphy riots prime among them—had given the organization an undreamed-of boost. It would be tempting to conclude that fate was playing an awful trick on the people of Northern Ireland by arranging matters in such a disastrous fashion, but there was a common factor in all these happenings that explained why events were spiraling out of control. Just as unionist obduracy had played the role of midwife to the new IRA, so the same need to placate Protestant extremism and prop up the Stormont government led Britain to take an increasingly tough line against the communities from which the IRA sprang. As that conflict worsened, the notion that as long as unionists held power and were supported by Britain, nationalists could expect no fair dealing gained more support and sympathy. The events of 1970 nourished the view that Northern Ireland was incapable of being reformed and that only its destruction could end the nationalists’ nightmare. For the first time in the history of the state the extreme republican agenda and the IRA’s violent methods were winning the allegiance of a sizable section of the Catholic community. Unionists had created the Provos, and now they were sustaining them. Only in Ballymurphy had republicans manipulated events; everywhere else there was no need to.

3. “The Big Lad”

The year 1970 was without doubt a seminal one for the IRA, as an early Volunteer recalled in vivid terms: “At the start of it I remember picketing outside British Army dances with placards and being spat at by local [Catholics] who attended those dances. We were nothing at that point. By the end of the year we had an organisation capable of taking on the British Army.”[1] A few months later the war between Britain and the Provisional IRA was raging, and the most violent conflict in the tangled and tragic history of Britain and Ireland was under way.

As 1970 drew to a close, the IRA’s leaders could be confident that their fortunes were improving and that 1971 would almost certainly see that trend continue, although not even the wildest-eyed IRA man could have foreseen just how sharply the curve would rise. In Gerry Adams’s home estate in Ballymurphy, as elsewhere in Belfast, the realization that a new plateau had been reached forced a change in tactics.

The riots in Ballymurphy faded around Christmas 1970, but in the middle of January 1971 they flared again with greater ferocity. Now, however, Adams and the Ballymurphy republicans were ready to bring them to an end, to patch up their differences with the Belfast Brigade leadership, and to unleash the gunmen. The local historian Ciaran de Baroid explained why: “The Ballymurphy republicans felt that [the riots] had served their purpose; the people were cemented together; the British army was humiliated and demoralised; alienation between the people and the state was complete and irreversible; and self-confidence and an efficient infrastructure of organisation had been developed within the area.”[2] Ballymurphy IRA members moved to quell the rioting, and British army commanders were not slow to notice the change in tactics.


Discreet negotiations were opened up between them and a policing deal was struck in which, according to the IRA version, the Army agreed that the RUC would stay out of the area, and policing in the Second Battalion area would be left to the IRA. The British would also evacuate the area. Almost immediately the deal, if such it was, fell apart, and seven hundred troops invaded the estate and began searching for arms. Fierce rioting once more engulfed Ballymurphy. Then the Reverend Ian Paisley announced that he had found out about the secret talks, while an increasingly beleaguered Chichester-Clark extracted a pledge from the Tory home secretary, Reginald Maudling, that British troops could now take “the offensive” against the IRA.

On February 2, 1971, in response to pressure from unionists and ministers in London, British soldiers cordoned off the Clonard area of West Belfast, the scene of the burning of Bombay Street in August 1969, and began a punitive series of house searches. These sparked vicious riots, in which loyalist workers from a nearby engineering factory joined. Clonard was in the Second Battalion area and was supposed to be covered by the secret policing deal. IRA commanders saw this as evidence of British army bad faith and responded accordingly. The Clonard riots spread to other Catholic districts, and soon the IRA went into action. British troops came under gun and bomb attacks in North Belfast, and then on the night of February 6 an IRA unit under the command of Billy Reid from the New Lodge Road area ambushed a British patrol, killing one soldier, Gunner Robert Curtis, and wounding four of his colleagues. He was the first of 503 British military personnel to die in the Troubles.[3]

By the end of 1970 support for the Provisionals in Belfast was increasing in direct proportion to the weight of the British military presence in nationalist areas. The unionist government’s answer to every downward twist in the cycle of violence was to ask Britain to send more troops to police nationalist districts. Invariably the British refused the more outrageous demands, but unionists still got a good deal of what they wanted. Chichester-Clark was in a strong position. Without him the British would have been forced to intervene directly in Northern Ireland’s affairs and might well have had to abolish the local parliament and government and rule directly from London, a leap in the dark that no British leader yet wanted to take. That might have had to happen at some stage, but in the meantime, the British calculated, the only viable policy was to give the unionist leader what he needed to stay in office.

Sending more troops into nationalist areas was, however, a bit like throwing gasoline on a fire. The troops were trained battle soldiers, not policemen; their instinct was to hit every problem on the head with a club, and, with some exceptions, few of their officers showed much understanding of the political cauldron in which they had landed. Some, like the Parachute regiments and the Marines, were particularly ill suited to the delicate task of keeping peace and quickly developed a name for casual, horrific brutality. For some incomprehensible reason the British insisted on sending into some of the worst troublespots, like Ballymurphy, Scottish regiments, many of whose recruits came from Orange backgrounds in Glasgow and elsewhere and were every bit as staunch as their Belfast brethren. If the riots in Ballymurphy were particularly bitter and bloody, it was due in no small measure to this ingredient.

The troops were also operating in a political environment that became more and more warlike, not least because the views of their commanders were hardening. During 1970 British commanders had given their troops permission to open fire and shoot dead any gasoline bombers who ignored warnings, and after the death of Gunner Curtis, Chichester-Clark declared that his government was at war with the Provisional IRA. As the rhetoric grew more bellicose, the troops on the ground became more aggressive, and increasingly the operational distinction between ordinary Catholics and IRA activists became blurred. Belfast Catholics repaid the soldiers’ hostility and anger with interest, and a self-perpetuating, self-nourishing cycle began that no one seemed able to break. It was all a boon to the fledgling IRA, as MacStiofain recognized: “The fact was that… the British soldiers caused friction, resentment and problems that had not been there before.”[4]

By 1971 there were unmistakable signs that the Provisional IRA was getting bigger and more dangerous than Goulding’s Officials. In June 1970 some five thousand people had attended the Provisionals’ Bodenstown ceremonies; within twelve months the crowd had nearly tripled, to fourteen thousand. This was reflected in growing IRA recruitment figures. The IRA Convention of September 1970, which regularized the makeup of the Army Council and Executive, was, in MacStiofain’s opinion, the largest he had ever attended, as was the following month’s Sinn Fein Ard Fheis.[5] In Belfast the Provisionals had scored two moral victories over their rivals in the Officials, both in Adams’s Second Battalion area, and these resulted in a shift of support to them in a key district of the city.

It was Adams’s old colleagues in D Coy in the Lower Falls Road area who swung matters the Provisionals’ way. The members of D Coy had performed well against the British army, better than the Officials, some believed, during the Falls curfew, even though the fighting had taken place in an Official IRA stronghold. Then, in March 1971, D Coy’s commander, Charlie Hughes, was killed by the Officials in circumstances that strongly suggested duplicitous dealing by the Goulding supporters. A feud had broken out between the two groups, and each side had kidnapped hostages. Negotiators intervened and patched together a settlement, but just as rival leaders endorsed it, Hughes was shot dead. Hughes was widely respected in the area; he had fought off a loyalist mob with a Thompson machine gun during the August 1969 riots, and it looked as if the Officials had killed him out of resentment at his local standing. The Officials claimed that the responsible gunman had not been told of the mediation in time, but few in the Lower Falls believed them and local sentiment shifted to the Provisionals.

With the political and military circumstances shifting in their favor the Provisional leaders decided to intensify the campaign, to move from defensive mode into retaliation and attack. The way the IRA signaled the change was also a harbinger of the depths to which the violence could and would descend in the years to come. On the evening of March 8 three young members of the Royal Highland Fusiliers, two of them brothers aged seventeen and eighteen from the Scottish town of Ayr, were lured from a bar in central Belfast and taken to a lonely mountain road overlooking the northern suburbs of the city; there they were each shot in the head, apparently as they were relieving themselves after the night’s hard drinking. The extreme youth of the victims, the fact that the killers would likely have known that two were brothers, and the supposition that girls may have played a part in enticing them to their deaths made it an operation that the IRA was not keen to boast about. The IRA issued a statement actually denying responsibility, and the episode was written out of the organization’s annals. The killings, for instance, did not feature in MacStiofain’s own account of this period or in the IRA’s official history of its early years,[6] but the truth was that the operation had been authorized by the Belfast Brigade and was carried out by personnel from the city’s Third Battalion.[7]

Such ruthlessness paid off. The loyalist reaction was instant and widespread, fueled in no small way by the cultural and ethnic links many Protestants had with the part of Scotland the soldiers hailed from. Thousands of Protestants marched to the center of Belfast, where the Reverend Ian Paisley led them in an impromptu memorial service; angry Protestant workers in the shipyard laid down tools and joined them. Teenage loyalist bands calling themselves Tartan Gangs in tribute to the dead soldiers appeared on the streets, adding a violent and unpredictable element to the growing unionist unease. Within days the beleaguered Chichester-Clark resigned. The shipyard workers and other loyalists had demanded that IRA leaders be immediately interned, and the Ulster Unionist Party’s choice for their new leader was evidence that this measure was now on the agenda. Brian Faulkner, an ambitious member of both O’Neill’s and Chichester-Clark’s cabinets, had been minister of home affairs during the IRA’s 1956–62 campaign when the use of internment on both sides of the Border had helped end the IRA’s violence. Convinced that internment could and would work again, he promised his hard-line supporters that his would be a law-and-order administration.

The prospect of internment also alarmed the IRA leadership, especially in Belfast, where the killing of the three Scottish soldiers and Faulkner’s elevation combined to produce leadership changes that would have profound and lasting effects on the course of the Troubles. In April, days after the killings, the Belfast commander Billy McKee and his adjutant, Proinsias MacAirt, or Frank Card, as the British insisted on calling him, fell foul of a new get-tough approach. They were stopped by a military patrol in West Belfast, and their car was searched both by the soldiers and by Scotland Yard detectives, who were on secondment to the military because of the RUC’s inability to operate in nationalist districts. One of the detectives triumphantly brought out an automatic pistol. McKee and MacAirt were arrested and charged. The IRA alleged that the men had been framed. Both men had participated in the secret talks about policing Ballymurphy, and the British knew exactly how important they were. They were also under constant surveillance. A powerful British army searchlight nightly illuminated the front of MacAirt’s house in the Lower Falls, where early brigade staff and other IRA meetings had been held. The pair knew how foolish it would be to drive around the city carrying arms. Their defense, however, was rejected and each was sentenced to five years in jail.

McKee’s removal created gaps in the Belfast leadership’s battle order. The Second Battalion commander, Joe Cahill, took charge of the Belfast Brigade while Seamus Twomey replaced MacAirt as second in command and was made brigade adjutant. Cahill took McKee’s position on the Army Council, thereby preserving the precedent set at the September 1970 IRA Convention that the Belfast Brigade leadership would be heavily represented at leadership level. Twomey, a veteran of the 1940s’ IRA campaign, had been in charge of building up the Auxiliaries, but now he was set on a course for national prominence. Known as Thumper because of his habit of slamming the table with his fist when angered, Twomey would later leave an indelible imprint on the history of the Troubles.

McKee’s departure opened opportunities for Gerry Adams. McKee disliked Adams and distrusted his motives for joining the Provisionals, believing that he was secretly sympathetic to the Goulding faction. The April 1970 riots in Ballymurphy had soured relations between them even further; Adams’s defiance had angered McKee, and it was clear that as long as McKee was in charge of Belfast, Adams’s chances of promotion would be slim. “He always thought of him as a Stick,”*recalled a contemporary.[8]

Temperamentally the two were at opposite poles. McKee was first and foremost an operator and believed that the best way to lead was by example. When the Short Strand was under attack in June 1970, he immediately drove over from West Belfast to take his place beside the East Belfast IRA, crouching behind crumbling headstones to return the fire of loyalist snipers. The wounds he received that night left him with a permanent illness. Adams on the other hand led more from behind, earning a name not for his physical courage or operational valor but for his organizing abilities and tactical canniness.

The British army had unwittingly cleared the way for Adams to rise through the ranks. Not only was McKee now out of the picture, but the elevation of Joe Cahill, a longtime IRA contemporary and friend of Gerry Adams Sr., meant that Adams had a friend at the top of the organization. As new Belfast commander, Cahill had the right to appoint his own commanders, and so in April 1971 he put Adams in charge of the Second Battalion. Adams had been Second Battalion adjutant when Cahill was battalion commander and according to one contemporary had managed to exercise “great influence” over his superior officer.[9] Now Cahill’s old job was his, while Seamus Twomey became Cahill’s new deputy, the adjutant of Belfast Brigade. Twomey was another friend of the Adams family, and both he and Cahill looked to Adams as someone who knew the minds of young Volunteers who were now flocking to the IRA. Adams’s influence over Cahill and Twomey gave him a line right into the Army Council. Only fourteen months after the birth of the Provisional IRA, Gerry Adams’s voice would be heard, albeit indirectly, in its highest councils.

The appointment of Faulkner and events on the streets combined to put the Belfast Brigade, and in particular the Second Battalion, in combative mode. It was clear that Faulkner would press the British to introduce internment, and since the IRA itself, from bitter experience, knew how effective a weapon it could be, it was important that the threat was met correctly. Adams realized that the great weakness of the British was their lack of hard intelligence on the burgeoning IRA. The no-go areas in Belfast limited the ability to mount searches, screenings, and various other surveillance and information-gathering exercises while the IRA itself had changed radically, expanding way beyond the small, well-known family networks of the 1940s and 1950s to include people who had no republican background and were unknown to the RUC’s intelligence wing, the Special Branch.

Adams decided to force the pace, as a contemporary recalled:

[I]n the Second Battalion, the leadership, which was making much of the running at that time, were aware that internment was on the way in so we took a strategic decision to force the British hand on the matter. We mounted a concerted bombing campaign against the barracks in the battalion area. Some of them we bombed two or three times. We could not afford to allow them to bring in internment when they were ready. Had it been introduced a year later the British would very probably have had their intelligence act together and would have hit us badly. We knew at the time that their intelligence was bad so it was to our advantage to force internment much sooner than they would have liked.[10]

The official explanation given by the IRA for launching its bombing campaign in April 1971 was twofold—it would stretch the British army on the ground, and it would inflict economic damage, which the exchequer in London would have to pay for.[11] But the truth was that the IRA wanted to force Britain into premature and hasty action. The number of bombing operations, mostly in Belfast, steadily rose: 37 in April, 47 in May, 50 in June, and 91 in July. The targets were not just military and police bases but increasingly included government and commercial premises.

As the summer progressed, the pressure for internment and on Faulkner grew. In the early hours of the Twelfth of July, ten bombs exploded along the route in Belfast to be used by Orangemen later in the day. During the same week a brand-new printing plant built on the western edges of Belfast for the Irish edition of the popular British tabloid the Daily Mirror was destroyed in a daylight raid by a large IRA force. The bombing was a serious blow to Faulkner’s policy of attracting foreign investment to his ailing economy. By then the number of British troops on duty in Northern Ireland exceeded ten thousand, and the Tory home secretary, Reginald Maudling, declared that a state of “open war” now existed between the IRA and British forces. Faulkner calculated that if he did not get internment he was doomed, while the British still saw propping him up as preferable to dismantling unionist rule, a course they feared would cause more Protestant unrest.

On August 9, 1971, a day earlier than planned because of widespread rioting in Belfast, troops raided hundreds of homes in the hope of arresting and interning IRA leaders and activists. But the operation was every bit as disastrous and counterproductive for the British as Gerry Adams and his colleagues in the Second Battalion had hoped. As they had forecast, RUC Special Branch and British army intelligence on the IRA was either hopelessly out of date or inaccurate, and only a few handfuls of activists were rounded up in the initial swoop. Forced into premature action, the British army seized people who were in no significant way central to the IRA’s war effort. The IRA also had excellent intelligence and knew several days beforehand that the raid was about to happen. “Those capable of running an effective war machine escaped,” recalled one of Adams’s Second Battalion colleagues, “and went on to direct the war.”[12] To rub salt in the British wounds, Joe Cahill held a press conference in the heart of Ballymurphy right under the noses of patrolling troops to declare that the IRA was intact. The press conference was organized by Gerry Adams, who had taken on the role of media adviser to the Belfast commander. It was evidence of Adams’s great media and PR skills, which he would put to good use throughout his career.

Internment was a triumph for the IRA in political terms as well, not least because it had been introduced in such a completely one-sided way that its effect was to enormously increase nationalist alienation on both sides of the Border. Although loyalist violence was also growing, the operation was directed solely against republicans, and even then political activists who were in no way associated with the IRA, student civil rights leaders, for instance, were included in the swoop. In Dublin the prime minister, Jack Lynch, had been toying with the notion of introducing the measure in tandem with Faulkner but the one-sided nature of the Northern operation meant he had little choice but to abandon the idea. As sympathy for their cause in the Republic exploded, IRA fugitives could now find sanctuary across the Border, safe in the knowledge that the Gardai would not throw them behind bars. Internment also pushed the levels of violence to record heights. Sectarian rioting flared across the city, driving up to seven thousand Catholics and two thousand Protestants from their homes, while for several days fierce gun battles raged, in many cases pitting loyalists and British soldiers together against the IRA. Once again the IRA could boast that it had answered the call to defend Catholic areas.

The figures spoke for themselves. In the whole seven months up to August 9, 34 people had been killed in conflict-related incidents, but in just three days following internment 22 people died violently. The death rate continued at a high level afterward; a further 118 were to die during the rest of the year, an average of nearly one a day. In Ballymurphy, British troops were involved in two days of savage gunfire and violence, which left a Catholic priest and 7 civilians dead, shot in circumstances that led to allegations the troops had killed with wanton abandon. There was little doubt that internment had exacerbated the violence.

The consequent alienation and anger in the Nationalist community took two catastrophic forms. Scores of young men and women, eager to strike back, flocked to the IRA, while older and more moderate nationalists registered their disgust by resigning from public positions. At Stormont, the nationalist opposition party, a pro-reform coalition called the Social Democratic and Labour Party, or SDLP, had already withdrawn in protest against British security policy, but now its leaders announced plans to establish a rival parliament. Internment had united Northern Catholics against the state in a way nothing else had done since 1921. It also soiled Britain’s name abroad and brought protests from respected human rights activists and intellectuals. Special interrogation methods used against twelve of those arrested landed Britain in the European Court of Human Rights, accused by the Republic of Ireland’s government of breaching the human rights charter and found guilty, the first of many occasions in which events in Northern Ireland would see an embarrassed British government carpeted at the international tribunal.

As the violence intensified and it became clear that internment had failed, unionists looked elsewhere for scapegoats and revived an old favorite—the IRA, they said, was still active because the Border with the Irish Republic was wide open. This simplistic conclusion flew in the face of the reality that most of the violence was taking place in Belfast, at least forty miles from the Irish Republic and rather too far for lightning cross-Border raids. Nonetheless in mid-October the British army obliged unionist anxieties and began to crater Border roads with explosives in an effort to make them unusable. In response, angry members of local farming communities who needed the road links to conduct day-to-day business promptly filled in the craters. The British army would arrive to stop them and there would be riots, often spread across fields.

The effect of all this was to antagonize a broad swath of rural Catholics and to energize the IRA outside Belfast, in Counties Tyrone, Armagh, and Fermanagh in particular, where new units, battalions, and brigades of Provisionals were formed or expanded. Existing units that were still unsure of their allegiance after the 1969 split now decided to plump for the Provisionals. Internment enlarged the IRA into a six-county-wide army and transformed it into a force that could now seriously challenge British rule in Northern Ireland.

Inasmuch as Gerry Adams’s Second Battalion had, by its bombing and his strategic foresight, helped to precipitate this disaster for British policy in Northern Ireland, the IRA had benefited from his strategic talents, and this enormously boosted his standing within the IRA. Cahill and Twomey became even more dependent upon his advice. Not the least of the effects of internment was that the IRA’s ranks were filled with new, angry, young recruits. The folklore of the IRA at this time is full of stories of young men and women rushing to join, some returning from as far away as North America and Britain. Numbers in the Second Battalion in particular soared, and soon its four companies were each able to field up to 100 volunteers. D Coy led the way with 120 members on active service at one point. By the end of 1971 the IRA in the whole of Belfast was more than 1,200 strong, a far cry from August 1969 when the entire organization was hard-pressed to mobilize more than 50.

With its ranks bloated, the IRA went on the offensive. All units, not just those in Belfast, were encouraged to take part in a commercial bombing campaign against businesses and offices. IRA operations multiplied. The number of bombings rose to nearly 200 in September 1971, the first full month after internment. In its official history the IRA claimed that all but a tiny number of the violent incidents, shootings as well as bombings, logged by the British after August 9 were its responsibility: 999 in September, 864 in October, 694 in November, and 765 in December.[13] One weekend in November saw no fewer than 100 IRA attacks, 60 of them carried out on the first day.[14] The death toll also soared. Killings by the IRA climbed to 86 in 1971, more than four times the number in the preceding two years, while those ascribed to the British army rose by more than sixfold, to 45. Forty-four soldiers were killed in 1971, more than two-thirds of them after August 9, while IRA casualties rose threefold, to 23, all but 4 of whom died after internment. Civilians made up the largest category of violent deaths, as they would throughout the Troubles; 92 died in 1971, compared with 16 since 1969.[15]

During all this time Gerry Adams led a highly furtive and clandestine existence. This was especially so after August 9, when British troops raided widely in West Belfast in the hope of catching those who had escaped the first swoops. Like other senior IRA figures, he moved from one safe house to another to sleep and eat—“billets,” in IRA language—and not always in the Second Battalion area. It was during this period that some IRA leaders realized that South Belfast, a mixed, mostly middle-class area with a large transient student population, was an ideal hiding ground, and they based themselves here. When it was necessary to talk to other members of his battalion staff or to pass on instructions, Adams used “call houses,” operational HQs in the houses of sympathizers in West Belfast, where it was considered safe to meet but not to stay overnight.

Generally, as he related in his autobiography, he rarely ventured onto the streets during daylight hours. As a result Adams was something of a mystery figure to the British army. The Ballymurphy riots had made him a well-known IRA personality, but his precise rank and status in the organization remained a puzzle to the authorities. One former British intelligence officer recalled the poverty of intelligence on him. “We had a trace on Adams as Second Battalion staff, Lower Falls, St. James, the Rock, and Ballymurphy, but not much more than that. We had no trace of his involvement in any act of violence except what you would call B2 grade; in fact we didn’t have much intelligence on him at all. We certainly didn’t know where he was when he was ‘on the run.’”[16]

The British army was not even very sure what Gerry Adams looked like. Only one photograph of him existed; it showed a bespectacled, clean-shaven figure wearing an IRA beret at the funeral of the Provisional icon Jimmy Steele, whose coffin he had helped to carry. Now that he wore a beard his appearance was an even greater mystery. Adams’s own story seems to bear this out. In his autobiography he writes that British soldiers kidnapped his pet dog, which they took with them on patrol around Ballymurphy, apparently in the hope that the animal would identify him. Had they been sure what he looked like, that would not have been necessary. His account of his life in this period is littered with stories of near-escapes, of being stopped, questioned, and let go by British patrols who failed to recognize that one of the most senior figures in the Belfast IRA had just slipped through their fingers.

At rank-and-file level in the IRA it was a different story. Adams was beginning to acquire a celebrity status that he strengthened by placing a distance between himself and ordinary IRA Volunteers. He was never one to go for a drink with the lads or to hang around the many illegal drinking clubs that were sprouting up all over West Belfast. And he acquired a prestigious nickname to fit his image, as one IRA contemporary remembered: “Adams was talked about with great reverence. It was ‘The Big Lad says this and the Big Lad says that.’ Adams loved being called the Big Lad because it evoked images of Michael Collins, the Big Fella.”[17]

If the Second Battalion and the Ballymurphy IRA had acquired a fearsome name by mid-1971, it was due largely to the activities of a small group of “operators” Adams had recruited to A Coy in Ballymurphy. Jim Bryson, Tommy “Toddler” Tolan, and Paddy Mulvenna were Ballymurphy’s equivalent of Dan Breen and Sean Treacy, legendary gunmen of the 1919–21 Anglo-Irish war, and their often bloodcurdling exploits helped to construct the Adams myth. “They were the brawn, he was the brain,” remembered a contemporary.[18]

The most fearsome of the three was Bryson, who took command of A Coy when Adams was promoted to lead the Second Battalion. A terrifying and even reckless figure, Bryson would think nothing of patrolling the streets of Ballymurphy armed to the teeth on the off-chance of meeting a British patrol. Other members of A Coy lived in terror of being ordered to accompany him. “He was a controlled psychopath, someone with ice water in his blood. He would do things no sane man would ever consider,” concluded an IRA colleague.[19] His favorite weapon was a vintage Lewis machine gun, known as Big Louie, with which he terrorized the British army. But he also earned a name as a deadly one-shot sniper. His relationship with Adams was complicated. Since Bryson’s death Adams has claimed him as “a dear friend of mine,”[20] but contemporaries say the friendship was not returned. “Bryson didn’t trust Adams, because he had never fired a shot,” remembered one. “He was such a hard bastard, and I think Adams was basically frightened of him.”[21] When Adams needed to curb Bryson, to put him on a leash, he would send someone else, usually a fellow operator for whom Bryson had respect. He never did the job himself.

Bryson died in September 1973, a few days after being shot by an undercover British army unit in Ballymurphy. He was on the run at the time, as was Toddler Tolan, who survived him by a mere four years. The pair became IRA legends when in January 1972, along with five other IRA prisoners, they succeeded in escaping from the prison ship Maidstone, which had been berthed in Belfast docks first as military accommodation and then as an overflow to the city’s Crumlin Road prison, where there was just not enough room to hold the scores of IRA internees. Within hours their feat was being celebrated in song. Tolan was shot dead during a vicious Provisional IRA–Official IRA feud in 1977, while the third member of the group, Paddy Mulvenna, a brother-in-law of Gerry Adams, was killed by the same covert unit that fatally wounded Bryson. Following the escape Bryson had insisted he stay in Belfast to fight. He was recaptured in September 1972 but made another extraordinary escape, overpowering prison guards and fleeing in a stolen uniform. After a short spell across the Border, he came back to Ballymurphy and to his death. “He was born to be killed,” concluded an associate.[22] Bryson was just twenty-six when he died, Mulvenna twenty-two, and Tolan thirty-one.

Adams’s leadership of the IRA during these early years received much of its shape and direction from his relationship with two influential comrades, both of whom were to leave lasting marks on IRA history. One was Ivor Bell, his adjutant in the Second Battalion, and the other, Brendan Hughes, an early commander of D Coy. Bell had been in the IRA during the 1956–62 campaign but disagreed with the decision to call a cease-fire and had quit. As much an anarchist as a republican, he rejoined in 1970 and became commander of B Coy in the Kashmir Road area. “Ivor and Gerry were a team. We looked to them for political direction, for strategy, and for interpretation,” recalled a colleague.[23] Hughes, from the Grosvenor Road, was an early recruit to the Provisionals and a disciple of the slain Charlie Hughes. Known as the Dark because of his swarthy features, Hughes was happier as a rank-and-file “operator” than as a leader and often teamed up with Bryson for operations. After Cahill’s elevation to the post of Belfast commander, he became Adams’s Battalion operations officer. Together, Adams, Bell, and Hughes were synonymous with the Belfast IRA of the early 1970s.

After internment the trio played an even more crucial role in the Belfast IRA. MacStiofain decided that leaving Joe Cahill in charge of Belfast was a public relations risk too high to take. Cahill had humiliated the British command with his Ballymurphy press conference a week after the failed internment swoop, but he was now a marked man and, as a result of the media coverage, well known too. It would only be a matter of time before he was arrested, and the British would be sure to make much of his capture. MacStiofain ordered Cahill down to Dublin, where he combined his Army Council position with the post of GHQ director of finance. Twomey was promoted to Belfast commander, and Adams became his adjutant, effectively his second in command. Bell became Brigade staff operations officer, while Hughes took over the Second Battalion. The trio had each moved up a step on the IRA’s ladder.

INTERNMENT USHERED IN a new phase in the IRA’s development, especially in Belfast. Its ranks were flush with new and angry members eager to strike back at the British, supplies of cash and weapons increased, particularly from Irish-American communities in the United States that had watched unfolding events with a mixture of astonishment and mounting atavistic fury, and soon IRA violence was at an unprecedented level.

The defining characteristic of the IRA in the weeks and months after internment was the utter spontaneity and unpredictability of its violence. There was virtually no central control from Dublin or even the Belfast Brigade. Aside from special operations that required coordination and planning, IRA companies were encouraged to go their own way. “There was an incredible amount of activity at this time,” remembered one activist. “We would mount five or six operations every day—a bank would be robbed, a bomb downtown, a booby trap for the Brits, snipes, a float.”[24] A float was a particularly hazardous operation in which three or four heavily armed IRA members would drive randomly around their streets in the hope of encountering a British army patrol to fire upon.

At the start the British army was naïve and suffered badly in the unfamiliar warren of tiny streets that made up much of West Belfast. The first armored cars, Saracens, which were nicknamed Pigs by military and IRA alike, came onto the Falls Road with pictures of the Virgin Mary or Christ pinned to their radiators, apparently in the belief that the IRA would then not dare shoot at them. Their knowledge of where they were or whom they were dealing with was fashioned by ignorance and bigotry. Early on, the IRA discovered that they could easily trick soldiers into firing at their own patrols: “Whenever the Brits came in they would come in big ‘duck patrols.’ They would patrol in parallel and we would snipe at one, and before you knew what was happening the Brits would be shooting at each other and we would withdraw and watch them.”[25] Such naïveté did not, however, last long.

Adams kept one or more steps ahead of the British army for months after August 1971, but eventually he made a mistake. In July 1971 he married Colette McArdle, an activist who came from a well-known republican family, whose mother, Maggie, was a republican veteran and a friend of Joe Cahill. Gerry Adams and Colette McArdle had met in 1970 at a point when republicans were mounting pickets outside British army bases where discos were being held and Ballymurphy girls were fraternizing with the troops. Adams had a progressive attitude toward female involvement in military matters, but he barred Colette from involvement in them, although he encouraged other women to join his Second Battalion on the same terms as men, much to the fury of Cumann na mBan leaders, as one female member recalled. “That was in the days when the Cumann na mBan came under the authority of the IRA; in fact even a Volunteer in the IRA could give orders to a ranking Cumann na mBan woman. The Cumann na mBan stopped taking orders from the IRA when he [Adams] allowed women to join it; they wouldn’t take orders from other women; they were very resentful of them and refused to work with them.” Adams was, by contrast, keen to ensure that Colette never got involved in activities that could put her in danger, which her association with him might well have done, as one contemporary remembered. “We were told that he had given a direction [to Cumann na mBan leaders] that Colette was not to go on any operations, not even to carry a weapon or papers, nothing.”[26] The purpose may well have been to make it more difficult for the British to track him and other IRA leaders down, but the order unsettled some colleagues.

The newlyweds had a difficult first few months, always on the run and constantly afraid that at any moment they could be parted by Adams’s arrest or worse. Colette suffered a miscarriage, and the pair were rarely able to spend more than a few hours or at most a day or two in each other’s company before Adams had to move on to a safer “billet.” Regularly changing habits, routes, and lifestyle was the key to survival for IRA leaders in 1971 and 1972. Despite the strain Colette appeared delighted to have married the IRA leader, as an associate recalled: “I remember once going to a call house where we had to meet Adams, and as we were sitting there in the midst of an Army meeting Colette came bursting in and sat down on Adams’s lap, and it all got kissy-kissy, stroking his hair and so on. We had to change the conversation to general things, and someone asked, ‘What would you do if your partner was unfaithful?’ And I remember Colette flourishing her hand and pointing to the wedding ring, saying this is all I care about, as long as I’ve got this, and Adams was giggling.”[27]

In a bid to create some stability in the relationship, Adams arranged to rent a terraced house in Harrogate Street in the Clonard district, where at least Colette, now pregnant, but destined not to go the full-term, could be assured of a more settled life even if her husband could visit her only occasionally. They moved in, but they were “hardly there,” to use Adams’s words, when the house was raided at dawn on March 14, 1972, by British troops and he was taken away to Springfield Road RUC barracks. Adams suspected that an informer had betrayed the house, because the troops knew whom they had come to get, although they were not really sure he was the Gerry Adams on their list.[28] He chose to play on this by claiming he was someone called Joe McGuigan. The soldiers believed him, but when they brought in a veteran RUC Special Branchman, Harry Taylor, who identified him, Adams was transferred to the main military interrogation center at Palace barracks in Holywood, on the eastern outskirts of Belfast. Palace barracks had a terrifying name among IRA members; it was where the special interrogation methods had been used in the first internment swoops and the victims, Adams’s cousin Kevin Hannaway included, suffered longterm psychological and physical damage.

Adams was given a rough time by his interrogators. He was beaten badly and subjected to mental terror; his captors pretended they were about to kill him, and an attempt was made to inject him with what he was told was a truth drug. Years later Adams would be credited with introducing the IRA to systematic anti-interrogation training, which had at its core the principle that if IRA Volunteers stayed silent and avoided creating any relationship with their interrogators, they would survive the experience. But during his own interrogation he chose the dangerous course of conversing with his questioners and persisted with the attempt to pass himself off as Joe McGuigan even though Harry Taylor had destroyed the ploy. In his autobiography he claimed the tactic helped him withstand the experience, but in the end he admitted who he was after the Special Branch told him that if he was interned under the name Joe McGuigan, he would not be able to get visits from Colette. Once the police and army were satisfied they had seized the right man, he was transferred to the Maidstone, where he found his maternal uncles Liam and Alfie Hannaway waiting. Later, when the Maidstone was closed down as a prison ship, he was moved to a World War II prisoner of war–style internment camp at Long Kesh, on the western outskirts of Belfast, where his father and brother Paddy were imprisoned. There was no shortage of Adams family members in jail.

SECURITY SUCCESSES like the arrest of Gerry Adams were slowly restricting the IRA, but the British military was still denied the sort of face-to-face confrontation that its generals were confident could deal a knockout blow to the republicans and give the politicians the time and space to construct a deal. Since the IRA was unlikely to make a gift of such an opportunity to the British, it was left to the British to create one or at least to take full advantage if even half a chance came along. On January 30, 1972, in Derry, an opportunity did present itself, but far from providing the British with an opening to give the IRA a bloody nose, the horrific events of that day were to be a watershed in Irish history, one that would propel the Provisional IRA measurably closer to the goal of forcing the British to withdraw entirely from Ireland.

The event on January 30 was a march organized by NICRA to protest against the continuing use of internment, a march that was bound to attract many thousands of demonstrators and lead to the sort of stone-throwing confrontation that might tempt the IRA to come out into the open. The British chose to send one of their crack outfits, the First Battalion of the Parachute Regiment, to police the march in the apparent expectation that its battle skills would be needed.

Later IRA leaders concluded that the British were hoping that the presence of the First Para would lure the IRA into a gun battle that the Provisionals would surely lose. If that was the plan, it was misconceived; the IRA had no intention of falling into such an obvious trap. Ten thousand people marched to the Bogside, and just after a small section broke off to throw stones, bricks, and iron bars at the troops, the Paras went into action. An hour later thirteen men lay dead and seventeen wounded, one of whom died a few weeks later. None were in the IRA, and eyewitness testimony said they had been killed in cold blood. The deaths on Bloody Sunday outraged nationalist Ireland. A wave of anger swept through the entire country. In Derry the local SDLP leader, John Hume, said that the mood in the Bogside was now for “a united Ireland or nothing,” while at Westminster the civil rights MP Bernadette Devlin physically attacked the British home secretary, Reginald Maudling, in the chamber of the House of Commons. In Dublin a crowd of twenty thousand besieged the British embassy before IRA members arrived to bomb and burn it down. Even ministers in Jack Lynch’s Fianna Fail government in Dublin were moved to unaccustomed militancy by the bloodshed. As he arrived in New York to speak at the United Nations, the Irish foreign affairs minister, Dr. Patrick Hillery, declared, “From now on my aim is to get Britain out of Ireland.”[29] In the Bogside, meanwhile, young people were said to be queuing up by the hundreds to join the IRA.

The most significant outcome of Bloody Sunday is that it sounded the death knell for unionist rule at Stormont. Within weeks the British finally concluded that the cost of sustaining Faulkner in power was too high. On March 24, 1972, after a contrived dispute with Faulkner over control of the security forces, the British prime minister, Edward Heath, announced a yearlong suspension of the Stormont parliament pending an agreed political settlement. In the meantime Britain would take direct responsibility for governing Northern Ireland on a day-to-day basis. A cabinet minister would run the government until the suspension was lifted, but everyone knew that in practice the suspension would be much longer than a year.

The fall of Stormont was a major victory for the Provisionals, but it also marked a watershed in their campaign. Most nationalists were jubilant. Stormont symbolized unionist domination and Catholics’ second-class status. Its collapse was a reason to celebrate and also to suspend the violence. This dramatic change in the psychological climate coincided with a series of badly bungled republican military operations, the combined effect of which was to stimulate demands for a cease-fire.

The botched operations followed one upon the other. On February 22 the Official IRA bombed the Paras’ headquarters in Aldershot, England, in revenge for Bloody Sunday, but instead of killing soldiers, five cleaning women, a British army chaplain, and a gardener were blown to smithereens. On March 4 a bomb exploded without warning in the middle of a busy Saturday afternoon in the Abercorn restaurant in downtown Belfast, killing two women and injuring seventy people, mostly shoppers, some of them terribly. The IRA was blamed for the bombing, and while it strongly denied the charge, the allegation stuck. Sixteen days later the IRA loaded two cars with bombs and parked them in Belfast city center. A number of conflicting phone warnings were given, with the result that police moved fleeing crowds in the direction of one of the bombs in Lower Donegall Street; it exploded and killed seven people, five of them civilians. The Provos had engineered their own Aldershot. Then, on May 21, the Official IRA in Derry kidnapped and killed a local Catholic, William Best, a nineteen-year-old member of the locally recruited Royal Irish Rangers, a regiment of the British army barred from serving in Northern Ireland. Ranger Best was seized while he was on home leave from Germany, and there was a strong local reaction against the killing, which affected the Provisionals as much as the Officials. As small peace groups, many led by women, sprang up in nationalist neighborhoods, the Official IRA leadership declared a cease-fire, and this in turn added to the pressure on the Provisionals.

THE SUSPENSION of Stormont and the imposition of direct rule from Britain represented a major victory for the IRA. In its own account of the period, the IRA described the day when Heath suspended the Stormont parliament as “one of the most momentous… in Irish history.”[30] The IRA had good reason to celebrate. The destruction of the Stormont parliament had been a declared Provisional IRA war aim from the outset, as Sinn Fein’s president, Ruairi O Bradaigh, made clear when he told a July 1971 rally in Derry, “We’re on the high road to freedom, and what we need to do now is to rock Stormont and to keep it rocking until Stormont comes down.”[31] Most Catholics, moderate as well as militant, heartily agreed—but with an important qualification. Many Catholics believed that, having achieved this success, the IRA should then at least review its options. There was also war weariness in many Catholic districts of Belfast. The shootings and bombings had transformed many nationalist areas into terrifying war zones, where people ran a daily risk of running into gun battles or being caught up in nerve-jangling bomb explosions. The collapse of unionist rule brought the hope that this could all be near an end, as one commentator noted: “This major victory having been won, the feeling grew in the Catholic community—fostered by the SDLP, the clergy and the Dublin government—that the Provos ought now call a halt to their campaign. At the very least, thought many nationalists, we deserve a bit of a respite from the past six months of non-stop violence.”[32]

While the moderate nationalist party, the SDLP, welcomed the arrival of the new British minister for Northern Ireland, the avuncular and genial gentleman farmer William Whitelaw, and showed an eagerness to re-enter political dialogue, the IRA greeted the fall of Stormont by stepping up its violence. With its ranks swollen by the anger at Bloody Sunday and other instances of state violence, the IRA carried out twelve hundred operations in May 1972, many of them in rural areas, and more the following month. While the IRA leadership hoped that this would drive the British to the negotiating table, the truth was that the fall of Stormont had opened up a fault line within nationalism that would never really close. Moderate nationalist opinion now sought a political deal and reform, while the IRA fought on for revolution and the elusive republic.

From all the available evidence it seems that the subsequent truce was entered into halfheartedly by both the British and the IRA and in an atmosphere of intense mutual suspicion and distrust. After weeks of maneuvering, the cease-fire began on June 26, 1972, and ended just thirteen days later, on July 9, only two days after Whitelaw and an IRA delegation had met in London. It appears that neither side was all that sorry when it ended. A squalid sectarian dispute over the housing of Catholics in a loyalist-controlled part of West Belfast was allowed by the British to get so serious that it broke the cessation. If the British were secretly glad to see the truce end, so was the IRA in Belfast. Fears that the longer the cease-fire lasted, the more damage it would cause the IRA were very strong. The halt in hostilities had tempted hitherto unknown IRA members to break cover; they were mixing in public with known IRA men, and this was all very visible to the British. Something similar had happened during the 1921 truce. The IRA then had greeted the Treaty negotiations as a victory and celebrated accordingly, emerging in public to receive the adulation of their communities. Collins’s negotiating hand had been badly weakened as a result. Nevertheless the 1972 cease-fire gave something to both sides. The British got an opportunity to take a close look at the key Army Council and military leaders, while the IRA could now say to nationalists that at least it had tried to negotiate terms with the British, and if it had failed it was no fault of theirs.

It was a measure of Adams’s status within the IRA even at that early point that part of the truce preconditions included a demand that he be released from Long Kesh to join the IRA leadership delegation and to assist Daithi O Conaill in making the detailed arrangements for the Whitelaw-IRA summit. According to one account the cease-fire would not have happened had Adams not been freed: “The leadership of the Belfast Brigade at this time was heavily made up of those who commanded the Second Battalion in the run-up to internment. When MacStiofain announced internally that we were going for a truce he was told by that element in the Belfast leadership that there would be no ceasefire in Belfast unless Gerry Adams was released from internment.”[33]

It was men from Adams’s Second Battalion who fired the shots that ended the truce, a fusillade directed at troops stopping homeless Catholics from occupying houses in the Lenadoon housing estate. One of them recalled it vividly: “I remember quite clearly, Jim Bryson, Tommy Tolan, and another figure were told by Twomey to go up to Lenadoon with a Lewis gun— Big Louie—and two Armalites. They were told to wait for a signal from Twomey, who was negotiating with the Brits. When he raised his arm they were to open fire, and the cease-fire would be over. He couldn’t, because of the crowd that was confronting the Brits, but eventually Twomey got them back, and they opened up followed by the Andytown men.”[34]

MacStiofain, ever eager to keep on good terms with the Northerners, had packed his London team with hard-liners from Belfast and Derry. Adams was joined by Seamus Twomey, Ivor Bell, now Belfast adjutant, and Martin McGuinness, the young leader of the Derry IRA who had risen through the ranks since internment. MacStiofain and O Conaill were the sole Southern IRA leaders on the delegation. Conspicuously absent was the Sinn Fein president, Ruairi O Bradaigh.

Afterward only MacStiofain appeared positive about the encounter, a view with which Adams openly disagreed. Adams and the Belfast IRA men suspected that Britain was seeking to draw the IRA into a long cessation, as an associate recalled: “The reasoning of the Belfast leadership… was that the British wanted the truce to continue. One of the delegation saw previously unknown Volunteers sitting outside pubs drinking with known IRA people, and he was deeply concerned. [The IRA’s] guard was dropping, and he was determined to see the truce broken. As far as I recall, the Belfast delegation left the Whitelaw talks with the clear intention of breaking the truce.”[35] Adams sided with the hawks, according to a contemporary. “Adams and Bell were very, very skeptical about the cease-fire, they warned us that things weren’t good, they didn’t trust the Brits and thought they were playing for time. They were fully in favor of breaking it. If Adams had been opposed, it wouldn’t have happened.”[36] The truce duly broke down amid violence that claimed twenty lives over the next three days.

The Belfast IRA had its own reasons for wanting to break the truce. There had been developments on the military front that gave the IRA in the city reason to believe that it could stretch the British further. The IRA, they calculated, was not yet at its strongest and, if things went well, could put more pressure on the British before returning to the negotiating table in a more powerful position. In both cases Adams’s contribution was crucial.

The first was the decision by the Belfast Brigade to import quantities of the American-made Armalite rifle, a powerful semiautomatic weapon that came to symbolize the IRA of the 1970s. The Belfast Brigade first heard of the gun when a Falls Road seaman showed a member of Second Battalion staff a U.S. magazine article about the weapon, known as the AR-15, in the autumn of 1971. Weighing only seven pounds and fitted with a collapsible butt, the Armalite was easy to hide and could even be dumped in water. It fired a high-velocity .223 round, which tumbled through the air with the same deadly effect as a dum-dum bullet, and it was highly accurate. When brigade staff heard about the weapon, the decision was instantaneous: the Armalite could make the IRA better armed than the British army. The conclusion was simple, as one source familiar with the episode recalled: “If we could lay our hands on these guns [maybe] we’d win the war.”[37]

Adams ordered his operations officer, Brendan Hughes, to travel to New York to arrange for the purchase and shipment of the guns. The Americans involved in gunrunning for the IRA had old-fashioned ideas about what the best weapons were, and GHQ in Dublin agreed with them. They were the weapons of their age. GHQ instructions were to acquire standard World War II U.S. infantry weapons, principally MI Carbines, Garand rifles, and the ubiquitous Thompson submachine gun. When Hughes arrived in New York looking for a completely different sort of weapon, he met enormous opposition from the locals. It strengthened a growing view at the top of the Belfast IRA that many in the Dublin leadership, especially in GHQ, were out of touch with the needs of those fighting the war in the North. Adams’s orders, however, were that the Armalites had to be acquired even if that meant bypassing GHQ. By the spring of 1972 the Belfast Brigade had organized its own supply route, and the Armalite began to appear in the city, where it was tested in combat by D Coy in gun battles with the British. The weapon was judged a huge success, and arrangements were made to import larger quantities. In May and June 1972, when preparations for the truce were in progress, two hundred AR-15s arrived in Belfast, smuggled aboard the transatlantic liner QE2 and then transported to Belfast from Southampton in England.[38] There were enough Armalites to equip every active-service unit in the city. As Adams and Bell flew out on board the RAF plane to meet Whitelaw, they knew that back in Belfast the IRA was better able than ever to take on British troops.

The second military advance was the development of the car bomb, another weapon synonymous with the IRA of that era. The car bomb was discovered entirely by accident, but its deployment by the Belfast IRA was not. The chain of events began in late December 1971 when the IRA’s quartermaster general, Jack McCabe, was fatally injured in an explosion caused when an experimental fertilizer-based homemade mix known as the “black stuff” exploded as he was blending it with a shovel in his garage on the northern outskirts of Dublin. GHQ warned that the mix was too dangerous to handle, but Belfast had already received a consignment, and someone had the idea of disposing of it by dumping it in a car with a fuse and a timer and leaving it somewhere in downtown Belfast. “It was a bomb in a car rather than a ‘car bomb,’” recalled a Belfast IRA source familiar with the episode. “A young Volunteer took it in [to the center of Belfast] and we could feel the rattle where we stood. Then we knew we were onto something, and it took off from there.”[39] The car bomb enabled the IRA to increase significantly the amount of explosives it could deliver in each individual operation while exposing fewer operatives to arrest or premature death.

The “black stuff” mix was perfected and made safer to handle within weeks of McCabe’s death, and this meant that the IRA now had unlimited supplies of homemade explosives. It no longer had to depend on gelignite, supplies of which were gradually being cut off by the British and Irish intelligence authorities. The car bomb was, however, a double-edged sword. The sheer size of the devices greatly increased the risk of civilian deaths in careless or bungled operations. The IRA bombs that killed seven people in Little Donegall Street, for example, were among the very first car bombs deployed, and no one could doubt that the episode was a public relations disaster for the Provisionals.

The new explosives mix could be used outside the city as well. It was an ideal explosive to use in rural land mines, which were usually hidden in culverts, or drainpipes, which ran under country roads at regular intervals to draw off rainwater. The land mine came into its own during the early summer of 1972, making military and police patrols in Tyrone and Fermanagh hazardous and eventually rendering South Armagh a no-go area for motorized British units. The truth was that among Armalites, car bombs, and land mines, the IRA felt it was on a string of successes when the 1972 cease-fire was called. Seen in that perspective, the chances that IRA leaders would enter talks with Whitelaw in a mood of compromise were virtually nonexistent.

WITHIN TWO WEEKS all had changed utterly. The IRA had forced the destruction of the Stormont parliament and fought the British to the negotiating table. Not since 1920 had the organization wielded such power or been so well placed to influence and shape events. But in a trice all that was to disappear. On Friday afternoon, July 21, the Belfast Brigade sent twenty of the new car bombs into the city and detonated them in just over an hour, killing 9 people and injuring 130, in one of the worst days of violence yet seen during the Troubles. At the height of the bombing the center of Belfast resembled a city under artillery fire; clouds of suffocating smoke enveloped buildings as one explosion followed another, almost drowning out the hysterical screams of panicked shoppers. Six people, two of them soldiers, were killed by a car bomb at a bus station, and three civilians, two women and a fourteen-year-old schoolboy, died when another device was detonated in North Belfast. Television pictures of firemen shoveling human remains into plastic bags compounded the horror.

The deaths were not deliberate but the result of careless planning by the Belfast Brigade leadership. Brigade commanders made the fatal error of assuming that the British army and RUC could deal with so many bomb alarms all happening more or less at the same time in different parts of the city. Bloody Friday, as the day was called, was the unionist equivalent of Bloody Sunday, and it was an unmitigated disaster for the IRA. The IRA later tried to blame the British for deliberately ignoring telephoned warnings, but that cut little ice with anyone except its most loyal supporters. Most people accused the IRA of using sheer terror tactics. Bloody Friday had a speedy political impact. Moderate nationalists put even more distance between themselves and the IRA and intensified efforts to seek negotiations with the British and Faulkner’s unionists.

It took years for the IRA to admit that Bloody Friday was its fault. “We put it down to the Brits allowing bombs to go off, but the real reason was it was too much for the Brits to cope with, the bombs went off too close together, the town was too small, people were being shepherded from one bomb to another,” conceded an IRA activist of the time.[40] Bloody Friday had been planned by Twomey and his staff before the June 26 truce and was intended to pressure the British to come to the negotiating table.[41] Adams was in jail at the time and played no role in its conception, but after the collapse of the truce he was back at the center of events, acting with Bell once again as adviser to Twomey.[42] The plan was revived to demonstrate that the IRA was still in business, but it backfired badly; Adams did not initiate Bloody Friday, but he was involved in its organization. The British quickly realized that events had moved to their advantage. Ten days later, on July 31, in an operation code-named Motorman, hundreds of British troops invaded the no-go areas of Derry and Belfast, an action that would have been unthinkable before Bloody Friday. Within weeks military forts were constructed right on the IRA’s doorsteps, and the organization’s freedom of movement was severely curtailed. Now able to put the IRA under close surveillance in both cities and to screen thousands of civilians for IRA sympathies, British army intelligence on the IRA improved markedly. Within two years the British grip on areas like West Belfast was so tight that the Belfast Brigade was forced to move its operational headquarters to the southern outskirts of the city, to the affluent Malone area.

Bloody Friday was the one great black mark against Adams’s strategic record in the early 1970s. The man himself was incandescent with rage after the botched bombings, according to IRA sources. “Most of the bombs that day came from the Third Batt area, from Ardoyne and the Markets, and some from the Second Batt; it was a BB [Belfast Brigade] operation though. Afterwards the word was that: ‘the Big Lad’s doing his nut, about the warnings not being phoned in or being bungled.’ There was an undercurrent of blame being put on the Third Batt leadership.”[43] In his autobiography Adams played down the implications of Motorman, saying that it merely required the IRA to readjust its tactics. But the reality was that Bloody Friday and its aftermath marked a watershed in the IRA’s fortunes. From then on the organization would be on the defensive both politically and militarily.

The first IRA casualty of Bloody Friday was the Belfast commander Seamus Twomey. Adams and Bell were convinced that Twomey was “too fiery” for the new circumstances, and they combined to persuade him to quit Belfast and to move to Dublin, where he would be better placed to look after the Northern IRA’s interests. In September, Twomey acquiesced. Adams replaced him and appointed Bell as his adjutant, and Hughes became his operations officer. In the view of some IRA members Adams was merely formalizing what already existed: “Twomey was commander in name only; Adams [always] really called the shots.”[44]

Soon Adams would have an ally at the very top of the organization. In November 1972 Sean MacStiofain was arrested as part of the Dublin government’s accelerating crackdown on the IRA leadership. Sentenced to six months in jail on IRA membership charges brought on the basis of a radio interview he gave to the Irish broadcasting service, RTE, MacStiofain immediately embarked on that most traditional of IRA protests, a hunger strike, which ended fifty-seven days later, inconclusively and amid charges that the Provos’ chief of staff had cheated during the fast. Upon his arrest MacStiofain immediately lost his IRA rank, and thanks to the opprobrious circumstances of his fast’s conclusion, his republican career effectively ended. A tough and uncompromising leader, MacStiofain had overseen the birth and development of a formidable fighting machine, and for that Northern IRA leaders retained for him a degree of affection and respect that survived his ignominious departure. Joe Cahill took over as the Provisionals’ second chief of staff. Second Battalion veterans now occupied the two most important posts in the IRA, its national leadership and the command of Belfast, the cockpit of the IRA’s war against Britain.

GERRY ADAMS was to be Belfast commander for the next ten months, during which time his already established reputation as the IRA’s key strategic thinker was significantly enhanced. But he also earned a name for ruthlessness that would make many a potential rival pause for thought before considering any challenge to his authority.

One event marked out Adams’s period in command as special in the eyes of other IRA members, and that was a strike against British military intelligence that was reminiscent of the triumphs organized by IRA leaders during the Tan War. It looked like an astonishing counterintelligence coup, but in reality what happened was more a chance affair, which owed much to the alertness of an observant junior IRA member. But the operation that followed persuaded many that the new Belfast commander had penetrated the core of British intelligence.

The story of what became known as the affair of the Four Square Laundry began with an admission by a rank-and-file volunteer in the Second Battalion’s D Coy that he had been working as an informer for the military. The volunteer, Seamus Wright, from the Lower Falls Road area had come under suspicion because he was so often absent from Belfast, apparently spending much of his time in England. Under interrogation by Second Battalion staff, Wright admitted that all the time he had actually been in the company of a special military unit based at Palace barracks in Holywood, County Down, where IRA suspects were taken for routine interrogation before being interned. The unit was known by the initials MRF, which the IRA believed stood for Military Reconnaissance Force, a group subsequently alleged to have been involved in two drive-by shootings in the summer of 1972 that were blamed at the time on loyalist gangs. Wright admitted he had agreed to work for the MRF.

The MRF ran plainclothes military patrols in republican areas of the city, but it also had built up an agent-running capacity and had set aside a special section of Palace barracks to house informers where debriefings were conducted and operations planned. A favorite tactic was to drive these agents through nationalist districts in military vehicles to identify and photograph other IRA activists through the slits in armor plating. Wright named another D Coy volunteer as a fellow MRF agent. The IRA arrested Kevin McKee, and his questioning added significantly to the IRA’s knowledge of MRF operations.

McKee revealed that the MRF had constructed an intricate undercover intelligence network that included a massage parlor, ostensibly run by English prostitutes, an ice cream business, and the Four Square Laundry, which operated in West Belfast. The Four Square operation was simplicity itself: a van would tour housing estates offering cut-price laundry sevices so as to acquire clothing to be analyzed for traces of explosives and gunpowder and so to identify IRA houses. Within the roof space of the large van used to collect and return laundry lay two British operatives who would photograph suspects on the streets. By the standards of later British intelligence operations against the IRA, it was an amateurish operation but also an indication of how little the British knew about the IRA in those early days.

After McKee’s interrogation, the Belfast Brigade—Adams, in other words—took over the operation from Second Batt, and plans were made to ambush the various MRF teams. During the midmorning of October 2, 1972, gunmen ambushed the Four Square Laundry van as it made its rounds in Twinbrook, a sprawling housing estate on the edge of West Belfast. The driver was shot dead and the roof compartment sprayed with automatic fire. At the same time in North Belfast gunmen from the Third Battalion shot up the massage parlor. The third premises identified as an MRF front, city center offices, turned out to be vacant when the IRA arrived. At the end of the day the IRA claimed to have killed five undercover British soldiers, but the British would admit to only one dead, the van driver killed in Twinbrook. Despite this uncertainty the IRA regarded the day’s work as a major victory over British intelligence, and Adams compared the operation—and implicitly himself—to the counterintelligence exploits of Michael Collins, whose famous “squad” had wiped out the bulk of Britain’s secret agents in Dublin in one violent day, the first Bloody Sunday, fifty-two years earlier: “It was a devastating blow, on a par with Michael Collins’ actions against British Intelligence in November 1920…,” Adams later wrote.[45]

A devastating blow it certainly was, but the incident was no less serious for the IRA. Although badly hit by the Belfast Brigade, the episode had demonstrated that the MRF had clearly managed to infiltrate the IRA’s crack Second Battalion, and that this had only been discovered by chance. It was also apparent that, thanks to Wright and McKee, British intelligence now knew the names of all D Coy’s members and many of the Second Battalion’s secrets. The chances of there being other Second Battalion agents on the MRF’s payroll, recruited as a result of intelligence passed on by the pair, had to have been high. More than anything else the penetration demonstrated that the IRA had no systematic counterintelligence capability. After all, Wright and McKee had been caught by luck, not by any IRA system. While the task of pursuing traitors was part of the Coy intelligence officer’s job, it was only one part of a substantial job description—more time and energy went into identifying targets than into uncovering agents. The Four Square Laundry affair exposed major deficiencies in the way the IRA conducted its business and it raised embarrassing questions. Not least it exposed the damaging consequences that would result if the intelligence officer himself or herself had been turned by the British. Without a foolproof counterintelligence capacity, the IRA had no protection against that level of penetration. The Four Square Laundry operation was loudly praised, and Adams basked in the adulation, but the hard questions were just not asked. Instead, the accidental discovery of the MRF’s penetration and the weaknesses in IRA command and control which it had exposed were literally covered up.

The IRA sentenced the MRF agents Wright and McKee to death after courts-martial, but in an unprecedented twist the Belfast Brigade ordered that they be buried in secret after execution and their treachery kept hidden.[46] This was contrary to one of the central principles of the IRA’s informer-hunting practice, which was that those found guilty of working for the British must be exposed publicly in order to discourage others from imitating them. In practice that meant that the bodies of dead informers must be left literally by the roadside and a public explanation given for the killing. The thinking behind this was that anyone tempted to follow suit would be persuaded to think again. That in this case, contrary to IRA rules, Wright and McKee were put in secret graves suggests another motive.

The justification for “disappearing” the men that was put forward at the highest levels of the Belfast Brigade—at the level of Adams, Bell, and Hughes—was disarming. Since Wright and McKee were both members of influential republican families, it was agreed, the IRA would be sparing their relatives considerable embarrassment if they were just quietly buried and news of their fate kept hidden.[47] Wright was related to the Hickey family, and a sister-in-law, Eileen Hickey, went on to become a senior figure in the Belfast Brigade and the commander of IRA prisoners at Armagh women’s prison. McKee was a nephew of Billy McKee, Adams’s old adversary. Secret executions and burial would spare their familes embarrassment, although anguish at their unexplained disappearance would surely outweigh that. But the secret manner of their deaths served another purpose; the extent of British intelligence penetration of Adams’s IRA units, particularly in his own Second Batt and D Coy, went to the grave with them and the luster of the Four Square Laundry operation left untarnished. The pair were ferried secretly down to South Armagh, where they were held for six weeks before the orders to kill them arrived. By that time their jailers had built up such a strong rapport with them that the South Armagh IRA requested that others do the deed. IRA men from Belfast, in fact from Adams’s old IRA unit in Ballymurphy, were sent down to carry out the killings.[48] The execution and burial of the two agents was kept a secret until March 1999 when the IRA, under pressure from relatives and the demands of the peace process, finally admitted part of the story of what had happened. Despite intensive efforts their graves have never been located. Seamus Wright and Kevin McKee carried to their secret graves the untold story of how British intelligence bested Gerry Adams. Their fate ensured that the story would never see the light of day.

THE CLANDESTINE BURIALS of Wright and McKee set a precedent that was repeated at least seven more times in subsequent years. The Belfast Brigade, under Adams’s leadership, had set a grisly precedent. From there on, if the IRA had an informer on its hands whose existence would embarrass the organization or its supporters, it would think little of disposing of the problem under the clay that covered a secret grave. The third secret interment during Adams’s reign as Belfast commander fell into that category, although the circumstances of the disappearance of Jean McConville were to leave a much more lasting and damaging stain.

The death and disappearance of Jean McConville was made possible by another important military innovation pioneered by Gerry Adams. With an eye on the special “squad” that Michael Collins had constructed in 1920 to subvert British intelligence operations in Dublin, Adams did the same thing in the Belfast of the early 1970s when he got Twomey’s go-ahead to set up two secret cells in the city to carry out special operations on behalf of the Belfast Brigade. These cells reported directly to Adams and received their instructions only from him, a chain of command that was formalized when Adams replaced Twomey as Belfast commander. They became known in IRA folklore as “the unknowns.” One unit was located in West Belfast and led by the brigade staff intelligence officer, a figure from the Turf Lodge area. The other was in the Third Battalion area in North Belfast and commanded by a renowned IRA leader, who was later shot dead at his home by loyalists. The cells were very small at the start, consisting of only three members each, later expanded to four. The idea was that “the unknowns” would be self-sustaining and independent of the rest of the Belfast IRA; they carried out their own intelligence work and acted on it themselves, although from time to time they borrowed personnel from other units for larger operations, especially D Coy in the Lower Falls.

Jean McConville was a thirty-seven-year-old East Belfast Protestant who had married a West Belfast Catholic and converted to Catholicism after their wedding. In 1972 she had been living for two years in the Divis Flats complex in the Lower Falls area with her ten children. Her husband, Arthur, a former soldier in the British army who had quit in 1964 to become a builder, had died a year before. Now widowed and poverty-stricken, Jean McConville was struggling to raise her family. In December 1972, suddenly and without any reason, she disappeared from the face of the earth. One widely accepted and repeatedly given explanation for her disappearance was that she had angered the IRA by comforting a seriously wounded British soldier who had been shot by a sniper outside her front door. In revenge, the story continues, the IRA abducted her and took her to a house in the Beechmount district of West Belfast, where she died during interrogation, allegedly suffocating when her questioners placed a plastic bag over her head in a bid to make her talk. Rather than admit what they had done, the IRA then decided to bury her quietly and afterward spread stories that she had deserted her children and run away to England with a British soldier.

The truth was much more complicated, as it tends in such stories to be, although there is no doubt the IRA lied about what happened to Jean McConville—and continued to lie for many years thereafter. The real story behind McConville’s death, the sad and squalid truth of her killing pieced together from sources in the IRA active at this time, is that she died because she was a small and not very important cog in the British army’s intelligence-gathering machine who had the misfortune to cross paths with two ruthless men. One was the British officer who ran her as an agent; the other was the senior IRA figure in Belfast who decided that her secret death would suit his purposes.

THE BACKGROUND to the tragedy of Jean McConville was set in Divis Flats, then a large sprawling complex of apartments and tower blocks whose stairways and corridors provided a perfect stage for snipers and bomb throwers. In the early 1970s the IRA operated virtually openly in Divis. “Everyone knew who the IRA in Divis Flats were; they walked around with guns and so on,” remembered one of their number.[49] For the British army it soon became a priority to place a reliable spotter in the flats who could warn them of IRA activity and planned ambushes. Jean McConville agreed to be one of those spotters, but by all accounts she was not very good at her job and showed a too obvious interest in the IRA’s affairs. It was not long before the local unit tired of her unending questions and began to suspect her. Her apartment was raided, and sure enough the IRA found a radio transmitter that she had been using to communicate with the British army. “It was taken off her, and she was warned never to do that again; she was a woman and the mother of a large family, and so we let her off,” explained one IRA member familiar with the events.[50] But it was just that, a warning. Next time, she was told, there would be no warning.

Inexplicably McConville went back to spying on the IRA, this time with fateful consequences. Although by this stage the British army must have been aware that the IRA knew all about her activities and that she was now in terrible danger, her handlers carried on regardless and supplied her with a second transmitter. Her spying recommenced, and it did not take long before the IRA worked out that she was back in business, once more betraying IRA volunteers and operations. The Belfast Brigade decided that this time she had to die, but its senior members disagreed violently about what to do with her body. The question bitterly divided the Belfast Brigade staff. Some argued that her body should be dumped in the street so that her fate would act as a deterrent to other would-be informers in accordance with IRA custom and practice. But one figure disagreed, arguing that the publicity attached to her death, the fact that she was a widowed mother of ten, would work strongly against the IRA, and he urged that she be buried in secret and effectively disappeared.[51]

The job of “disappearing” Jean McConville was given to “the unknowns” in a move that guaranteed that the story of what had really happened to her would be confined to the smallest number of IRA activists. According to one well-informed source, the order to “disappear” McConville was given to the Turf Lodge-based commander of one of the “unknown” units by a senior member of the Belfast Brigade. Whether, as alleged by one well-informed source, or not the order was given by Adams himself, it is inconceivable that such an order would have been issued without his knowledge. Her court-martial had been held—although McConville was not present to defend herself—she had been found guilty and sentence pronounced. The task of the “unknown” was to fetch her and carry out the sentence.[52] McConville was taken down to a beach near Carlingford just across the County Down–County Louth Border, where she was shot in the back of the head and her body buried in the sand.

Jean McConville’s death and disappearance came back to haunt Sinn Fein during the height of the peace process in the mid-1990s. A campaign to discover her fate was launched by her children, all of whom had been dispersed to foster homes after her abduction, and their efforts won the support of President Bill Clinton and the Irish government, two of Adams’s strongest allies in his new foray into constitutional politics. Adams met the chief campaigners, McConville’s daughter Helen McKendry and her husband, Seamus, but initially denied all knowledge of events. Only after Clinton’s intervention did he and the IRA admit that she had been “disappeared” by the organization. Even then strenuous efforts were made to distance Adams from the affair. Sinn Fein spin doctors suggested that he had been in Dublin at the time of the killing, implying that he had played no part in the decision to kill and secretly bury Jean McConville. Adams himself, according to her son-in-law Seamus McKendry, tried to claim that he could not have been involved, since he had been interned at the time: “He told Helen and I [sic]: ‘Thank God I was in prison when she disappeared.’ ”[53] In fact Adams was very much at large at the time of Jean McConville’s disappearance and must have known all about the circumstances at the time. He was not arrested and imprisoned until July 1973, more than six months after her abduction and execution.

Public disquiet over the IRA’s treatment of the “disappeared,” especially Jean McConville, forced the Irish and British governments to set up a cross-Border commission to coordinate the search for missing remains. The IRA’s Army Council appointed its director of intelligence, Bobby Storey, a close ally of Gerry Adams, to revisit each case and to question the IRA members involved in an effort to discover where bodies had been hidden, while the police on both sides of the Border dug up bogs, beaches and basements. In the case of Jean McConville, extensive searches and excavation of beaches in the Carlingford area by Irish police in 1999 and the following year failed to find anything. But in August 2003 her remains were discovered, apparently by chance, by members of the public walking on a nearby beach that had not been searched. A postmortem examination revealed she had been shot once in the back of the head. While many in Sinn Fein were relieved to see this grisly and embarrassing case apparently closed, for others the killing and disappearance of Jean McConville had become an enduring metaphor for the Adams stewardship of the Belfast IRA. It also seems likely that McConville’s “disappearance” will haunt the Provos and Adams for some time. In July 2006, Police Ombudsman Nuala O’Loan said that an investigation by her office, launched at the request of the McConville family, had failed to find evidence Jean McConville had been an informer, a claim that forced the IRA to publicly repeat its assertion that the dead woman had been executed for spying on behalf of the British army. With the McConville case thus re-opened and an unwelcome spotlight directed at himself, Gerry Adams felt obliged to express concern at the IRA’s past behavior, although managing to avoid mentioning the word “disappeared” or his own possible knowledge of events: “Whatever about the circumstances surrounding Jean McConville’s killing, the burial of her remains was a great injustice to the family.” He also urged the Irish government to “act speedily” so that more excavations to find others who had been “disappeared” by the IRA could take place.

THE IRA’S FORTUNES had waxed and waned in the two years since the Ballymurphy riots ended in December 1970. By the end of 1972 the political and military balance was swinging in the British favor and the IRA was very much on the defensive. Not least of the factors now working against the IRA was that the British military’s intelligence was improving all the time, thanks to the recruitment of a growing number of informers. Since the IRA was increasingly restricted on its home patch, the idea grew among Belfast activists that it should try to break out of the straitjacket, to take the war directly to the enemy. The idea came not from Adams himself but from others in Belfast Brigade, although in IRA mythology it is credited as his and recorded as yet another example of his military and strategic skill.

“Towards the end of 1972 we started working on the plans,” recalled a well-placed IRA source.

The first priority was to recruit unknown Volunteers with no records. It was Adams who went to the three battalions to get them; he told them that the operation was a very big one, that it could be a hanging offense, as it was treason. There were rooms full of Volunteers, and when he said that and that anyone who didn’t want to go should leave, he was nearly knocked down in the rush. The result was that the team ended up with red lights, people like Gerry Kelly who was on the run for murder and others who had been interned.[54]

Elaborate planning went into the bombs that would be placed in London. Although Chief of Staff Sean MacStiofain was told about the plan and approved it, GHQ’s involvement from Dublin was kept to a minimum, not least because the clannish Belfast IRA did not fully trust their Southern colleagues. The date of March 8 was chosen deliberately, for it was the day on which a British government-organized poll to decide whether the Border should be retained was to be held. With the result a foregone conclusion, thanks to Northern Ireland’s built-in Protestant majority, it was the IRA’s way of showing contempt for the idea that unionist consent should ever be a precondition of Irish unity.

The choice of bombing team was to prove controversial. Among the six IRA members who went to London were two young sisters, nineteen-year-old Marion Price and her twenty-two-year-old sister, Dolours, from a staunchly republican family in West Belfast. They had been among the first to volunteer for the mission. The Price sisters had been brought up in a family atmosphere in which sacrifice to the republican ideal had been sanctified. An aunt had been cruelly maimed in the cause, losing her sight and both hands in a bombing that had gone wrong in the 1956–62 campaign. MacStiofain objected, saying that they were too young and that both couldn’t go, because they were sisters. He relented only when one of them became visibly upset.[55] The operations organizers in Belfast, by contrast, had expressed no such qualms, and they were sent on the mission.

The bombings duly went ahead. Car bombs were driven over on the Dublin–Liverpool ferry and taken down to London, where at 9:00 A.M. they were parked beside their targets. March 8, 1973, was not an abnormal day by the standards of the time in Northern Ireland. That morning a twenty-one-year-old British soldier was shot dead by an IRA sniper as he guarded a polling station in West Belfast; a thirty-one-year-old soldier shot earlier in the week in South Armagh died in hospital of his wounds; and the body of a forty-five-year-old married Catholic man was found in a Protestant district of North Belfast, shot in the head, apparently by loyalist gunmen. Six bombs exploded in Belfast that day and five in Derry, but it was the blasts in London that captured the world headlines. One car bomb detonated outside the Old Bailey courthouse and another exploded in Whitehall, at the epicenter of the British government. One man died of a heart attack and 180 people were injured. The IRA had bombed targets in England before, during the Forties Campaign, but never on this scale in London. Now, as a result of the efforts of the Belfast Brigade, the IRA’s war had come to Britain, and the extent of the subsequent media coverage taught the IRA a lesson its members would never forget: one bomb in London was worth a dozen in Belfast.

The bombing team had, however, made a simple error that caused its downfall. The group had fitted false British license plates to the cars ferrying the bombs before they arrived in Liverpool but had made them up randomly. What they had failed to realize was that British plates were year coded, and the age of the cars did not match the code on the plates. An alert police patrol noticed that a car parked outside Scotland Yard, the headquarters of the London metropolitan police, had the wrong plates and raised the alarm. The second car bomb, parked in the West End, was discovered in the same way, and both devices were defused several hours before they were due to go off. The police headed straight for Heathrow Airport, where they found the bombing team queuing for the flight to Belfast. The plan had been that the bombers would all be safely back in West Belfast by the time the car bombs detonated, but it had all gone badly wrong.

It looked as if a simple oversight at the planning stage had landed the London bombers in jail, but there is evidence to suggest that the operation had been compromised somewhere in Ireland. The Belfast Brigade’s original plan had been to take ten bombs over to London, but that was scaled down to six, one for each member of the team. Two days before the bombs were set, a decision was made to reduce the number to four, after British customs officers had taken a close interest in one of the car bombs as it was driven off the Dublin–Liverpool ferry. A message was sent to the Belfast Brigade to leave the fifth and sixth car bombs in Ireland, just in case the customs interest was more sinister.[56]

It later became clear that the British knew all about this change of plan. At the bombing team’s trial, depositions from the prosecution side revealed that British police had sent out a bomb alert warning of four devices—not the six or ten in the IRA’s original plans—and had sealed ports and airports at 6:00 A.M. on the day of the bombings, long before the bombs had even been put in place. The intelligence was very specific and suggested the existence of an agent somewhere at the top of the organization. In addition, the police had distributed photographs of the Price sisters with instructions to stop them—it was clear that the police in Britain knew not only that a bombing was planned but who was involved. The evidence appeared to point to a serious leak somewhere in the Belfast Brigade, but attempts to hold an internal investigation into the affair were resisted.

THE NET HAD CLOSED over the London bombers, and in a wider sense it was also closing in on the Provisionals. On the political front the fracture within nationalism brought about by the fall of Stormont deepened. Whitelaw cajoled the SDLP into talks and won the party’s support with a promise that any deal negotiated would need to embrace “the Irish dimension” and that other nationalist concerns would have to be addressed. In March 1973 Whitelaw published a white paper that recommended a power-sharing form of government that would give the SDLP a guaranteed say in running Northern Ireland. There would also be a Council of Ireland to give meaning to the Irish dimension.

The principle of consent—the doctrine that Northern Ireland would stay British as long as a majority wanted—was reaffirmed, but there were other, balancing concessions for Catholics. Oaths of allegiance to the British crown, which were obligatory for civil servants, teachers, and local councillors and which Catholics considered discriminatory, were abolished. Most nationalists found Whitelaw’s proposals to their liking, and gradually the middle classes dropped their post-internment boycott of the state and returned to public life. There were other positive developments to encourage the Catholic middle class. In mid-1973 the first loyalist paramilitaries were interned, thus answering a long-standing nationalist complaint that the implementation of internment had been biased and one-sided.

The Provisionals meantime sought refuge in ideological purity and as a consequence became more isolated. In January 1973 Sinn Fein announced that it would boycott local council elections planned for later in the year. These went ahead, and the SDLP managed to establish itself as the North’s largest nationalist party. In June an SDLP councillor became mayor of Derry, once the capital city of unionist discrimination. When elections were held to Whitelaw’s power-sharing Assembly, the IRA urged nationalists either to boycott the poll or to spoil their votes. Another Stormont parliament would hinder the achievement of “a just and lasting peace,” it said.[57] Most nationalists ignored the IRA and voted. The SDLP got a mandate, 22 percent of the votes and one-fifth of the seats. When the results came in, the SDLP leader in Derry, John Hume, declared, “The IRA have now heard the voice of the people and it is time they listened.”[58]

Operation Motorman had meanwhile tightened the British army’s grip on the previously unchallenged no-go areas of republican Belfast and curtailed the IRA’s freedom of movement. The numbers of IRA suspects arrested and either interned or charged with criminal offenses increased steadily, according to official claims: one hundred by November 1972, a thousand by the following April. In June 1973 the Northern Ireland Office claimed that 500 IRA members had been convicted and sentenced since Motorman, eight of whom had been given life sentences, the rest an average of four years in jail apiece. Slowly the number interned fell—it stood at 450 in early 1973 but at 330 in June—as the British relied increasingly on the courts to put their adversaries out of action.

By this time Adams had found a safe billet in the University area in neutral South Belfast where he lived with Colette, by now expecting their first son, Gearoid. Like some sort of revolutionary commuter, Adams traveled daily from the safety of his middle-class hideout into the war zone of West Belfast to direct IRA operations. The fact that the IRA’s senior figure in Belfast was now unable to live among his own people was eloquent testimony to the extent to which events had put the IRA on the defensive.

On July 18, another signal that the IRA’s fortunes were slipping came when the Northern Ireland Constitution Bill, a product of Whitelaw’s political negotiations, became law and enshrined once again in section 1, part 1, the principle of consent, that Northern Ireland would remain part of Northern Ireland unless and until a majority of people voted otherwise in a poll. The passage of the bill symbolized the gravity of the political reverses suffered by the IRA since the suspension of the Stormont parliament in March 1972. The British move to strip unionism of power had succeeded in dividing nationalism and diluting opposition to the state, while the IRA’s stubborn support for violence alienated more and more Catholics.

The largest section of nationalism, represented by the SDLP, which was supported by the Catholic middle classes and church hierarchy, welcomed British direct rule and quickly entered into talks with the British and eventually the unionists about a deal that at its core would, inevitably, recognize the constitutional integrity of Northern Ireland. It was a decisive break with the Provisionals. The Catholic middle-class boycott of the institutions of Northern Ireland, which had begun in sympathy with and protest against the internment of IRA men in August 1971, slowly evaporated, and soon constitutional nationalist politicians and Catholic clerics would be condemning IRA violence with the gusto and vehemence they had once reserved for British excesses.

PART TWO
Taking On the Old Guard


Gerry Adams speaks to jubilant supporters after his election as Westminster MP for West Belfast in June 1983. (Derek Speirs/Report, Dublin)

4. Cage 11

At around the same time on July 18, 1973, that Queen Elizabeth signed Whitelaw’s Constitution Bill into law, a British army patrol was briefed to carry out one of the most important arrest operations of the Troubles so far: the capture of most of the Belfast Brigade’s key staff. Early the following morning Adams, his operations officer, Brendan Hughes, and another staff officer, Tom Cahill, arrived at the appointed call house in the Iveagh district of the Falls Road, and this was the signal for the soldiers to swoop. Adams, Cahill, and Hughes were dragged off to Springfield Road RUC barracks, where another senior brigade officer, Owen Coogan, joined them not long afterward.

Adams was stripped and for several hours was badly beaten by his interrogators. Eventually the beatings ended, and Adams and his comrades were trussed up and then photographed, almost as if they were prize trophies, before being flown by helicopter to Long Kesh internment camp. When Adams was welcomed once again by IRA comrades into the huts and cages of Long Kesh, it was two days short of the first anniversary of Bloody Friday.

The IRA was never quite sure how the British army knew about the brigade staff meeting in Iveagh, but it was clear that a well-placed informer had been at work. The loss of the Brigade staff made this a disastrous day for the IRA, but worse was to come. Only hours after the capture of Adams, Hughes, Cahill, and Coogan, the entire Third Battalion staff was arrested at a house in Ardoyne in North Belfast. That day the Belfast IRA saw at least sixteen of its most skilled and experienced leaders incarcerated by the British.

It was not until two years later that the identity of the informer responsible for these losses was established with any certainty, and by then the damage he had done elsewhere to the Provisional IRA’s personnel, resources, and structures was irreparable. The informer’s activities were to have major long-term significance for the IRA, but they were also eloquent testimony to the IRA’s failure in Belfast to sustain the brief counterintelligence successes enjoyed under Adams’s command. The Four Square Laundry operation of late 1972 had been a flash in the pan.

The name of the informer was eventually pieced together in the Long Kesh prison camp, where IRA inmates worked out that their brigade quartermaster in Belfast, Eamon Molloy, was the one figure who had featured in each of their sad stories. A message was smuggled out to trusted colleagues suggesting that he should be closely questioned, and Molloy was tricked into admitting his secret role. Under interrogation he confessed that he had been working for the British since early 1972 and had betrayed dozens of IRA members and revealed the whereabouts of enormous amounts of arms and equipment. After a court-martial in the summer of 1975, Molloy was killed, felled by a bullet to the back of the head, and, like Seamus Wright, Kevin McKee, and Jean McConville, his body was buried in a secret grave, by now the Belfast IRA’s established way of dealing with its embarrassing secrets.

Molloy’s remains were recovered by the IRA, placed in a coffin and left in a graveyard near the Irish Border in May 1999, a year after the Good Friday Agreement was signed. The scandal of the “disappeared,” which surfaced as the peace process gathered pace, had forced the IRA to end years of lying and admit that it had abducted, killed and hidden nine people, mostly in the early and mid-1970s, and all but one from Belfast. To that list have been added five other names who were “disappeared” by the IRA in later years. Some sources have hinted that there is yet another category of such victims, people who could be said to have been “double-disappeared,” i.e. their secret executions and burials have never been acknowledged by the IRA. There may be two or even three such victims, the first of whom was executed and “disappeared” after his alleged misuse of IRA procedure, weapons and personnel led to a serious clash with the rival Official IRA and the death of one Official IRA member. Another of the possible “double-disappeared” was an alleged undercover MI5 agent killed by the IRA in County Kerry. The remains of only five of the fourteen people that the IRA admit to having “disappeared” have been recovered despite claims from the IRA to have provided details of the location of their bodies to the authorities.

That Molloy was able to inflict such damage was due in large measure to the refusal of his immediate superior to believe that one of his own men could be a traitor. It was a weakness in IRA commanders that the British learned to exploit well. Brian Keenan, the IRA’s quartermaster general at this time and Molloy’s commander, has entered IRA mythology as one of its


hardest men, a skilled and ruthless commander who was as determined a revolutionary as existed anywhere in the IRA—but like many military-minded men he had a weak spot in his vanity, and Molloy’s handlers in British intelligence exploited this flaw with consummate skill.

The problem for the British was simple but not straightforward. Molloy had been picked up by the military and had quickly broken during interrogation at Castlereagh RUC station in East Belfast, which was fast becoming the major holding center for IRA suspects, and had agreed to work for military intelligence. But if he was to be of any value to the authorities, he had somehow to be put back into circulation, back into the IRA’s higher circles in Belfast—and fast. The difficulty was that too many of Molloy’s senior IRA colleagues knew he had been arrested, and since the fate of most IRA men who fell into the hands of the authorities was usually immediate consignment to an internment cage, Molloy’s reappearance in West Belfast would automatically attract suspicion. A way had to be found to explain away his safe return from Castlereagh. Molloy’s handlers devised an audacious cover story that worked only because Brian Keenan fell for it, although it required nerves of steel on Molloy’s part when he explained it to the IRA’s quartermaster general. The story concocted by British intelligence was extraordinary: Molloy would tell his IRA colleagues that during a break in questioning one of his interrogators had carelessly left the key to his cell in the lock on the outside of the door. He had spotted this, he said, slipped a sheet of paper taken from the detectives’ carelessly discarded notepad under the door, and using the policemen’s similarly abandoned pen then worked the key free, and fortuitously, it fell onto the paper. He slid this back, unlocked the cell door, and calmly crept out of the station unnoticed by any of his jailers. It was a measure of the hold that his captors had over him that Molloy was prepared to return to the IRA with such a fanciful tale. But incredibly it worked. “Keenan gave him a clean bill,” recalled a colleague.[1]

At the time of Molloy’s arrest and “escape,” Brian Keenan was already one of the IRA’s most important leaders and a key ally of Gerry Adams. Throughout his long IRA career, even during fourteen long years spent later in the high-security sections of various British jails, Keenan, whose IRA nickname was “the Dog,” would stay loyal to Adams and was always there to give support to whatever new direction Adams was advocating. At one point in the peace process he appeared to back dissidents who were plotting Adams’s downfall, but this, it later appeared, was a ruse devised to infiltrate and undermine the Sinn Fein president’s enemies. This hard-line image was a huge asset to Adams. When it came to selling a strategy that filled the republican grassroots with doubt and uncertainty, Keenan’s support helped to win over skeptics. Had he seriously opposed Adams, the outcome of the peace process might have been very different.

Born in Swatragh in rural South Derry in July 1940, Keenan had been in the republican movement since his early twenties. He came from a much more eclectic background than most other Northern republicans of his time and was certainly more politically aware than the scores of recruits flooding into the organization after 1969. The son of a junior member of the British Royal Air Force, Keenan left home when he was just sixteen and emigrated to England, where he worked for a while as a television repairman. Accounts of his early life conflict. One version says that when he went to England he joined the Communist Party and embraced left-wing politics, while another says that he got the name for holding Marxist views only because he was a Goulding supporter when he joined the IRA in the early 1960s.

Much has been written about Keenan’s links to radical Middle Eastern and former Soviet satellite countries in Eastern Europe—he is said to have had extensive links with East German intelligence, for example— but little has ever been substantiated in the way of definite links to such regimes. He has often been portrayed as the sole, uncompromising Marxist revolutionary in the IRA leadership, but this portrait suffers from one major flaw. It was the Goulding group, the Official IRA, which obtained political, financial, and other assistance from the Soviet Union and its allies, not the Provisionals. The Officials received cash, guns, and other aid both from East Berlin and from Moscow, but no evidence has ever been produced to indicate that the Provisionals enjoyed such generosity, which presumably would have been the case had Keenan been as close to these governments as has been suggested. Occasionally Keenan revealed his politics publicly. In an article written in the February 1988 edition of the IRA prison journal Iris Bheag, Keenan, writing under the penname Pow-Wow, revealed that he got his real political inspiration from the neocolonial struggles of South and Central America rather than from the dull orthodoxy of Eastern Europe.

At thirty years of age Keenan was considerably older than most other IRA recruits when he switched to the Provisionals, but he quickly rose through the ranks. By 1971 he had become Belfast quartermaster and two years later was made quartermaster general, succeeding the rebel Catholic priest Father Patrick Ryan, who had been promoted when Denis McInerney was captured on board the Claudia as it made its way from Libya packed with Colonel Qaddafi’s weapons. Keenan’s elevation was another piece of evidence that Northerners were not as excluded from the IRA’s upper reaches in those early days as the Adams camp would subsequently claim. His appointment as QMG also gave Adams another important ally at leadership level, one who was also in charge of a crucially important IRA department.

Keenan’s impact was felt very quickly. While his predecessors in the QM’s department had mostly turned toward conservative Irish-America, the Clann na Gael, and other support organizations for supplies of arms and cash, Keenan was not afraid to seek assistance from revolutionaries elsewhere, even those in strange foreign lands. He was one of the first to realize that the strongly anti-British regime of Libya’s Colonel Muammar Qaddafi could be a rich and regular source of weaponry. By 1972 Keenan was importing RPG-7 rocket launchers and missiles from Libya to add to the IRA’s already fearsome arsenal of Armalite rifles, car bombs, land mines, and endless supplies of homemade explosives.

Keenan had overseen and approved Eamon Molloy’s rise through the quartermaster’s department, from deputy QM to brigade QM, and in 1972, after his arrest and apparently remarkable escape from Castlereagh police station, cleared him of all suspicion of treachery, apparently unwilling to believe that someone he had mentored could turn against him. The British double agent was now free to wreak havoc on such a scale that within two years his activities played a major role in forcing the IRA to call a cease-fire.

Molloy’s treachery led to important arrests. In February 1974 Ivor Bell, Adams’s successor as Belfast commander, was arrested in Andersonstown on the basis of information that was subsequently shown to have come from Molloy. Bell soon escaped by swapping places with a prison visitor on April 15 but was recaptured less than a fortnight later, hiding out in a flat in the affluent Malone Road section of South Belfast. In December 1973 Brendan Hughes also managed to escape from Long Kesh, hidden in an old mattress, and returned to active service as Bell’s adjutant. When Bell was captured, Hughes took over the Belfast command and moved into the IRA’s operational headquarters in Myrtlefield Park, also in the Malone area, where he posed as a toy salesman. He lasted until May 10, a fortnight or so after Bell’s second arrest. The IRA later learned from Molloy that the British had placed Hughes and the Myrtlefield Park house under constant surveillance and could have moved against him at any moment—but did not.

This aspect of Molloy’s spying activities convinced Adams, Hughes, and Bell that the British were using his information to shape and mold the composition of the Belfast Brigade staff to their liking. This was done, they suspected, to boost elements in favor of a cease-fire, by releasing more pragmatic figures from internment while simultaneously rounding up the IRA’s so-called Young Turks, those identified as hard-line associates of Adams, Bell, and Hughes. That way the brigade staff would be composed of figures more amenable to a peace settlement. The evidence for that, they later claimed, was the cease-fire of December 1974, a disastrous cessation that helped seriously debilitate the IRA. That cease-fire opened a fault line between the Belfast IRA and the national leadership, the basis for which was the conviction in the Adams camp that the new Belfast Brigade leaders and the national leadership of the IRA had been the victims of a British intelligence sting.

The British, however, may have had other reasons to put the youthful militants behind bars. In late 1973 the unionist leader, Brian Faulkner, had agreed to share power with the moderate nationalist SDLP and the middle-of-the-road Alliance Party and also to establish a cross-Border Council of Ireland to foster cooperation between the two parts of Ireland.

The Provisionals opposed the deal, known as Sunningdale, after the English civil-service college where the parties hammered out the deal, seeing it as something that had the potential to sideline and defeat the IRA— but it was also bitterly opposed by hard-line unionists who saw the deal as a plot to destroy Northern Ireland. There was widespread agreement within the republican movement that the IRA had better move to kill off Sunningdale before Sunningdale killed it, as a key strategist of the time recalled: “Our objective was to ensure that the Sunningdale Agreement would not succeed. [Daithi] O Conaill was pushing us to blow up [the] Stormont [parliament] with a massive bomb and the Belfast leadership was trying to devise a method of getting a bomb onto a ship and blow it up in order to block the main channel in Belfast harbour. We wanted to make our presence felt as a force without which there could be no solution that was not to our liking.”[2] But before these plans could be put into place, the British, very possibly armed with intelligence from Molloy, moved against the Belfast Brigade. Just four days before loyalists began a general strike aimed at killing off Sunningdale, the Myrtlefield Park headquarters were raided and troops arrested the brigade commander, Brendan Hughes.

The timing of Hughes’s arrest convinced the Adams camp that the British were manipulating the IRA leadership in Belfast and elsewhere. The Young Turks later discovered that secret contacts with the British were being opened behind their backs, contacts that eventually formed part of the negotiations for the 1974 cease-fire. Hughes learned that Jimmy Drumm, a veteran from the 1950s and a member of the IRA Executive, was talking indirectly to the Northern Ireland Office; Drumm was arrested by the Belfast Brigade and questioned, but he denied the story. Nevertheless Hughes complained directly to the Army Council. A week later he was arrested.[3] Among the Adams dissidents the view hardened that Hughes had been kept under surveillance but removed only when he started to threaten the British plans for the new cease-fire.

The conspiracy theorists found evidence from elsewhere to support their suspicions. From inside Crumlin Road prison—then a peeling, overcrowded Victorian hulk used to hold paramilitary remand prisoners—it emerged that the MRF, the British undercover group, had re-formed and was once again dabbling in black operations in a bid to destabilize the organization. The IRA staff in the jail uncovered what they were convinced was evidence of a plot to poison senior IRA figures who were in the jail awaiting trial, as well as plans to sow suspicion and distrust within the ranks. The story began when a young remand prisoner confessed that he had taken part in a number of MRF-type operations carried out on behalf of British intelligence. “They were involved in dirty tricks, MRF sort of stuff designed to get a sectarian war going or to discredit the IRA,” recalled a source familiar with the episode. “They planted a bomb in Corporation Street in a Protestant bar that killed two children and that was wrongly blamed on the IRA. A woman was raped and shot, garages robbed, and so on, all of which was blamed on the IRA.”[4] The remand prisoner, eighteen-year-old Vincent Heatherington from Andersonstown, claimed that British intelligence had trained and armed him and then given him, along with other double agents, a free hand to carry out shootings, bombings, and robberies that would be blamed on the IRA.

Heatherington named his alleged co-conspirators to the IRA leadership in Crumlin Road, many of whom were interned or imprisoned in Long Kesh. His claims were believed, and a reign of terror began in the jails. Those named by the young prisoner were tortured into making confessions to even the most far-fetched wrongdoings against the IRA. “They created paranoia in the ranks…,” admitted one of those involved in the affair.[5] It was of course a classic sting operation, as the IRA staff in Crumlin Road and Long Kesh eventually realized. The aim was to set the IRA prisoners at each other’s throats, and the ploy worked.

The British operation did more than that. Outside the jail it divided the IRA leadership in Belfast between those who took Heatherington’s allegations seriously and those who were either deeply skeptical or who believed that torturing people into confessions was fundamentally counterproductive. This division coincided with the fault line that had long existed in the Belfast leadership and that was now, as the 1974 cease-fire approached, asserting itself with a vengeance. The split was between those associated with Adams, Bell, and Hughes, who had all fallen for the British sting, and Billy McKee, at this point commander of IRA prisoners in Long Kesh, who had tried but failed to stop the torture and mistreatment of the IRA inmates named by Heatherington. The effect of the British operation was to discredit the Young Turks and strengthen the traditionalists who proposed and supported the coming cease-fire. When, in September 1974, McKee finally came to the end of his sentence on the 1971 arms charge and was released, he was not interned, as happened to many released IRA prisoners, but allowed to return to the IRA, where he slipped easily and unopposed back into command of the Belfast Brigade.

Eamon Molloy’s activities were meanwhile having a devastating impact on the IRA’s ability to sustain its war effort in Belfast. The loss of hard-to-smuggle weaponry in particular enervated the organization, and the ceasefire came as something of a relief, as one former IRA officer recalled:

The cease fire was a godsend. We had no weapons. In the Second Battalion of the Belfast Brigade there were three weapons. In fact the situation was so bad that at the start of 1975 while we were still waiting on [a] shipment [from the Irish-American gun-running ring] to come in we went to a member of the Sticks [Officials] and we told him that we felt there was going to be a repeat of the 1969 pogroms but on a larger scale. We asked him for weapons which he refused to part with. We told him that it would be on his conscience if such a situation developed. He then agreed to sell us twenty weapons without the knowledge of his leadership.[6]

The cease-fire arrived hesitatingly. Secret contacts with the IRA leadership were opened up by a group of liberal Protestant clerics, and there were other indirect conversations between the Army Council and the British government, mostly though its Secret Service, MI6, which produced a shortlived truce over the Christmas holiday of 1974 and early new year of 1975. The cease-fire was renewed in early February 1975, and there were reasons to believe that this time it would not be like 1972, and that the British wanted to hold serious talks. Although the IRA was not able to raise its violence to the levels of 1972, when nearly five hundred people had died, 1974 was nevertheless a bad enough year. Killings had risen by nearly 20 percent compared with 1973, and the notion that, in the bleak winter of 1974–75, in the wake of the collapse of Sunningdale and with no other viable options available, a frustrated British government might seriously contemplate talking to the IRA about withdrawal was not that far-fetched.

IF THE IRA LEADERS agreed to the 1974–75 cease-fire with high expectations that it would end with a British commitment to withdraw, they were to be sadly disappointed, although it was easy to see why at the time the Army Council might have entertained such thoughts. The collapse of the Sunningdale agreement sent British policymakers into despair. The power-sharing deal had offered something to everyone, yet it had been decisively rejected both by the Protestant working class—and less ostensibly by much of the unionist middle class as well—and by the supporters of the Provos. In a way the collapse of the Sunningdale experiment seemed to symbolize to the outside world the addiction of the parties in Northern Ireland to their ancient quarrel and spoke to an almost inherited inability on the part of the belligerents to entertain reasonable solutions. With IRA and loyalist violence continuing apace and no political settlement in sight, the British might, it seemed, be inclined to contemplate even the most extreme solution. So it was when, at Christmastime 1974, Sinn Fein’s president Ruairi O Bradaigh received an extraordinary message from an intermediary in Derry that the British wanted to talk about ways of disengaging from Northern Ireland, the Army Council did not hesitate in recommending that the advance be followed up.

As things turned out, the cease-fire was to be just another punctuation mark in a long story, a pause during which the British regrouped their forces and improved their intelligence in preparation for an assault on the IRA that very nearly brought its total defeat. But the cease-fire was to have other unexpected consequences, the most significant of which was that it paved the way for Gerry Adams and his allies to take control of the IRA.

Adams and a small group of like-minded activists grouped around him in jail led the opposition to the cease-fire, and the manner in which they did so created one of the most abiding myths of his career—that he rescued the IRA from the near defeat precipitated by the cessation. The Adams camp would argue that the Army Council of 1974 foolishly led the IRA into a carefully laid trap, the objective of which was to tempt the IRA into lengthy and inconsequential talks designed to buy the British enough time to construct the method of the organization’s downfall. From the confines of their cages in Long Kesh, Adams and his supporters not only put forward the political arguments against the cease-fire and those who had advocated it, but, the myth continued, they designed the practical changes and political-military strategies needed to drag the IRA back from the edge of the abyss.

A crucial part of their case was that the blame for the cease-fire disaster lay with an Army Council that was dominated by Southerners who had no feeling for those fighting the war in the North. But the facts suggested a massaging of the truth. Of the seven Army Council members who sanctioned the 1974–75 cease-fire, no fewer than five came from the North. These included the chief of staff, Seamus Twomey, who had been Belfast commander when Adams was his number two, while another key Council member, Joe Cahill, was a family friend and an ex-Belfast commander who had been particularly close to Adams. There were three other Northerners on the Council: J. B. O’Hagan from Lurgan, Kevin Mallon from Coalisland, County Tyrone, and Seamus Loughran, from West Belfast. The anti-Northern accusation was a potent, if unfair, charge which greatly appealed to the prejudices of those who made up the bulk of the fighting units north of the Border, while serving to undermine the position of Army Council members Daithi O Conaill and Ruairi O Bradaigh, two of the three major IRA figures who stood between Adams and his supporters and their bid to win absolute control of the organization. The third member was Billy McKee, who rejoined the Council upon his release from Long Kesh late in 1974 and who was a figure with whom Adams had repeatedly clashed ever since the Ballymurphy riots of 1970. That McKee was a fellow Northerner and a West Belfast veteran was later forgotten as Adams and his allies portrayed the struggle for the IRA’s soul as a battle between compromising, armchair generals pontificating from the safety of the South and those who actually had to take on the British in the mean streets of Belfast and Derry.

Adams exploited the divisions within the IRA to his own advantage, but so too did the Army Council in its efforts to commence talks with the British. In the run-up to the cease-fire the Army Council held discussions with a group of liberal Protestant clerics in the County Clare village of Feakle, and one of the clergymen vividly recalled the IRA leaders playing the Adams card: “O Conaill told me at the time that we would have one bite at the cherry, no more. He said, ‘That’s because behind every one of us on the Army Council there’s a young man with a gun in his hand who still has to make his name for Ireland and write his name in the history books. And when they take over there will be no more cease-fires.’”[7]

The 1974–75 cease-fire had three major negative consequences for the IRA, each of which would be critiqued by Adams during his subsequent bid to capture the organization. The first was the debilitating effect on the organization caused by both the sheer length of the cessation and the failure or refusal of the British to put flesh on the bones of the secret offer to talk about “structures of disengagement” from the North, as O Bradaigh had been promised by the Derry intermediary.[8]

Talks took place between the two sides, but nothing concrete ever seemed to emerge. There were three levels of contact. At the lowest level, incident centers were set up that linked Sinn Fein and British government monitors in Belfast, where there were seven such offices, and in Derry, Armagh, Newry, Enniskillen, and Dungannon. These were supposed to monitor the truce to ensure there was no 1972-style breakdown. Sinn Fein figures and British officials, communicating mostly via telex machines, would try to make certain that small problems did not become big ones. The enterprise was given an appropriate code name, after the mock-Gothic pile, Stormont Castle, which housed the British administration in East Belfast. Operation Ramparts conjured up a picture of British spies crouched over keyboards at the top of spiral stone staircases as they waited for the latest message of complaint from their IRA counterparts.

At the next level there were regular meetings in Belfast between the IRA Executive members Jimmy Drumm and Proinsias MacAirt and British officials that were held at a British government house called Laneside, in the affluent County Down seaside town of Hollywood. The Laneside talks happened quite frequently, much more often than the highest level of talks that were held in Derry between Army Council delegates, O Bradaigh and McKee, and senior officials from the Northern Ireland Office (NIO). One authoritative account of the contacts depicts the British as quite open about their wish to withdraw but reluctant to spell out the details, even resorting to evasion in response to IRA questions.[9] There were constant squabbles and complaints about breaches of the terms of the cease-fire by either side. The British could produce only one piece of evidence to support the notion that they wanted to be out of Northern Ireland, and that was the exclusion from British plans to nationalize the shipbuilding company of Harland and Wolff, the Belfast shipyard where the Titanic had been built some sixty-five years before. Nonetheless the Army Council perservered.

The British failure to make good on the secret promises gave the Adams camp powerful ammunition. The allegation he and his allies leveled was simple: the Army Council had been fooled into taking part in bad-faith negotiations and tricked into calling a cease-fire that was designed to last so long that it seriously eroded the IRA’s fighting ability. The result was that the British were given a breathing space in which to devise ways of inflicting even more damage on the IRA. With IRA members relaxing visibly as the cease-fire dragged on, the British built up their intelligence on the organization in preparation for major changes in security policy, the central feature of which was that control of operations against the IRA passed from the British military to the Royal Ulster Constabulary and a new emphasis was put on legal methods to put IRA members behind bars. The use of internment was phased out, and instead more conventional judicial methods were employed, primarily the processing of IRA suspects through the courts, albeit courts that had no juries and in which verdicts and sentences, fashioned by a barrage of new antiterrorist laws, were handed down by a single judge. The change affected the prisons as well; special-category status, which had been granted to IRA prisoners as part of the 1972 cease-fire deal and which recognized a political motive for paramilitary activity, was scrapped, and instead IRA inmates were to be officially regarded and treated as common criminals.

The British strategy was known variously as Ulsterization, normalization and criminalization, but whichever word was used the result was the same. The British had put in place a system that, from 1976 onward, saw IRA suspects trundled along a sort of conveyor belt that began with arrest and lengthy questioning in new RUC interrogation centers where confessions would be extracted, often amid claims of brutality. The next stage was in the courts, where, despite their often dubious integrity, the confessions were invariably accepted, and the last stage was in the Maze prison, the renamed and rebuilt Long Kesh camp but soon known to the world as the H Blocks, where IRA prisoners were to be treated in the same way as thieves and rapists. The new security policy cut swaths into the IRA, bringing it to the verge of defeat. The fault for this, the Adams camp was to insist, belonged to those who had negotiated the cease-fire.

Terrifying and cold-blooded loyalist violence, most of it directed at ordinary Catholics, particularly in Belfast, was the second result of the 1974–75 cease-fire. Unnerved by the IRA’s contacts with the British, loyalist paramilitary groups began murdering Catholics in an often gruesome fashion and in a way that challenged the Provisionals’ claim to be the defenders of their community. As the loyalist killing increased, the pressure to hit back in like manner intensified.

The loyalists’ logic was terrifyingly simple: the more Catholics they killed and the more horrible the manner of their deaths, the stronger the pressure would be on the IRA to stop its activities. The cease-fire was the trigger for the bloodshed. When the British and the IRA announced the terms of their truce, the loyalist groupings, especially the UVF, reacted with unusual violence. Within seven days they had killed five Catholics, and many more were to follow.

AS 1975 UNFOLDED, Northern Ireland, and Belfast in particular, entered what is arguably the darkest period of the Troubles, nearly two years of slaughter in which the loyalists and the IRA vied with each other in an often indiscriminate sectarian killing game, the latter refusing to admit its activities or seeking refuge behind fictitious cover names, the former glorying openly and proudly in the carnage. These were the years that saw the rise of chilling loyalist gangs with names like the Shankill Butchers, a group of UVF killers who specialized in cutting their victims to death with surgically sharp knives and axes. The IRA responded with bombings of Protestant pubs that invariably killed uninvolved civilians. Increasingly the IRA imitated the loyalists by kidnapping and killing innocent Protestants. The violence spread to rural areas of Tyrone and Armagh, where IRA members using cover names such as the South Armagh Republican Action Force to disguise their bloody deeds enthusiastically joined the slaughter. By the end of 1976 loyalists had killed nearly 250 people, most of them innocent Catholics, while 150 Protestants, similarly unconnected to those who were directing the murder campaign, had also met violent deaths, a majority of them at the hands of the IRA, however disguised or renamed.[10]

The upsurge in loyalist killing put the IRA in a quandary. To respond in kind threatened to drag republican values into the gutter, making a nonsense of IRA claims to be a nonsectarian movement in the tradition of the Protestant founder of republicanism, Wolfe Tone. Yet if the Provisionals stood aside, this would be to deny the atavistic forces that had brought them into being in the first place, the need to defend their streets and communities. Most Provisionals agreed, some with less enthusiasm than others, with the view of one IRA leader of the day: “Republicans had to hit back at the Loyalists. It was as simple as that. They were slaughtering the Nationalists. There was no other way round it.”[11]

The Adams camp chose to see things in a more negative light and courtesy of the same conspiratorial prism through which they had viewed the arrests of Hughes and Bell and the treacherous activities of Eamon Molloy and Vincent Heatherington. The sectarian warfare, they claimed, was being sponsored by the British, whose aim was to pull the IRA into a brutal but diverting tribal conflict in which the British could depict themselves as “the piggy in the middle,” as neutral arbiters, whose only interest was to stop the irrational violence of mad Irishmen. For this state of affairs Billy McKee, the defender of St. Matthew’s and once again Belfast commander, was held to blame. His motives were regarded as beyond justification. “McKee just wanted to get into a war with the Orangies,” concluded one of the Adams camp at the time.[12]

The third consequence of the cease-fire was an outbreak of vicious feuding between rival republican groups, a hangover from the split of 1969, but something the Adams camp also maintained to be in the British interest. Billy McKee was, in their mind, guilty of allowing the IRA to get involved in sectarian killings, and they blamed him for the feuding as well. The worst piece of internecine slaughter between Provisionals and Officials broke out in Belfast at the end of October 1975. It started with around a dozen attacks mounted by the Provisionals in the course of an evening during which one member of the Official IRA was killed and sixteen were wounded. Up to ninety Provisional IRA members were said to have been involved in the night’s violence. The feud ended a fortnight later with a total of eleven dead, including a six-year-old girl, fatally shot during an attack aimed at her father, a member of the Official IRA. Feuds between the Provisionals and the Officials were not unknown, but this was by far the most serious outbreak of such violence. McKee justified the attacks by arguing that the Officials were acting as an arm of the British in areas controlled by the Provisionals, passing on information about IRA activists, facilitating British community policies, and undermining Provo influence. But the feud, coming on top of the loyalist onslaughts, was deeply demoralizing inside nationalist areas, and Adams seized on this to criticize McKee, as he later recorded: “Republican feuding contributed significantly to feelings of alienation on the part of the Nationalist people, who had long provided the essential support for the IRA. It also coincided and dovetailed with a sustained British propaganda campaign to portray IRA members as ‘common criminals.’”[13]

The long-standing hostility between Adams and McKee, which had its roots in the Ballymurphy riots of 1970, came bursting to the surface again but this time in surrogate form. Even so, the dispute can with hindsight be seen as the first shot fired by Adams in his campaign to oust McKee and to engineer his takeover of the republican movement.

The target chosen by Adams was David Morley, the commander of IRA prisoners in Long Kesh. The former leader of the Provisional IRA in the County Down Border town of Newry, Morley took over the job of camp commandant from McKee just prior to McKee’s release in the autumn of 1974. The two men were on close terms, and when McKee assumed command of the Belfast Brigade and rejoined the Army Council, Morley became the principal point of contact between the leadership and the Long Kesh prisoners.

Largely because of McKee’s patronage, Morley had been elected camp commander by the rank and file in the jail, but his style was not universally popular. “Morley came from a British army background and ran the prison, much as McKee had done, on very militaristic lines,” remembered one former IRA prisoner who supported the Adams line. “There were roll calls, parades, tight discipline, and so on, all of which rankled with many people, especially young lads. Morley introduced a new IRA salute, two salutes to the head rather than one, so as to distinguish the IRA from the British army. Staff members wore Sam Browne belts and a holster with a wooden gun stuck in it.” Adams would describe it all as a cross between George Orwell and Spike Milligan.[14]

To the astonishment of many in the camp, Morley told the prisoners that McKee wanted them to scrap existing escape plans and not to bother hatching any more. “The Army Council told us that we were all going to get out, internees and sentenced men,” recalled one former prisoner. “First there would be 50 percent remission, then two-thirds, then all out.”[15] This was, the Morley-McKee critics complained, the ultimate in naïveté for at the same time this advice was being given, the IRA leadership was fully aware that the British were planning the construction of a brand-new prison, the H Blocks, to house IRA inmates. Morley had toured the first H Blocks, so named after the shape of the cell wings, and could see they were intended for something much more permanent. He was also allowed out of the jail on parole for talks with Jimmy Drumm and with British officials. “Mixing at this level with senior Brits went to his head,” suggested one critic.[16]

Morley’s elevation coincided with changes to Adams’s own prison circumstances. He had initially been housed in the internee section of Long Kesh, in Cage 6, which he shared with some one hundred other inmates in three huts. But after two unsuccessful escape attempts, Adams was sentenced to three years’ imprisonment and was moved to the sentenced-prisoner section of the camp, into Cage 11. Coincidentally, hostility to the leadership line had been growing over the cease-fire and the leadership style in jail; a large number of prisoners had been suspended from the IRA for disputing Morley’s authority. When Adams arrived in Cage 11, it was decided that a challenge should be made against the camp leadership, and Ivor Bell ran against Morley in the election for camp commander and fared surprisingly well. “The purpose of the exercise was to show there was unease with what was happening,” recalled an Adams supporter. Morley won but by a margin of just a handful of votes. Adams later stood in a separate poll and again ran Morley close, so close that the result badly scared the IRA leadership.[17]

One outcome of the election was that the Army Council changed the rules governing the selection of prison OC. From then on the rank-and-file IRA inmates could nominate whomever they liked, but there would be no election. Instead the Army Council would choose the commander it wanted out of the list of candidates submitted by the prisoners. The idea was to prevent splits from coming out into the open, but it enormously strengthened the leadership’s hand in the jail; ironically, this was later to greatly benefit Adams when he assumed control of the IRA and wanted to be sure of the loyalties of the prison leadership. Although he had been a victim of the change, he made no alteration in the rule when he and his allies assumed control of the IRA.

From early 1975 until Adams’s release in February 1977, two cages in Long Kesh were the focus for dissident criticism of the leadership in Dublin. One was Cage 11, where first Adams and then Brendan Hughes was OC, and Cage 9, commanded by Ivor Bell. But as the new British security strategy swung into action and the Castlereagh conveyor belt gathered speed, disgorging more and more IRA members into prison, Cages 11 and 9 also became think tanks devoted to planning new structures and policies designed to rescue the IRA from what every activist, inside and outside the jails, could see was imminent defeat.

The period was vividly symbolized by the British Labour government’s choice for Northern Ireland secretary in 1976, a blunt and diminutive former coal miner called Roy Mason, who saw his job in simple terms. He had come to Belfast not to indulge in dangerous political experiments like power-sharing but to crush the IRA. In one of his earliest public pronouncements he delighted unionists with a promise to roll up the IRA like “a tube of toothpaste.” And such was the initial success of the new police interrogation centers at Castlereagh, at Strand Road in Derry, and at Gough Barracks in Armagh in extracting confessions from IRA Volunteers that he felt strong enough to predict, “My view is that [the IRA’s] strength has waned to the point where they cannot sustain a campaign.”[18] The security statistics appeared to support his boast. The death toll for 1977 went down to 116, about a third of what it had been the year before. Even though this reflected a lower level of loyalist activity, there was no doubt IRA operational capacity was being undermined. Shooting incidents in 1977 were down 45 percent from 1976 and bombings by more than half.[19]

IRA MEN have always used their time in prison well. After the 1916 Rising in Dublin the defeated rebels were jailed in Britain, and one of their leaders, a young Michael Collins, used the time, leisure, and opportunity provided by his spell in the Frongoch internment camp in Wales to restructure the Irish Republican Brotherhood in preparation for the coming conflict with Britain. Gerry Adams, Ivor Bell, and Brendan Hughes spent their days in Long Kesh sixty years later doing something remarkably similar for the Provisional IRA. There was one major difference, however. The ideas produced in Cages 11 and 9, many of which Adams outlined in a lengthy report to the IRA leadership in 1976, were more than a rescue plan for the IRA; they were also a blueprint for his takeover of the movement. Their significance does not stop there. Within the plans can also be found the seeds of what, only a few years later, would become the Irish peace process, although it is doubtful if any of those responsible for drawing them up could see that at the time.

The conceptual foundation stone of the takeover plan was the doctrine that the war against the British was going to be a long-drawn-out affair and that the heady days of 1972, when the IRA could realistically imagine forcing a speedy military defeat on the British, were gone forever. As with other controversial ideas, Adams employed junior allies to float the notion first, a ploy that allowed him to gauge grassroots reaction before declaring his own hand. In this instance his choice was Danny Morrison, a bright young West Belfast IRA member who had been interned but then released as part of the 1975 cease-fire deal. Morrison, who had initially backed the McKee leadership against Adams and then switched sides, eventually replaced Daithi O Conaill as the IRA’s director of publicity and went on to become a strong ally of Adams on the Army Council.

One IRA activist of the mid-1970s can remember when Morrison first expounded Adams’s “long war” doctrine.

He was the first I heard it from sometime in 1975. We had one or two mini- [IRA] Conventions, one in Clare and one in Donegal involving twenty-five or thirty of the top people, and Morrison was saying we have sold the people a false bill of goods with slogans like ‘Victory in 74!’ and so on. People were getting cynical, and we would have to say instead that it is a long war. It grated with a lot of us. None of us believed it would be a long war. We were of the opinion that we could win it, we could force the Brits to pull out. We were young and in our twenties, we had seen the fall of Stormont, burned down the British embassy in Dublin, ran the Brits ragged in the countryside. We were of a generation that had seen Saigon fall and the U.S. defeated in Southeast Asia.[20]

The “long war” became IRA policy, but estimates of just how long the Long War would have to last varied enormously. Writing in the Belfast IRA paper under the pen name Brownie, Adams envisaged the conflict lasting for just seven more years, that is, until 1983.[21] But privately he and his allies talked of a twenty-year conflict.

Whatever the reservations, the “long war” doctrine was really just a statement of the obvious. Roy Mason had nearly defeated the IRA, and it was going to take years of careful rebuilding before it would again be in a position to challenge the British, if ever. The signal that this had been formally accepted by the Army Council as IRA policy came only four months after Adams’s release from Long Kesh, at the Bodenstown commemoration of June 1977, the highlight of the republican calendar. Adams’s choice of speaker to give the address sent its own signal. Jimmy Drumm, who had been one of the senior IRA figures involved in the secret Laneside dialogue, climbed the podium beside Wolfe Tone’s grave to declare that he and the other leaders associated with the 1974–75 cease-fire had all been wrong. In a script composed by Adams and Morrison, Drumm declared, “The British government is not withdrawing from the Six Counties. Indeed the British government is committed to stabilising [them] and is pouring in vast sums of money… to assure loyalists and to secure from loyalists, support for a long haul against the IRA.”[22]

The speech crafted for Drumm went further and contained a hint of what was to come, although few of those present at Wolfe Tone’s graveside that day would have realized the full significance of his remarks. “We find,” he went on, “that a successful war of liberation cannot be fought exclusively on the backs of the oppressed in the Six Counties, nor around the physical presence of the British Army. Hatred and resentment of the army cannot sustain the war…”[23]

Heavily coded though these remarks were, the import was clear. The first message was that if republican activists wanted to sustain the war effort, they would have to expand their support base in the North by becoming politically active. Adams had already described how he saw this being done, in a series of articles written in Long Kesh and published weekly in Republican News under the Brownie pen name. Dropping abstentionism, recognizing the partitionist parliaments, and running for office, at least in the North, was still forbidden territory, but Adams nevertheless put the idea onto the Provisionals’ agenda, albeit in a more subtle way. He called his approach “active abstentionism,” a philosophy that in theory involved republicans’ building alternative governmental structures and becoming relevant to the needs of ordinary people.

Borrowed from his days as a Goulding republican, “active abstentionism” was just another way of encouraging political activity in an organization in which “politics” was still a dirty word. The plans to construct alternative structures never really got off the ground, but the idea that republicans should “do things for the people” took root and later found more tangible expression when Sinn Fein started to fight elections and in an approach that was christened the “Armalite and ballot box” strategy. The really significant feature of the move was the way Adams sold it to his colleagues, largely on the basis that increasing political support would enable the IRA to intensify and sustain its war effort. It would not be the last time that he justified political activity in military terms even if, in the end, the effect was to undermine the IRA’s war effort.

The same military argument was used to justify another major theme in the Drumm speech, the idea that the republican movement should try to build political support in the South. Increasing political support south of the Border, Adams argued, would translate into more safe houses for the IRA to use, more money, and more recruits. The argument was uncritically accepted internally, even though it flew in the face of the reality of the Ireland of the mid-1970s. The truth was that by that point in the Troubles, the conflict in the North was something most Southerners wanted nothing to do with, and no amount of political work by Sinn Fein was likely to change that. Nevertheless the Adams dictum became IRA policy, and with hindsight it seemed a highly significant move. Its effect was to introduce a contradiction into Provisional politics that played a major role in the evolution of the peace process. Political ambition south of the Border and armed struggle in the North were mutually exclusive. One or the other could prosper but not both. The contradiction was sharper in the South but it also restricted Sinn Fein’s room for growth in the North, even though the IRA’s campaign had created a sizable support base. Adams had set the republican movement on a journey that would eventually reach a crossroads, where it would have to choose between politics and the gun.

When it came to putting flesh on the theory of “active abstentionism,” Adams and Bell turned to the Libyan regime of Colonel Muammar Qaddafi for inspiration. Qaddafi had outlined a system of people’s committees or popular congresses to help govern Libya. Organized in a pyramidal fashion, the people’s committees oversaw different aspects of government and elected delegates to a general people’s congress, Qaddafi’s version of the Libyan parliament. This system, Qaddafi wrote, is “the only means to achieve popular democracy.”[24] Adams and Bell copied the idea and suggested that people’s committees be established separately from British government structures in republican areas of the North. Adams extolled the idea in a Brownie article: “We need an alternative to the British administration in our country. Especially now in the 6 Counties and when the Republican movement has control and administration to some degree in all the Nationalist areas. So why not cement this into local government structures. In Belfast alone could not the three or four big Nationalist areas be organised into community councils?”[25]

The concept of people’s committees made a brief appearance after Adams’s release from jail, but by mid-1979 talk of building alternative government structures had virtually disappeared. The reason given was the arrest of the group of Sinn Fein activists charged with bringing them into existence, but the idea had proven to be simply impractical. One bit that did survive, however, was the Civil Administration, or Administrative IRA, which was created to police the areas within which the IRA had influence. With the Troubles now nearly a decade old, law and order had broken down in many nationalist districts and established value systems had been upturned. Crime, vandalism, and joyriding were endemic, and there was a demand for a policing system. Rather than see the RUC back in their areas, the IRA began to dispense its own system of rough justice and to mete out punishment shootings, beatings, and expulsions.

While it appears that some of Adams’s associates, Bell in particular, were motivated by an admiration for Qaddafi, the Long Kesh dissidents were also trying to ingratiate themselves with the Libyan leader in the knowledge that he had the wherewithal to arm and finance the IRA beyond its wildest dreams. This consideration was evident in two other structural changes advocated by Adams and Bell.

One was the Revolutionary Council, which came into being in 1976–77 and took its name from a central part of the Libyan system of government. In Libya the Revolutionary Council was the executive branch of government that sat atop the people’s committees. It consisted of Qaddafi, a general people’s committee chosen from the People’s Congress, and the ministers of the principal government departments. The IRA’s version was a sort of mini-Convention, a sounding board of IRA opinion representative enough of the grassroots, yet small enough to be able to meet safely. The regular IRA General Army Convention could draw as many as a hundred delegates, and the risks attached to convening such a large gathering were considered too great in the mid-1970s. The last Convention had met in September 1970, and much had changed in between. Adams and Bell saw the Revolutionary Council as a way of bringing the IRA leadership into touch with the rank and file, among whom their influence ran strongly.

A mini-Convention held in late 1976 agreed to set up the Revolutionary Council and determined its makeup. It would consist of the seven-man Army Council, the GHQ staff, and the commander, adjutant, and quartermaster of the Belfast, Derry, Mid-Ulster, East Tyrone, South Derry, and South Armagh Brigades. Allowing for overlapping at Army Council and GHQ level, the Revolutionary Council had perhaps thirty to thirty-five members.

Bell had intended the Revolutionary Council to effectively replace the Army Council as the IRA’s primary decision-making body. As a radical— some would say a Marxist, others an anarchist—Bell distrusted the middle-aged conservatives who dominated the Army Council, and he wanted to sideline them by subjecting them to IRA democracy. Adams, however, used the Revolutionary Council in a very different way, to control and bend the Army Council to his way of thinking but not to replace it. In his hands the Revolutionary Council became an instrument for taking over the leadership of the IRA, but the Army Council, a smaller body and thus easier to manipulate, remained, under his direction, the supreme decision-making body. The Revolutionary Council was phased out when Adams’s control of the Army Council was complete, although it was revived, to perform a similar function, during the peace process many years later. “It was at this stage that a gap started to open up between Adams and Bell,” observed a contemporary.[26] While Bell wanted to broaden and radicalize the IRA, Adams sought merely to curb and then control the Army Council.

Qaddafi had outlined his political philosophy, a compromise between communism and capitalism that he termed “the third way,” in a three-volume publication called the “Green Book,” so called because green is the color of Islam. Again in imitation of their putative arms supplier, Adams and Bell proposed that the IRA should have its own “Green Book,” which all IRA recruits would be obliged to read and digest before being admitted as full-fledged Volunteers. Thereafter such IRA members would be described as having been “Green-Booked.”

A cross between a political manifesto and a training manual, the hundred-page “Green Book” set out the IRA’s fundamental beliefs and political and military strategies. It also included a copy of the most up-to-date IRA constitution, often an extremely sensitive and revealing guide to the organization’s political disposition and intentions, which IRA leaders strove to keep secret. The result was that the Green Book’s circulation was unnecessarily restricted, bestowing upon it, in the public mind, a sense of mystery and importance beyond its merit. The IRA, it said, was the lawful inheritor of the First Dail and was the legal government of Ireland. The Dail in Dublin and any parliament at Stormont were thus bogus and illegal, the “puppet governments of a foreign power and willing tools of an occupying force,” according to the text.[27] The IRA’s long-term aim was to create a democratic socialist republic, and Volunteers had the moral right to kill to achieve it, not least because they were acting on behalf of the true Irish government. IRA strategy would be based on a number of tactics, the Green Book said, including a war of attrition against “enemy personnel,” a bombing campaign to deter inward investment, opposition to all attempts to create internal political stability, and a propaganda campaign in Ireland and abroad aimed at broadening support for the war effort.[28]

To most rank-and-file IRA members, however, being “Green-Booked” really meant they had gone through training in the anti-interrogation techniques outlined in the manual. The RUC’s holding centers at Castlereagh, Gough barracks, and Strand Road were having a devastating impact on the IRA, and there was an urgent need to instruct Volunteers in ways of resisting police questioning, as one member recollected: “Men were breaking in the police stations. We’d hear of people handing over twenty-five to thirty names at a time. In the first twelve to eighteen months of Castlereagh we suffered great damage.”[29] The lectures in the “Green Book” on anti-interrogation methods, which basically boiled down to tutoring IRA members—seasoned activists as well as novices—in how to remain silent during questioning, came into effect around late 1978 and early 1979. From then on the only defense that an IRA member could make to a proven accusation of informing was that he or she had not been “Green-Booked.”[30]

The Army Council chose a brutal way of demonstrating to Volunteers the importance it attached to the “Green Book” and its admonition to stay silent during interrogation. In July 1979 Michael Kearney, a twenty-year-old IRA Volunteer from Lenadoon, was found dead near Newtownbutler in County Fermanagh with a bullet wound to the head. In a statement the IRA said he had been killed “for breaches of general army orders in that he imparted information of vital importance to the British war machine.”[31] Later the organization’s spin doctors suggested Kearney had betrayed a major bombing operation in East Belfast aimed at destroying oil depots and industrial plants. But within the IRA the strong belief was that Kearney had been killed only because he had broken during RUC interrogation and that his death was meant as a warning to others. A few years before, when such examples of weakness had been commonplace, the worst that most IRA members could expect was to be boycotted and isolated by colleagues in jail. Kearney’s death was intended as a signal that this leniency had gone and that others should take the “Green Book” with deadly seriousness.[32] In September 2003, the IRA finally admitted that Kearney had not been an informer, thereby implicitly conceding that he had been killed only because he had violated the Green Book’s admonition to stay silent during police questioning.

British informers were high on the dissidents’ target list. In their reorganization scheme, Adams and Bell suggested the creation of a specialist counterintelligence unit, known as the security department, which was initially charged with debriefing IRA members who had been in police custody with the aim of discovering whether they had talked to their interrogators or been turned by them; initially confined to Belfast, the security department was eventually extended to Northern Command and, over the years, its powers to probe IRA affairs were expanded. The security department was also responsible for arresting, interrogating, and ultimately arranging the deaths of informers, a great number of whom would end their days hooded and trussed at lonely roadsides in places like South Armagh. After 1977 the number of deaths of alleged informers rose steeply; one estimate suggests that as many as 70 percent of all informers caught by the IRA during the entire Troubles were killed after that date.

The security department proved to be a double-edged weapon, however. Infiltrating the new department became a priority for the RUC and British intelligence for one very simple reason. The security department’s members knew many of the IRA’s most intimate secrets, including the identity of key gunmen and bombers, and a double agent placed within their ranks could cause havoc. The years since the department was set up have been characterized by persistent suggestions that this is just what happened.

The idea for the internal security department came in part from the special “unknown” cells set up by Adams in 1972. Part of the task of “the unknowns” had been to lure alleged informers to their deaths and to engage in counterintelligence work. “The unknowns” were also the inspiration for another part of the reorganization plan, and that was the refashioning of the IRA into secret cells and the dismantling of the old company and battalion structures. Modeled on the British army, the IRA’s old system of companies and battalions were based in well-defined geographical areas, and this made it relatively simple for the British to work out which units were responsible for what operations. The structure also made easier the task the RUC Special Branch and other intelligence agencies faced in mapping and identifying the IRA’s battle order so that when the interrogation centers swung into action the organization quickly buckled.

Adams was asked by the Army Council to design the reorganization plan, a draft copy of which was found by the Gardai when they arrested Chief of Staff Seamus Twomey in Dublin in December 1977. The kernel of Adams’s scheme was the creation of secret, four-man active-service units (ASUs), which would specialize in activities such as sniping, bombings, assassinations, and intelligence work. They would have a roving commission and could operate anywhere in the IRA’s war zone so that the British would find it more difficult to identify which ASU had been active where.

The restructuring of the IRA, its new political-military strategy, the “Green Book,” the internal security unit, and the course of anti-interrogation lectures for recruits and veterans were all parts of a strategy designed to facilitate the conduct of the “long war.” The theory was that between them the changes would enable the ASUs to ward off penetration by British intelligence and resist RUC efforts to force confessions out of the activists, thus enabling the IRA to survive and fight for years to come. At the same time the IRA’s new political activism would, again in theory, expand the support base and provide a steady supply of new recruits, safe houses, and so on.

On top of all this Adams proposed another major innovation, the idea that the IRA should have “permanent leadership” at all levels. The argument put forward in favor of this innovation was that the IRA lacked consistency at the commander level and was too easily disrupted by arrests and harassment. Since the start of the IRA campaign, Adams and his allies argued, there had been too many chiefs of staff, too many changes in Army Council and GHQ personnel, and, lower down too high a turnover at the brigade staff level. Each time a top commander was arrested and jailed, valuable talent was lost, and precious experience and knowledge would disappear and have to be replaced in a lengthy and often difficult process.

Adams’s proposal was backed by a powerful group of activists, including Danny Morrison, Brian Keenan, Martin McGuinness, and Adams’s cousin Kevin Hannaway. They got their way, and while the change was not always implemented, where it was enforced it did not come cost free. In an important way the concept of “permanent leadership” made the IRA more vulnerable to British penetration; a commander turned by the RUC or MI5 could be in place for a very long time. There was another unforeseen consequence. In practice the only way “permanent leadership” could work was for commanders to become operationally inactive so as to avoid any risk of arrest. That was a significant break with Provisional IRA tradition and practice, in which leaders had often led by example, fighting alongside Volunteers. One outcome was the development of a self-perpetuating elite at top- and middle-level ranks whose composition was often the result as much of loyalty to the political strategy of the Army Council, and ultimately to Gerry Adams, as of battle skills. The concept, the fact that leaders put away their guns, also paved the way for hitherto secretive senior figures to emerge as public personalities, an essential prerequisite for a successful electoral strategy.

Much of the Adams-Bell blueprint was beginning to be implemented before Adams’s release from Long Kesh. With Adams, Bell, and Hughes in jail, the task of selling the restructuring plans both to the Army Council and to the middle leadership fell in part to Martin McGuinness, the Derry IRA leader who had spent much of the cease-fire period in Portlaoise prison after a conviction for IRA membership. But it was mostly Brian Keenan who toured the country preaching the Cage 11 gospel to IRA members. “More than anyone else,” remembers one former IRA man, “Keenan was a roving ambassador for Adams.”[33]

The most important and meaningful part of the restructuring scheme was also its most controversial. Adams proposed the division of the IRA into two separate geographical entities. A Northern Command would be set up, comprising the six counties of Northern Ireland and five Border counties in the Irish Republic (Louth, Cavan, Monaghan, Leitrim, and Donegal), an area that coincided with what was effectively the war zone where military operations were planned and carried out. A separate Southern Command was proposed in the remaining twenty-one counties that would specialize in providing the logistics necessary for conducting the campaign in the North. The Southern Command would supply three essential elements: most of the IRA’s arsenals would be hidden there, most of the “factories” churning out homemade explosives and the IRA’s often ingenious improvised weaponry would be based there, and so would the organization’s training camps, often situated in isolated mountainous countryside or in underground firing ranges painstakingly excavated in lonely farmland and forests.

Important though these functions were, there is little doubt that the proposal for a Northern Command would herald a significant switch in the internal balance of power. The IRA’s principal business was to fight a war against British forces in Northern Ireland, and those who controlled how that war was conducted—and those who appointed them—would inevitably wield the greatest influence in the organization. The Northern Command would have its own staff, which would shadow that in GHQ, which promptly lost its responsibility for conducting operations in the North and was thus weakened. Thereafter GHQ’s direct military role was confined to the IRA’s international activities, principally in Britain and Europe. The construction of a Northern Command would make the position of Northern commander one of the most important positions in the organization. Not least of the Northern commander’s powers would be that of having a major say—in later years the final say— in deciding who became brigade commanders and who made up their staff.

The idea of creating a Northern Command was always going to be a contentious proposal, and so it was, especially with older republicans who had bitter memories of the last time, aside from the brief period before and after the 1969 split, when the Northern and Southern parts of the IRA had gone their separate ways. A Northern Command, composed of the six Northern Ireland counties and County Donegal, was established in 1939 just after the outbreak of the Second World War when cross-Border communication difficulties made the change necessary. The IRA had begun a bombing campaign in England around the same time, but within two years arrests by the British police had brought it to a halt. The IRA in the South, meanwhile, was enfeebled by arrests and arms seizures. The police forces in Ireland and Britain clearly had good intelligence, which the Northern commander, Sean McCaughey, and his allies suspected was being supplied by a traitor in the Dublin leadership. They kidnapped the chief of staff, the Wexford man Stephen Hayes, and held him for several weeks, during which time he was tortured, starved, and repeatedly questioned about failed operations and alleged Garda successes. After a court-martial, which predictably convicted and sentenced him to death, Hayes was forced to write out a lengthy confession. He managed to drag this out until an opportunity arose to escape. Hayes turned himself in to the Gardai and survived the ordeal, but McCaughey was captured and later died on a hunger and thirst strike. The Hayes affair and its consequences deeply divided the IRA on North–South lines. Many activists of the day blamed the Northern Command experiment for the acrimony and vowed never to repeat it.

The “Forties men” were the strongest opponents of Adams’s proposal, among them Joe Cahill, Proinsias MacAirt, and Billy McKee, who was probably the most bitterly against the idea. Their common fear was that a Northern Command would precipitate a split. But Seamus Twomey, the chief of staff and, in the words of one IRA militant, “Adams’s pet Rottweiler,”[34] backed the proposal, and Brian Keenan did the rest. Most of Belfast supported the idea, as did Martin McGuinness in the Derry Brigade, while Keenan, still the IRA’s QMG, won over crucial middle leadership people in three other vital areas, South Derry, East Tyrone, and South Armagh. Keenan’s argument was simple: a Northern Command would be responsive to the needs of the men who were fighting the war. “He told us we wouldn’t have to wait to see someone from Kerry,” explained a Northern activist. “We asked would a Northern Command mean that we would get more gear, and Keenan would say bigger, better, longer guns and then it was okay.”[35]

Another rural Northern IRA member takes up the story:

Adams was still in jail at this time, so Keenan was really the John the Baptist to Adams’s Christ. Into our midst during our perennial quest for gear would come Brian Keenan. Unlike other leaders he was an activist. He would arrive in jeans and denim jacket and would sleep in the ditches along with the Volunteers. He was good fun, clever, likable and would always arrive with a bit of gear wrapped up in plastic, fifty or a hundred rounds for some weapon. Brian could always turn up stuff for esoteric weapons. I remember we had a Chinese version of the AK-47 called the SKS, which used short 7.62 mm ammo as opposed to the British-NATO standard long round, and he got some for us. He would sit and talk things over with us; we’d go and get a carryout to eat; he charmed us and won us over.[36]

Keenan and Adams were knocking at an open door, and the establishment of the Northern Command became a foregone conclusion. In late 1976 a meeting of the Revolutionary Council passed a resolution recommending to the Army Council that the change be made; faced with the united voice of the fighting units in the North, the Army Council had little choice but to agree. Martin McGuinness was appointed the first Northern commander but on Billy McKee’s insistence the powerful Belfast Brigade was excluded from the new arrangement. McKee was still OC of Belfast Brigade, albeit hanging on by his fingertips, and correctly saw Northern Command as an effort by the Adams camp to extend their power base. Only when Ivor Bell took over from McGuinness, when he was appointed chief of staff in early 1978 in succession to Gerry Adams, was Belfast integrated into Northern Command.

The creation of the Northern Command had important and long-lasting consequences for the IRA. Not only did it facilitate coordinated Northern Ireland-wide attacks and make rapid alterations in military tactics more feasible, it established and refined central control over the IRA’s cutting edge. This was especially the case outside Belfast in militant areas like East Tyrone and South Armagh where the local IRAs were really under the control of clan chieftains—Kevin Mallon in Tyrone and later Tom “Slab” Murphy in South Armagh were examples—rather than part of a structured, centrally directed organization. “The leadership would never try to give them orders,” recalled a rural activist. “There was virtually no control from the center. They mounted operations against the British, and the job of leadership was to provide resources, training guns, explosives, etc. You just could not guarantee that they would vote at a Convention for the leadership’s political strategy. That independence all but disappeared with the Northern Command.”[37] Creation of the Northern Command marked a vital staging post on the road to Adams’s establishing control not just of the IRA’s military tactics but of its political direction as well.

The new structure also standardized the IRA in another important way. Prior to the Northern Command each area in the North had organized its own training camps and procedures. Separate camps had existed for Belfast, Tyrone, Derry, and Armagh IRA units, but after the change training was centralized and procedures regularized and, inevitably, improved. This development had its negative side. If a training camp was stumbled upon or betrayed to the Garda, Special Branch surveillance would now reveal much more about the IRA than before.

There were other adverse consequences from the creation of the new body. The IRA lost a great deal of the spontaneity and unpredictability that had made it such a difficult quarry to corner in the early 1970s. After Northern Command was set up, Adams sent emissaries throughout the countryside to propound the new gospel, with the result in some instances that well-protected safe houses were revealed and local command structures disclosed to outsiders. Some areas were forced to share hitherto jealously guarded technology. The IRA in South Armagh, for instance, had developed radio-controlled bombs but had refused to allow other areas to use them, for fear that the British would capture them and work out countermeasures. Once the Northern Command was set up and the technology shared, that is precisely what happened.

IN FEBRUARY 1977, after the best part of four years spent in the Long Kesh prison camp, Gerry Adams was released into the loving arms of Colette and Gearoid, his three-and-a-half-year-old son, who had been born while he was in jail. Adams had seen his young son grow up but only in the visiting rooms of the jail and under the watchful and often intrusive eyes of the prison guards, the hated “screws.” Now for the first time they could be a proper family, inasmuch as an IRA leader ever had a proper family life. Despite the risks associated with the IRA lifestyle, at least they knew that the fear of arrest no longer hung over the head of the family. The phasing out of internment meant that if the British wanted to put Adams back in jail, they would have to assemble a case that would pass the scrutiny of a court, even if it was one where only a single judge also acted as the jury.

Adams could be well satisfied with the fruits of his work in jail. Given the agenda devised in Cage 11, the IRA had been set on a course that would have lasting consequences for Anglo-Irish politics and for Adams himself. Reorganized to revive its flagging military fortunes, the IRA now had a political program to guide it, one that would ultimately propel it into electoral politics. It also had a structure that would inexorably and increasingly concentrate control and power in the hands of Gerry Adams and those he chose to have around him.

Adams did not emerge from Cage 11 and return to his IRA comrades empty-handed. As far as some Army Council members were concerned, the disastrous cease-fire of 1974–75 had been partly forced on the leadership by a weapons shortage in the Belfast Brigade area that was so severe that many doubted if the organization in the city could conduct anything approaching a sustained campaign. Within weeks of Adams’s release, however, a cache of sixty Armalite rifles, a batch smuggled by Irish-American sympathizers, was being distributed to grateful and astonished active-service units in the city. The sudden abundance of weaponry was credited to their released leader. To many activists, it seemed that Gerry Adams had not only the ideas to rescue the IRA from defeat but the means as well.

What none of them could have known was that the Armalites had arrived in Belfast well before the weapons famine hit the IRA in the city, before the resulting cease-fire, and had been secreted in five sealed dumps under the supervision of Brian Keenan and hidden even from Eamon Molloy’s prying eyes. The weapons had been put aside in case of an upsurge in loyalist violence, but when Adams was arrested and imprisoned along with many of his brigade staff, and eventually replaced by Billy McKee and his allies, no one told the new Belfast leadership that the weapons existed. Had they known about the guns, it is almost certain that the pressure to call the cease-fire would have been considerably less than it was. And if there had been no cease-fire, Gerry Adams would have been deprived of the cause around which to build his bid for control of the IRA.[38]

5. “Our Dreyfus”

Gerry Adams’s release from Long Kesh was eagerly anticipated by the IRA in Belfast and in the rest of Ireland. It was clear to even the most junior Volunteer that the organization had lost its sense of direction by the spring of 1977, that it had come out of the cease-fire in a battered and damaged state, and that its leadership was largely directionless and bereft of new ideas. Defeat stared the IRA bleakly in the face, and it was to Gerry Adams and the group around him that republicans now looked for salvation. The grassroots were aware of the divisions inside the jail, knew of the critique Adams had made of the Army Council’s policies, but they also knew that he had a reputation for strategic innovation, and although only a few were privy to the detail of his plans, most expected Adams to come out of jail with some sort of blueprint for regeneration.

He was supposed to go on a vacation south of the Border with his family immediately after his release from Cage 11, but within hours of being reunited with Colette, Gearoid, and the wider Adams clan in Ballymurphy, he changed his plans. Seamus Twomey, the IRA chief of staff, who was still in hiding in Dublin after his dramatic escape from Mountjoy jail, had sent a message summoning him to an urgent meeting. Adams had been able to communicate with Twomey and the Army Council from Long Kesh but only within limits. Written messages as well as the occasional typewritten document were smuggled out by visitors, and these had included a lengthy critique of the cease-fire and the state of the IRA that Adams had composed. His trip to Dublin after his release gave Twomey and Adams the first chance to discuss fully all the momentous events of the preceding four years and to bring Adams up to date on the implementation of the IRA reorganization plans hatched with Ivor Bell inside Long Kesh, not least the creation of Northern Command.

By the end of 1977, less than a year after his release from Long Kesh, Adams had become chief of staff in succession to Twomey, but not before another spell as Belfast commander. Debilitated by the 1975 cease-fire and a series of security successes by the British, the Belfast Brigade had corroded badly under McKee’s stewardship. Both the quality and quantity of IRA operations in the city had seriously deteriorated and when McKee was finally ousted, Gerry Adams was drafted in to reorganize and revive the organization.

Recapturing the Belfast Brigade from Billy McKee was an important milestone in Adams’s quest to dominate and mould the IRA nationally. In the late 1970s, as at the outset of the Troubles, Belfast was the engine for the IRA nationally. If the IRA in Belfast prospered so did the entire organization; if it faltered in the city, it faltered everywhere. And whoever dominated the IRA in Belfast, especially when it was doing well, inevitably carried huge clout in the national leadership.

Following his release from jail, Adams was armed with a powerful critique of the McKee leadership and had crafted a far-reaching plan to revive the IRA. But he also encountered stiff resistance to his agenda, even from those who had been part of his circle before his arrest. The explanation was very simple. The McKee leadership had taken pains to ensure that their supporters were in key positions throughout the Belfast Brigade and elsewhere in the IRA. No matter how convincing Adams’s case was against the cease-fire leadership, it made little impact on men who owed their positions and status to that leadership. Adams had to move with guile and caution to advance his agenda.

Assuming command of the Belfast Brigade brought Adams the breakthrough he needed. McKee’s allies in Belfast were removed and replaced with figures sympathetic to Adams and his reorganization plan. He also had the support of Seamus Twomey, the IRA’s chief of staff, who managed to get Adams on to the Army Council later in 1977. These developments paved the way for a decisive shift in the IRA’s direction that was cemented when Adams succeeded Twomey as the IRA’s military commander.

By 1977 two important allies from the North had joined the Council, and they would greatly assist the project. One was Martin McGuinness, the Derry IRA commander, whose IRA career had been launched when he was talent-spotted by Daithi O Conaill in 1971, and Brian Keenan, who by 1977 was IRA quartermaster-general, in charge of acquiring, hiding and distributing the IRA’s weapons.

McGuinness had spent much of the 1975 cease-fire either on the run across the Border or in Portlaoise prison, where he had been jailed after a conviction for IRA membership. Unlike Gerry Adams, he was not yet a national figure in the IRA but was known in Derry as a fearless and talented commander. Whereas Adams had a family history of association with the IRA, McGuinness’s parents were strong supporters of the old Nationalist Party and were typical of the vast majority of Catholics in a city never known for its adherence to the republican cause. Street violence, first involving the RUC and then the British army, had driven McGuinness into the IRA, initially the Goulding Officials, who had the greater support in the Derry of 1969–70. Unhappy at the Officials’ military timidity, McGuinness switched to the Provisionals just before internment removed the bulk of its activists. He suddenly found himself a general in a tiny army, but one that quickly mushroomed in numbers as conflict with the British intensified.

McGuinness had managed to stay out of the controversy caused by Cage 11’s campaign against the truce but had strongly supported Adams’s reorganization scheme and was rewarded with a place on the staff of the new Northern Command, first as operations officer (“double O,” in IRA parlance) and then as northern commander. Adams’s release and elevation to the Army Council meanwhile encouraged Keenan to abandon a long-standing reluctance to involve himself in the IRA leadership. Senior colleagues had often complained about Keenan’s repeated refusal to throw in his lot with them by joining either the Council or the IRA Executive, the twelve-person body whose main function was to select the Council’s members. But with Adams now on the Council, Keenan’s reservations dissolved. The effect of this crucial move was to strengthen Adams’s hand significantly.

The rest of the Army Council was a different matter. All the key figures were either obstacles to Adams’s ascent or ideological foes. Two in particular, Daithi O Conaill and Ruairi O Bradaigh, were major roadblocks. In 1977 the pair were still powers in the republican movement despite widespread criticism of their handling of the truce. Both men had enthusiastically backed the cessation and had played leading roles in bringing it about and nurturing it, and when it failed Cage 11 blamed them. O Conaill had acted as spokesman for the Army Council when senior Irish and English Protestant clerics had met the IRA leaders in an isolated hotel in Feakle, County Clare, in December 1974 to broker the cease-fire terms. His importance to the cease-fire was recognized by the British, who chose to send messages down the clerical pipeline to the IRA via O Conaill rather than anyone else. O Bradaigh was a member of a three-man Army Council delegation that met secretly with British officials in Derry throughout the cessation. His commitment to a successful resolution of the IRA’s campaign was more personal. When asked by the Feakle clergymen why he wanted a settlement, O Bradaigh had replied that the war could consume a second generation if it wasn’t brought to a halt.[1] There was no doubting the disappointment of both men when the cease-fire failed to secure the deal they had hoped for.

Notwithstanding that failure, O Conaill and O Bradaigh had an IRA record that ensured they had considerable support at grassroots level, particularly in the South, where they were well known and respected. If they had a weakness, it was that they were largely unknown quantities north of the Border, where the attentions of the British security forces meant it was too risky to circulate frequently among the rank and file. To most of those IRA activists who had joined since the early 1970s, they were distant figures, a weakness Adams would later exploit to his great advantage.

Soon Adams would have two close allies by his side in the Army Council. One was his old Second Battalion and Belfast Brigade comrade Ivor Bell; the other was Danny Morrison, the republican movement’s able public relations guru and editor of Republican News, who first handled publicity relations for Northern Command and then became the IRA’s overall director of publicity. According to one account of this period, Adams, McGuinness, and Bell tightened their control of the Army Council by taking advantage of a clause in the IRA constitution that allowed the Council to co-opt people to fill vacancies in its ranks as long as the Executive ratified the appointment later. But in practice, knowledgeable sources say, no ratification was ever sought, a snub to the Executive that exacerbated internal divisions.

As Adams and his allies consolidated their hold on the IRA leadership, Billy McKee emerged as the first target. McKee had become president of their Army Council during the 1975 cease-fire, a post separate from that of chief of staff, whose occupant normally fulfilled a plenipotentiary role on behalf of the leadership. Along with Ruairi O Bradaigh, he had led the secret Derry negotiations with British officials on behalf of the Army Council, and like O Bradaigh and O Conaill, he desperately wanted the cease-fire to work. A confirmed bachelor who lived with his elderly mother and later his sisters, McKee was a deeply religious man who attended Mass daily. A quiet talker, he was described by one of the Feakle clerics as looking “like a Baptist pastor or a Sunday school teacher.” The reason McKee gave the Protestant clergymen for wanting the cease-fire to succeed revealed much about his politics. “He was fed up with the fight because it was doing more harm to the Catholic Church than enough,” recalled one. “When [the war] started at first all the boys went to Mass, but now they weren’t going. He said, ‘I’m not fighting to destroy my church.’”[2]

Adams’s attack on McKee centered on his direction of the IRA during the cease-fire, in particular alleging that he had encouraged the IRA to fight a sectarian war and to feud with other republican groups. McKee, the Adams camp said, had fallen into a double British trap. The sectarian killings allowed Britain to say that the Troubles were a communal conflict, not an anticolonial war, while the shooting wars with the Officials just spread demoralization in the nationalist community, something the British welcomed. It was the latter that provided the reason for Billy McKee’s fall.

McKee’s approach to conflicts with Goulding’s Officials was simple, as an IRA activist of the day recalled: “His attitude was that ‘if any of my men are hit, I’ll hit back.’”[3] The 1975 feud had claimed eleven lives, and scores more were injured. This and the killings of Protestants had enraged the Cage 11 dissidents. “When I first met Adams, he was very angry over what had been happening outside while he was in jail,” remembered the same IRA source. “He was pointing out that we started the ’75 feud. Yobos in the Provos would get drunk, start fights, and away you go, and we are supposed to be surprised when it started up. In that feud a score or more ‘Sticks’ shot in one and a half hours, then retaliations. Meanwhile 150 Catholics are killed by loyalists, more retaliations but no British being killed. It was a total waste.”[4] The hostility toward McKee from the Cage 11 dissidents ran to such feverish levels that some even argued that they should arrange to get the Belfast commander shot dead.[5]

While Adams and those who thought like him undoubtedly saw McKee as a political and military liability, he was still a formidable opponent. When Adams was released from Long Kesh, McKee still had iconic status among the IRA grassroots, particularly in Belfast. McKee had two strikes in his favor. He had helped to defend St. Matthew’s Church in East Belfast against a loyalist mob, and his long hunger strike had, in IRA eyes at least, secured political status for IRA prisoners in the jails. As long as Billy McKee still held a leadership position in the IRA, Adams’s ambitions would be stalled.

The Revolutionary Council, the gathering of twenty to thirty IRA commanders and senior officers devised in Cage 11, was the instrument Adams used to purge McKee from the Army Council. McKee’s downfall began with yet another feud between the Officials and the Provisionals which broke out not long after Adams’s release from Cage 11.

AT THE EASTER commemorations in Belfast on April 10, 1977, a parcel bomb exploded just as the Official IRA parade was about to leave for the republican plot at Milltown cemetery from the assembly point at Beechmount Avenue in the mid–Falls Road area. A ten-year-old boy whose father was a member of the Officials’ political wing, the Republican Clubs, was killed, and several other Official IRA members and sympathizers were injured. The Officials assumed that the Provisionals had placed the bomb, and they set out to exact revenge. When the Officials’ parade eventually arrived at the gates of Milltown, the Provisionals were just leaving and the angry crowds clashed. Violent fistfights broke out, and shots were fired. Later that afternoon the dead boy’s uncle was shot dead by the Provisionals, and others were wounded. The feuding was stemmed when Catholic priests mediated a settlement, but the bad feelings simmered on for weeks even when it became clear that it had been loyalists, members of the UVF’s notorious Shankill Butchers gang, who had planted the Beechmount bomb, not the Provisionals. Nevertheless, at the end of July the Provisionals struck again, when gunmen killed a top-ranking Official IRA officer from North Belfast. Over the next four hours three more people were killed, two of them Provisionals and one a civilian mistaken for a member of the Officials. The last to die was Adams’s old friend Tommy “Toddler” Tolan, who along with Jim Bryson had helped make the Ballymurphy IRA unit such a formidable outfit back in the early 1970s.

His alleged failure to secure Army Council permission for striking against the Officials was the charge leveled against McKee at a Revolutionary Council meeting later that autumn. The accused IRA leader had no defense and appears to have been taken by surprise at the strength of the assault. According to one account McKee told the meeting that he could not remember the details of what had happened, and he was heavily criticized by one speaker after another, many of them members of the Adams camp. At a meeting of the Army Council held afterward, the censure continued, and he was out. Of the Council members present, only O Bradaigh spoke up for him. One version of what happened says that he resigned, another that he was asked to go.[6]

Whatever the truth, Billy McKee’s days as an IRA leader were over, and he was badly affected by the experience. “I saw him the next morning, and he was a shrunken man,” recalled a GHQ member who was at the Revolutionary Council meeting.[7] Shortly afterward he was admitted to hospital first in Belfast and then in Drogheda over the Border, and his illness was the official reason given to fellow republicans for his departure from the leadership.

Adams had chosen the right issue on which to confront McKee. Had he criticized McKee for allowing the IRA to kill Protestants, the outcome might have been very different. While important IRA leaders, Twomey in particular, were ready to accept that feuding with the Officials played into British hands, striking back against the loyalists was a different matter. The truth was that many IRA leaders, particularly those from Belfast, found little wrong with McKee’s uncompromising attitude to the loyalist gangs and had raised no objection when the retaliations were at their worst in 1975 and 1976. “Twomey was all about protecting Belfast from the Prods,” explained one contemporary.[8]

Adams’s criticism of the 1974–75 cease-fire was accompanied by promises that if he and his supporters had their way there would never be another cessation unless and until the British had committed to withdrawal, and it was this dual approach that appealed most to the Revolutionary Council. There had been a great deal of resentment within the republican grassroots at the way the 1974–75 cease-fire had been handled, and the Adams camp expertly exploited the unease. The IRA leadership had never spelled out the terms of the cease-fire or made public any of the promises allegedly made, and broken, by the British. Even the way the cease-fire had ended was never satisfactorily explained. Above all there was a suspicion that the Army Council had been tricked and manipulated by the British, who had used the breathing space afforded by the cease-fire to reorganize and refocus their drive against the IRA. The overwhelming sentiment after the truce among activists at all levels in the IRA was that never again must a cease-fire be called unless the IRA had the British on the rack. It was against this strident background and distaste for cease-fires that Adams and his supporters made their pitch.

The message that they delivered was that as far as he and his supporters were concerned, there would be no more cease-fires, no repeat of the disaster of 1974–75. Presenting a hard-line, militant face was a strategy that Adams was to use again and again to take his leadership colleagues and the IRA rank and file down paths they otherwise would have shunned.

The anti-cease-fire message was repeated in public as well as privately. Using the Brownie pen name he had adopted in 1975, Adams had started sending it out as early as May 1976, using a Republican News critique of the leaders who had declared the 1921 Truce to make a thinly veiled denunciation of those behind the 1975 cessation. An admission of IRA membership in one Brownie column led Sinn Fein spin doctors during the later peace process years—when Adams was emphatically denying any association at all with the IRA, past or present—to claim that others in Cage 11 shared the Brownie by-line with Adams and that the offending article had actually been penned by Richard McAuley, Adams’s aide and constant companion during the peace process. Cage 11 veterans insist, however, that Brownie’s work was the product of only one hand, while the Republican movement as a whole regarded the Brownie articles as carrying Adams’s imprimatur. “The weakness of the IRA of that period,” wrote Brownie,

was that instead of pursuing the war to its bitter end come what may, they allowed unscrupulous politicians and so-called “Peacemakers” to gain the upper hand. The result was the betrayal of the Fight for Freedom followed by a vicious and brutal Civil War and of course partition. It is to be hoped that the lesson of that period will not be lost on today’s leaders. There is only one time to talk of peace and that is when the war has been won not while it is raging. The time to talk of peace is when the British have left Ireland, otherwise they will find some excuse to remain.[9]

Within two years the message had become much more explicit while public condemnation of the 1975 leadership was barely concealed. In an interview with Vincent Browne, the editor of Magill magazine, a GHQ spokesman was blunt. Asked what attitude the IRA now had to talks with Britain of the sort that had happened during the truce, the spokesman replied, “We now regard such talks as entirely futile and the only time we will talk to the British again is when they come to us and ask our help to secure their immediate departure from Ireland.” Pressed on whether or not any consideration had been given to another cease-fire, the spokesman did not mince his words: “None. There is absolutely no question of another ceasefire or truce. In my opinion the last one went on far too long and it would be almost impossible for anybody to persuade the Volunteers that another one would be in the interests of the Movement or its objectives. Our aim now is to win the struggle on this occasion and we are prepared to make whatever sacrifices are necessary to achieve this.”[10] To drive the message home, Republican News reproduced the Magill interview in a special double-page center spread.

The American writer Kevin Kelley was given the same story in, if anything, stronger terms:

[T]he IRA today asserts that almost certainly it will not enter into a cease-fire agreement again, no matter what the bait might be. The Provos’ postmortem on the truce of 1975 is that, on balance, it proved to be seriously damaging to the movement, politically as well as militarily. As one volunteer observed…, “Even if the Brits reintroduced internment tomorrow and managed to pick up most of our guerrillas and all our weapons, we still wouldn’t ask for a cease-fire. The attitude would be, ‘Right, let’s get some new recruits and some more guns and keep fighting.’ Cease-fires are just not on.”[11]

It was hardly surprising that, presented with such uncompromising sentiments, the Revolutionary Council was so ready to give Adams his way and the Army Council so unwilling to oppose him.

The other reason was that the new leadership was quite simply delivering the goods. Although the IRA was never able to reproduce violence on the level and scale reached in the early 1970s, the years immediately after Adams joined the Army Council nevertheless saw a significant recovery in its fortunes, one which even the British were obliged to concede.

In 1977 Roy Mason had boasted of squeezing the IRA like a toothpaste tube, but eighteen months later, after the changes introduced by Adams had begun to take effect, his successor, Margaret Thatcher’s nominee Humphrey Atkins, was forced to admit to the British House of Commons that the situation had changed radically. In July 1979 he told MPs, “The first six months of this year have shown a marked rise in the level of terrorism and have demonstrated that we are up against a more professional enemy, organised on a system of self-contained, close-knit cells which make it difficult to gather information. Their weapons are more powerful and their operations have a different emphasis.”[12] Atkins was not saying that the IRA was able to present the sort of threat it had posed in the early 1970s, but his words were an acknowledgment that the defeat of the IRA was no longer within easy reach.

Toward the end of 1977 Adams’s grip on the IRA leadership strengthened. In November, Belgian customs officers, possibly acting on an intelligence tip-off from the police in Dublin, discovered six tons of Russian-and French-made automatic pistols, explosives, mortars, rockets, and rocket launchers and ammunition hidden in electrical transformers on board the MV Towerstream, which had docked at Antwerp after a voyage from Cyprus. The weapons had been smuggled from the Middle East and were being sent to a front company in Dublin established by a GHQ officer called Seamus McCollum. The Garda Special Branch put him under surveillance and on December 2 swooped on a house in Martello Terrace in the scenic Sandycove area of South Dublin, where McCollum was arrested. The operation was a singular success, but the detectives got two apparently unexpected bonuses. In the flat detectives found a draft of the cellular reorganization plan put together by Adams at the request of the Army Council, while outside they found Seamus Twomey sitting in a parked car. After a frantic car chase that ended outside Fianna Fail’s headquarters in central Dublin, Twomey was captured. Four years earlier he and other IRA men had made world headlines when a helicopter had swooped into the exercise yard of Mountjoy jail in Dublin and carried him off to freedom. It was one of the most dramatic jail breakouts in the IRA’s history. Now he was back behind bars.

The circumstances surrounding the capture of Twomey have long been a matter of conjecture and controversy within the IRA, not least because his departure paved the way for Adams to become chief of staff for the first and only time. Adams was the sixth chief of staff in the Provisional IRA’s history, and his tenure is distinguished by two features: his reign as military commander was the only one that can be precisely dated, and it was also the shortest. He took over immediately after Twomey’s arrest but lost the post seventy-eight days later, on February 18, 1978, when he was arrested by the RUC along with over twenty other republican suspects, as a wave of condemnation swept Ireland following one of the most horrific IRA incidents of the Troubles.

THE BOTCHED BOMBING of the La Mon House hotel on the southeastern outskirts of Belfast was one of the worst atrocities of the IRA’s campaign, its twelve uninvolved victims exceeding the death toll of Bloody Friday. The bombing was a political and public relations disaster for the IRA, tarnishing its attempts to present a new efficient military face to the world and once again marking the IRA with a sectarian stigma—all twelve of the dead were Protestants. The La Mon bombing also held up the implementation of the Adams military and political agenda by nearly a year, much to the dismay of his supporters and allies.

The dead, seven of them women, had been attending the annual dinner dance of the Irish Collie Club when a blast incendiary bomb hung on a window of the restaurant exploded, sending a huge fireball billowing through the room and incinerating everything and everyone in its path. The IRA later admitted that the warning it had phoned to the RUC was inadequate. The bombers could not find a public phone box nearby, and by the time they did, the bomb was just minutes away from detonation. There was simply not enough time to evacuate the building. The outrage caused by the size of the death toll and the horrible manner in which the victims met their end was intense and widespread, and orders were issued to arrest Adams. He was picked up by the RUC in West Belfast, questioned at Castlereagh holding center, and then held in Crumlin Road jail in Belfast on an IRA membership charge for the following seven months. But the case against him collapsed before it reached a full trial. It had been based on what proved to be flimsy evidence, principally clips from a BBC TV Panorama program featuring him making a Sinn Fein Ard Fheis speech in which he used words like “billet” and “war zone.” Much to the anger of the British, the North’s senior judge, Lord Chief Justice Sir Robert Lowry, threw out the case.

Adams was once more free, but he would never again hold the post of chief of staff. The disaster at La Mon had happened on his watch as chief of staff, and while he was clearly not responsible for the bungled warnings that caused the deadly inferno, the use of La Mon-style incendiary bombs had been approved by an IRA leadership of which he was then a crucial part. As soon as he was arrested, his place had been taken by Martin McGuinness, who occupied the position for the next four years while Ivor Bell replaced McGuinness as Northern commander; on his release Adams became McGuinness’s deputy, the IRA’s adjutant-general. The takeover of the IRA begun inside the cages of Long Kesh was complete and the Adams—McGuinness era had begun.

The months after Adams’s arrest were quiet as the IRA attempted to recover the swaths of political and propaganda ground lost after La Mon. Operations diminished in both number and scale, and the use of the deadly blast incendiary bomb virtually ceased. But in November 1978, two months after Adams’s release from jail and his assumption of the Northern Command post, the IRA offensive resumed. On the night of November 30, sixteen towns were bombed in the space of a one-hour period; altogether that month more than fifty bombs exploded across Northern Ireland, injuring nearly forty people. At the same time the Army Council authorized a new bombing campaign in England, and that Christmas police leave in London was canceled in expectation of a bombing blitz. By coincidence or otherwise, Adams’s release from prison signaled an upsurge in IRA violence.

Humphrey Atkins’s view that the post-1977 IRA was becoming a greater menace received its most powerful endorsement from one of the British army generals charged with combating the organization. An assessment of the IRA threat written in November 1978 by Brigadier James Glover, an intelligence specialist who later became the British army’s commander of land forces in Northern Ireland, fell into the IRA’s hands and was released by the IRA’s publicity department in May 1979. If Gerry Adams had been asked to write Glover’s report, he could hardly have done a better job. The brigadier’s central conclusion was an alarming one for the British government but music to the IRA’s ears. The Provisional IRA, he wrote, “has the dedication and the sinews of war” to maintain the then current levels of violence for the foreseeable future. He went on, “The Provisionals cannot attract the large number of active terrorists they had in 1972–73. But they no longer need them. PIRA’s organisation is now such that a small number of activists can maintain a disproportionate level of violence… though PIRA may be hard hit by Security Force attrition from time to time, they will probably continue to have the manpower they need to sustain violence….” He also paid a compliment to Adams’s skills in revamping the IRA: “[B]y reorganising on cellular lines PIRA has become less dependent on public support than in the past and is less vulnerable to penetration by informers.”

For years the official British propaganda view of the typical IRA Volunteer painted a picture of thugs motivated by subhuman criminality, but Glover recognized not only that this was rubbish but that the IRA itself was a great deal more sophisticated than was ever publicly admitted. “Our evidence of the calibre of rank-and-file terrorists does not support the view that they are mindless hooligans drawn from the unemployed and the unemployable. PIRA now trains and uses its members with some care…. They are constantly learning from mistakes and developing their expertise… there has been a marked trend towards attacks against the Security Forces and away from action which, by alienating public opinion, both within the Catholic community and outside the province is politically damaging.” Glover’s final judgment underwrote the entire Adams project: “The [republican] Movement will retain popular support sufficient to maintain secure bases in the traditional republican areas,” he wrote.[13]

Meanwhile Adams was using his post as adjutant-general to consolidate his standing in the eyes of the rank and file. His brief included enforcing discipline and reviewing IRA unit strength all over the country. The circumstances of the late 1970s were very different from those of 1973, when Adams was last on active service. Internment had been phased out, and while the authorities would dearly have loved to put him back in jail, they lacked the evidence to convict him in court, as Roy Mason had learned to his cost. Adams was now able to move around freely, and as he did so his influence outside Belfast grew, as a contemporary recalled:

Unlike any of his predecessors he was by this stage not on the run. He could go anywhere and spend days at a time reviewing units. He would arrive at a house, for instance, where there had been a death in the family and sympathize, telling them that he and Colette had prayed at Clonard for whoever it was, you know, showing a charming, personal touch. He made contact at a human level in a way his predecessors couldn’t, and that helped him to disseminate his message and win support.[14]

LORD LOUIS MOUNTBATTEN was arguably the best-known and possibly best-respected member of the British royal family. A naval hero, he had served with distinction during the Second World War, commanding a British destroyer, the HMS Kelly, which was torpedoed several times in an incident immortalized in Noel Coward’s 1942 film In Which We Serve. He went on to serve as Supreme Allied Commander in Southeast Asia between 1943 and 1946, after which he was named viceroy of India, in which capacity he oversaw the turbulent and violent handover of power to the Congress Party and the partition of India into separate Muslim and Hindu states. It was, however, his role as adviser and confidant of the British royal family that really marked him out as a significant figure. As cousin to Queen Elizabeth II and mentor to Prince Charles, the heir to the throne, Mountbatten was eagerly sought out for his advice and experience by a dynasty going through painful and unwelcome changes in the media-conscious, iconoclastic latter half of the twentieth century.

The seventy-nine-year-old Mountbatten had a soft spot for Ireland. Since the early 1970s he had vacationed for part of each summer at Mullaghmore on the Bay of Donegal between Bundoran and Sligo on the northwest coast. It was a dangerous spot for such a high-profile member of the British establishment to spend his vacations. Bundoran was a popular resort with Derry folk, and inevitably the summer crowds enjoying a vacation break would include IRA members and sympathizers who would be bound to hear of Mountbatten’s presence and might be tempted to strike a spectacular blow against the British royal family. But Mountbatten ignored the security advice to think twice about spending time there and continued to enjoy his fishing and boating expeditions off a section of the Irish coastline that everyone agreed was spectacularly beautiful.

His stubbornness cost him his life. On August 27, 1979, he died instantly when a fifty-pound radio-controlled bomb exploded on his thirty-foot pleasure craft, reducing the vessel to matchwood. Mountbatten died alongside his fourteen-year-old grandson, his daughter’s mother-in-law, and a fifteen-year-old boat boy from Enniskillen. It was clear that the IRA had known all about Mountbatten’s vacations for some time but had deliberately chosen this moment to move against him. The new British prime minister, Margaret Thatcher, had just taken office. She was a hardline opponent of Irish republicanism who was known to sympathize with the unionists and whose close friend and adviser on Northern Ireland matters, Airey Neave, had been killed by a bomb planted by the violent splinter group, the Irish National Liberation Army (INLA), during the 1979 British general election campaign. Mountbatten’s killing, as the IRA must have known, could only drive the new prime minister to adopt tough security measures of the sort that in the past had sustained and nourished the IRA.

The British establishment and media were only beginning to digest the enormity of this disaster when, later the same day, news of another catastrophe came in. On the shores of Carlingford Lough, a picturesque stretch of sea that divides County Down in the North from County Louth in the Republic, a two-truck convoy of British paratroopers was making its way from the soldiers’ base in Ballykinler to Newry when a huge bomb, hidden beneath hay on a trailer parked by the roadside, was detonated by a radio signal. The explosion devastated the convoy. Six soldiers died instantly, and when their shocked and disoriented colleagues came under sniper fire from across the bay, they sought refuge in the ruins of a gate lodge at a spot called Narrow Water. The frantic survivors radioed for assistance, and within twenty-five minutes a large Wessex helicopter carrying soldiers from the Queen’s Own Highlanders arrived and cautiously landed in a nearby field. No sooner had the reinforcements disembarked than another huge bomb, hidden near the gate lodge, was detonated. Another twelve soldiers were killed and a score or more seriously injured. Among the dead was the CO of the Queen’s Own Highlanders, Lieutenant Colonel David Blair, who was so close to the explosion that his body was literally vaporized by the force of the blast. It was a classic guerrilla ambush that drew the grudging admiration of the British, but there was no concealing the scale of the calamity. The death toll was the highest of the Troubles for the British army and represented the worst casualties suffered by the Paras since the Battle of Arnhem in 1944. In the annals of Anglo-Irish conflict it equaled or surpassed the famous Kilmichael attack of November 1920 in which the West Cork IRA commanded by the legendary guerrilla leader Tom Barry had ambushed a convoy of auxiliary police and, depending upon the version of the incident, killed seventeen or eighteen of their number.[15]

Between them the combined slaughter at Narrow Water and the assassination of Earl Mountbatten threatened to pitch Northern Ireland into the sort of security crisis relished by the IRA. There were rows between the RUC and the British army over who should lead the security battle against the IRA, and Mrs. Thatcher was forced to referee. She appointed Sir Maurice Oldfield, a former head of the British Secret Intelligence Service, MI6, as security coordinator in an effort to improve interservice intelligence operations. Despite the provocation, the incident did not, however, tempt Thatcher into the sort of precipitate response the IRA had hoped for. The IRA, under Adams as under any of its leaders, operated on the principle that the more Britain resorted to crude repression, the greater the degree of sympathy and support, passive and active, the organization could count on from nationalists. The classic example of that was the one-sided internment operation of 1971, which had boosted IRA ranks enormously. For years afterward the IRA lived in hope that the British would repeat that mistake. But Thatcher disappointed them.

THE REAL LONG-TERM significance of the events of August 27, 1979, was that they enormously fortified Adams’s status in the IRA and thus his control of and influence over the republican movement’s political direction. In its review of 1979, An Phoblacht – Republican News gave the credit for Mountbatten, Narrow Water, and other military successes to the changes pioneered by Adams: “Last year was one of resounding Republican success,” crowed the paper, “when the IRA’s cellular reorganisation was operationally vindicated, particularly through the devastating use of remote-control bombs.”[16] The IRA had been delivered from the disaster of 1974–75.

It has become an accepted part of the mythology of this period in the IRA’s development that Adams’s reorganization plan rescued the organization from certain defeat and was responsible for a dramatic turnaround in its fortunes between 1977 and 1979. The truth may be more complex. To begin with, the cell system was largely a Belfast phenomenon. Some rural areas successfully fought to maintain their old structures and the operational spontaneity and local control that came with them. The South Armagh Brigade, whose members had planned and carried out the Narrow Water attack, was perhaps the best example of this. Even in Belfast the reorganizing was far from complete or universal, as one former battalion commander recalled: “The cells in Belfast were never really divorced from the old company structures, because they [the companies] were needed for logistics, safe houses, call houses, and so on.”[17]

Those involved in fighting the war believed that a much more important factor in reviving the IRA after the 1975 cease-fire was the fact that the British had ended internment, and as a consequence scores of released IRA prisoners returned to active service. Internment was brought to an end officially in December 1975 as part of the cease-fire deal, and after this IRA suspects began to be processed through the courts and treated like ordinary criminals. As far as the Cage 11 dissidents were concerned, the criminalization policy that followed internment represented a major political setback, but according to another IRA commander it brought an unanticipated bonus for the organization:

Myself and [Billy] McKee analysed the political situation at the time [late 1974]. Internment had decimated the ranks of the IRA and in Ardoyne we had only four active Volunteers. I was never consulted about the possibility of a ceasefire and I near blew a gasket when I heard in prison that the Feakle talks had taken place. But I can’t understand these people who say that the truce wrecked us. In my view it strengthened us. We had a lot of internees coming back in for active service. It was so unlike the situation in 1974 when we had four active Volunteers. By the start of 1976 we were bursting at the seams.[18]

Nevertheless Adams received most of the credit for the IRA’s resurgence, and as his position strengthened he set his sights on removing the last obstacles in his way to unchallenged control of the republican movement. With Billy McKee ousted, the two most formidable remaining opposition figures were Daithi O Conaill and Ruairi O Bradaigh. The Belfast leader had once confided to a colleague on the Army Council the tactical approach he favored when going about the destruction and removal of political enemies, and it was this line of attack that he adopted to remove the Southern veterans: “ ‘You don’t confront people,’ he would say. ‘You isolate and marginalize them and then get rid of them.’ I often heard him say that,” the figure recalled.[19] It was to be a long, arduous, and at times painful campaign against O Conaill and O Bradaigh, but in the end it succeeded—the pair was isolated, marginalized, and then discarded.

Gerry Adams’s drive against O Conaill and O Bradaigh had actually started before the purge of McKee when, in July 1977, the Revolutionary Council was convened to expel a fellow West Belfast man, Gerry O’Hare, a former public relations man for the republican movement in Belfast who had risen to become editor in Dublin of An Phoblacht, the Provisionals’ Southern weekly paper. There were two reasons for the heave against O’Hare; one was his political friendship with O Conaill and O Bradaigh, and the other was the fact that as long as the editorship of An Phoblacht was in his hands, Adams would be unable to influence republicans south of the Border. The pretext for O’Hare’s removal was, however, a much less straightforward matter.

O’Hare’s wife, Rita, one of the first women to join the male-dominated IRA and a formidable activist in her own right, had been badly wounded during an ambush of British soldiers in Andersonstown in 1972 in which she had taken part. She was arrested and charged, but because of her severe injuries she was given bail. On the eve of her trial she absconded across the Border and took up residence in Dublin. Like many IRA relationships, the O’Hare marriage had been battered by the Troubles and weakened by separation and worry. Gerry had been interned in the North and jailed in the South while the authorities relentlessly pursued his wife in the Republic and attempted to have her extradited back to Belfast for trial. Under the stress and strain, the couple had begun to drift apart. In 1975 the marriage was dealt a devastating blow when Rita attempted to smuggle a stick of gelignite into Portlaoise prison during a visit to an IRA prisoner and was caught. The three-year jail term meted out by the Special Criminal Court in Dublin sounded the death knell for their relationship.

With the couple now married in name only and Rita confined to Portlaoise prison, sixty miles south-west of Dublin, Gerry O’Hare struck up a relationship with Grainne Caffrey, Daithi O Conaill’s sister-in-law and a first cousin of Ruairi O Bradaigh, and it was this that Adams and his allies seized upon to undermine the An Phoblacht editor. Internment and imprisonment had badly weakened family life in republican areas in places like Belfast, and one of the most divisive problems faced by IRA leaders was the constant allegation that prisoners’ wives were sleeping around, often with IRA men still at liberty. At a Revolutionary Council meeting in July 1977, O’Hare was accused of setting a bad example to IRA Volunteers because of his relationship with Grainne Caffrey, and he was dismissed from the editorship of An Phoblacht. “O Bradaigh defended him saying that but for the absence of divorce legislation they [Gerry and Rita O’Hare] would have regularized their situation, but Adams and Co. were pitching their appeal to Twomey and McKee, who were both very conservative Catholics,” recalled one delegate.[20]

The merciless ousting of O’Hare was the first move in Adams’s push to take control of the Provisionals’ public relations arm. O’Hare was replaced as editor of An Phoblacht by the Dubliner Deasun Breathnach, a member of a distinguished republican clan, but in reality control fell into the hands of Adams and his allies. While Breathnach became editor, power was transferred to an Adams appointee, Mick Timothy, who became manager of the weekly, and soon there were loud and bitter complaints about its content from the O Bradaigh–O Conaill wing.

The takeover of An Phoblacht was interrupted by Adams’s arrest and imprisonment after La Mon, but following his release in September 1978 the campaign resumed with vigor, and the Army Council authorized what was officially termed a fusion of An Phoblacht and Republican News. The new weekly, An Phoblacht–Republican News (AP-RN), was unveiled on January 27, 1979, and its first lead story announced that the purpose of the fusion was to “improve reporting of the war in the North.” But there was a hint of another objective, one that would pitch the organization into ideological turmoil. “We also intend to provide an improved and widened forum for Republican debate on building a new Ireland,” it declared.[21]

The fusion was a logical move from a number of viewpoints. It made economic sense to produce and distribute one rather than two weeklies, and the existence of separate Southern and Northern papers flew in the face of the republican objective of destroying partition. All these points were made in a paper prepared for the leadership by Danny Morrison, who was slated to be the new paper’s first editor. But the real significance of the merger became clear only after it had happened. It was at first not so much a merging of the two papers as a takeover of An Phoblacht by the Belfast paper, to the extent that the new weekly even looked and read like Republican News. “Effectively the Republican News people came down from Belfast and took it over,” recalls one spectator.[22] Even so the real control of the new paper would lie in the hands not of its editorial board, or even the Sinn Fein Ard Comhairle, or Executive, but with the IRA leadership, and as a result the paper would become a powerful vehicle in the effort to undermine the remaining influence of O Conaill, O Bradaigh, and their allies. Republican News had started life as the news sheet of the IRA’s Belfast Brigade, and APRN would be the creature of the Adams-dominated Army Council. Having captured the IRA Army Council, Adams was now determined to remove O Conaill and O Bradaigh’s influence from the last forum where their voice was still strong, in the leadership of Sinn Fein. The takeover of AP-RN was a vital first step in that drive.

THE FOUNDERS of the Provisional Republican movement had devised a policy program that they believed would satisfy the two unresolved issues from the 1921 Treaty settlement. Known as Eire Nua, or New Ireland, the program was designed to create political structures that, its architects believed, would calm Protestant fears that a united Ireland would mean their subjugation and eventual absorption by nationalist and Catholic Ireland, while its economic ideology was intended to correct the grave wrong wrought to those who had done most to give Ireland its freedom back in 1919–21. The great economic imbalance that gave the east coast, and especially Dublin, such a huge advantage over the rural west, southwest, and north-west, where much of the IRA’s 1919–21 campaign had been fought, had to be rectified, and the sacrifice made by lower middle classes, particularly the small farmers, who had provided the manpower for the independence struggle, would be recognized in the new order.

No republican policy was more personally identified with O Conaill and O Bradaigh than Eire Nua. It had been adopted as IRA policy in June 1972 when the Army Council endorsed it. Sinn Fein followed suit, and at that year’s Ard Fheis the party’s constitution was changed to encompass the plan. Eire Nua outlined a decentralized federal scheme that would consist of a central government drawn from a federal parliament, half of whose members would be elected nationally via a system of proportional representation and half drawn from four provincial parliaments that would have strong powers over economic policy. The provincial parliaments would be based on the four ancient provinces of Ireland—Munster, Leinster, Connaught, and Ulster—and underneath them would be two further structures, a series of regional development councils and a system of district councils.

The important layer was at the provincial level, and the keystone of the whole edifice was the Ulster parliament. Although based on the pre-1921 province of nine counties, and not on the six counties of Northern Ireland, the Ulster parliament, or Dail Uladh, was intended to safeguard Protestant rights in an independent Ireland and to sweeten the bitter pill of British withdrawal. “Dail Uladh would be representative of Catholic and Protestant, Orange and Green, Left and Right,” declared the Eire Nua document. “It would be an Ulster parliament for the Ulster people. The Unionist-oriented people of Ulster would have a working majority within the Province and would therefore have considerable control over their own affairs. That power would be the surest guarantee of their civil and religious liberties within a New Ireland.”[23] At the same time the inclusion of Counties Cavan, Monaghan, and Donegal, with their significant nationalist majorities, meant that the overall Protestant majority in Ulster would be a thin one and that compromise with Nationalists would be necessary to make the scheme workable.

Eire Nua held other attractions for the republican movement of the early 1970s. By creating strong provincial governments, Eire Nua intended to adjust the economic and political imbalance that had developed in the Republic as a result of the overdevelopment of the east coast and in particular the spectacular growth of the greater Dublin area since the 1960s. One result of this was that the west of Ireland and the southern and northern edges of the Southern state, including the Border counties, felt excluded and discriminated against. Eire Nua promised to change that. It was no accident that the bulk of Provisional supporters in the South came not from the east coast or from Dublin but from these poorer and more isolated fringes of the country.

A profile of the typical rural Provisional supporter of that time would show him or her to be a member of the small landowning and small business class, what one of their number called “peasant proprietors”[24] and an Adams supporter once scornfully dismissed as “Fianna Failers with guns.”[25] These were the people who stored weapons and explosives for the IRA’s Northern war, raised money, gave shelter to fugitives, and allowed their land and farms to be used as training grounds, meeting places, and bomb factories. Family ties to the losing side in the civil war motivated many, and few had shared in the benefits of Irish independence; above all they were overwhelmingly Catholic and conservative in their outlook. Eire Nua’s social and economic program appealed to all these elements. Based on the ancient philosophy of Comhar na gComharsan, Eire Nua decreed that the main instrument of economic policy would be the cooperative. A firmly neutral Irish state would control the finance sector and major industries; large ranchers would be dispossessed and their land broken up, and even though private enterprise would still play a role, it would be subservient to the cooperative principle. Non-nationals would be barred from owning a controlling interest in any Irish industry, while the strengthening of the Irish language and culture would be a priority in the new order. All this was, as the Eire Nua document boasted, a compromise between individualistic Western capitalism and the Soviet socialist system, a spot on the political spectrum that ideally suited the Provisionals’ Southern support base.

Both O Conaill and O Bradaigh strongly supported the Eire Nua policy and firmly believed that it was the only scheme that stood a chance of winning Northern Protestants to the idea of Irish unity and independence. But, for Gerry Adams and his allies, hostility to Eire Nua became the route by which they would undermine the O Conaill–O Bradaigh leadership. The assault on Eire Nua that followed took place on two fronts, one within the IRA and the other inside Sinn Fein, and in each case the tactics were markedly different.

Winning over the IRA was the easy part. With the organization now dominated by Northern Command, support for the Eire Nua policy was sapped by appealing to the most sectarian of sentiments—namely that the federalist scheme would leave Northern nationalists in the same subservient situation vis-à-vis the unionists as had existed before British withdrawal. The very reason for waging armed struggle would be questioned in the minds of many Northern and Belfast IRA activists if Eire Nua was implemented. It was, its critics claimed, a sop to loyalism.

Once again the Revolutionary Council was employed to win the argument for Adams. Now dominated by his allies, it rejected Eire Nua at a meeting in Donegal in July 1979 and shortly afterward the Army Council endorsed the decision. Eire Nua was no longer IRA policy, although it was still Sinn Fein’s program. The extraordinary situation now existed where the military and political wings of republicanism held diametrically opposed views on what shape Ireland should take after the British had been forced out of Northern Ireland. Within weeks the Army Council attempted to exert its authority over the Sinn Fein leadership on the issue, and that is when the trouble began, as a participant recalled:

One fine day we were sitting in an Ard Comhairle meeting when a certain gent appeared and announced he had a statement to read, that we had to listen and it was of vital importance that it shouldn’t go outside the room. There was to be no discussion, and after he had read what he came to read we were to move on to the next business. He had been asked to deliver a message saying that the Army Council no longer supported Eire Nua, and all documents and leaflets, stocked and on shelves [dealing with Eire Nua], were to be taken away and boxed.[26]

A fissure had opened up in the Provisionals that was unseen by the outside world but starkly visible to those at the top of the movement. While Sinn Fein continued to support the idea of Irish federalism, the IRA leadership and increasingly the organization’s weekly paper An Phoblacht – Republican News opposed it. Furthermore, O Conaill and O Bradaigh showed every sign of fighting to hold the precious ground of Eire Nua, playing on what they perceived to be Adams’s fear of a public and damaging split that could cost Northern Command much of its logistical support in the South. At the 1979 Ard Fheis, Sinn Fein delegates overwhelmingly passed a motion proposed by the County Kerry republican Richard Behal and seconded by Daithi O Conaill urging that Eire Nua be “retained, promoted and publicised.” The 1980 Ard Fheis the following November endorsed a similar motion. The two resolutions were as near to an open defiance of the IRA leadership as it was possible to get. That Easter the Army Council replied with a declaration that deliberately omitted any mention of federalism, saying instead that only a unitary state—“a 32 County Democratic Socialist Republic”—could bring unity and peace between Catholics and Protestants.[27] The two sides were at war, and occasionally, as at that Easter time, their skirmishes became publicly visible.

The warfare was fought on a number of fronts, political and organizational, but at one Sinn Fein Ard Fheis after another O Bradaigh and O Conaill were slowly but firmly sidelined. The determination of the Adams camp to destroy its enemy was absolute. To establish their militant credentials the Adams camp first persuaded Sinn Fein to back a policy of demanding “immediate” British withdrawal from Northern Ireland and the simultaneous disbandment and disarming of the mostly Protestant RUC and Ulster Defence Regiment. This was pie-in-the-sky stuff of course—there was little chance of republicans ever being able to enforce it—but that was not the purpose. The intention was to contrast Adams’s militancy and determination with the vacillation of O Bradaigh and O Conaill, whose own preference was for a phased and gradual British withdrawal so as to lessen the chances of a violent Protestant reaction, causing a civil war to break out. Meanwhile on the organizational level Adams and his allies pushed through measures that allowed for the co-option of their allies to Sinn Fein’s ruling executive. He also secured approval for the appointment of deputies to the party’s officers, a measure that soon brought charges that Adams’s allies were being placed strategically in places that would allow them to undermine the old guard.

The effect of all this was to gradually strengthen Adams’s grip on the Sinn Fein leadership. In 1977 he could count on the support of at most three other members of the party executive, but five years later he and his supporters had a total of ten out of the sixteen Ard Comhairle members elected by the Ard Fheis on his side and perhaps half or more of the twenty-one co-options.

It was, however, the turn to the left, charted and pioneered by Adams and his supporters, that sharpened the divisions almost to the breaking point. The Adams camp was picking at an ancient scab with this move. It was the extreme socialism of the Goulding leadership that had motivated many of the Southern republicans to side with the Provisionals in 1969 and 1970, and their views had not changed much in the intervening years. The reintroduction of socialist ideas by the Adams faction in the late 1970s deeply unsettled O Conaill, O Bradaigh, and their allies, but it also created a dilemma for them. Goulding’s leftward movement was accompanied by a scaling down of the IRA and military methods and was thus easier to denounce, whereas Adams presented his socialism as part of a revolutionary agenda of which an enhanced armed struggle was a vital and integral part. Opposing Adams’s socialism in these circumstances made O Conaill and O Bradaigh appear as if they were against the IRA at a time when many of those doing the fighting in the North identified fully with other revolutionary movements elsewhere in the world and saw the IRA’s struggle as fully consistent with them. As one of their number recalled, “[We were]… delighted to see the Khmer Rouge take over Phnom Penh, reveled in the liberation of Saigon, thought it fantastic when the Cubans chased the South Africans out of Angola, and identified with the Left in Europe, the ANC, and the Zimbabwe liberation struggle.”[29]

THE MOVE TO THE LEFT was first signaled not by Sinn Fein but by the IRA in bloody and dramatic fashion on the evening of February 2, 1977, when fifty-nine-year-old Jeffrey Agate, the English-born managing director of the giant multinational chemical company Du Pont, arrived home after a day’s work to find IRA gunmen waiting for him. He died instantly in a hail of bullets. Agate was the first businessman shot dead in the IRA’s new campaign of assassination aimed against the employer class. He died just before Adams’s release from Long Kesh. A month later, after Adams had joined the Army Council, the IRA killed its second businessman victim when forty-five-year-old English-born publicity consultant James Nicholson was shot dead as he made his way to Belfast airport following a one-day business trip to a struggling hi-fi factory on the edge of nationalist West Belfast. The killing of Agate and Nicholson and possibly as many as eight other locally based business figures in subsequent weeks and months 29 was justified by Chief of Staff Seamus Twomey. Employing the unfamiliar left-wing rhetoric of Cage 11, Twomey told a French TV interviewer, “All British industrialists are targets. They are exploiting the Irish working class… everyone directly connected with British imperialism are definite targets.”[30]

The move to the left was announced at same time as the “long war” doctrine was spelled out, at the 1977 Bodenstown commemoration. Declaring “We need to make a stand on economic issues and on the everyday struggles of people,” the Army Council spokesman Jimmy Drumm called for “the forging of strong links between the Republican Movement and the workers of Ireland and radical trade unionists.” The alliance, he predicted, “will ensure mass support for the continuing armed struggle in the North.”[31] Two years later it was Gerry Adams’s turn to give the Bodenstown address, and he amplified the message, this time not as a Belfast troublemaker fresh from the Long Kesh prison camp but with the authority of a former chief of staff and current Northern commander.

The move to the left became the backdoor way of attacking the Eire Nua policy. The target became not the federalism of Eire Nua itself but the economic and social program attached to it. To transform Eire Nua’s mild radicalism into a left-wing revolutionary program, Adams relied heavily on advice from outside the republican movement and in particular from a figure who quickly became, in the eyes of the O Conaill–O Bradaigh camp, the new version of Roy Johnston, Goulding’s éminence grise. Phil Shimeld was an English writer on the Trotskyist weekly Red Mole, the newspaper of the London-based International Marxist Group (IMG). He had made contact with IRA prisoners during the mid-1970s and they had invited him to Belfast and asked him to write for Republican News. Much to the irritation of the republican old guard, Shimeld supplied leftist-oriented articles, often under the nom de plume Peter Dowling, a tactic that enraged allies of O Conaill and O Bradaigh.[32]

The IMG was the British section of the Fourth Socialist International and traced its political roots to Leon Trotsky himself. Among the group’s leading lights was Tariq Ali, a left-wing celebrity of the 1970s who had hit the headlines when he helped organize mass demonstrations in London against American involvement in the Vietnam War. Although the IMG was hostile to the Soviet Union, as most Trotskyist movements were, the distinction was lost on the older republicans, who saw Shimeld as Adams’s version of Roy Johnston and Tony Coughlan. The allegedly baneful influence of Goulding’s pro-Communist aides had badly divided the IRA and Sinn Fein in the 1960s, and although they may have been unaware of these niceties of Marxist ideology, O Conaill and O Bradaigh suspected that history was about to repeat itself.

Following the IRA’s rejection of Eire Nua, Adams came under pressure from the old guard on the Ard Comhairle to come up with an alternative policy, which eventually he presented to his senior colleagues late in 1979. It became known as “the gray document” because a fault in a photocopy machine in Belfast had darkened the copies made for the rest of the Sinn Fein leadership. It was a slim document—at most two pages long—that still advocated the decentralization that characterized the Eire Nua document but strengthened central government at the expense of the provincial parliaments. Adams reserved his assault on Eire Nua for its economic and social program. The alternative he advocated was unequivocally socialist, and it appalled the conservatives in the Sinn Fein leadership.

Declaring that political control in a post-British Ireland would be worthless without control of the country’s wealth and economic resources, the Adams document continued:

Furthermore with James Connolly, we believe that the present system of society is based upon the robbery of the working class and that capitalist property cannot exist without the plundering of labour, we desire to see capitalism abolished and a democratic system of common or public ownership created in its stead. This democratic system, which is called socialism, will, we believe, come as a result of the continuous increase of power of the working class. Only by this means can we secure the abolition of destitution and all the misery, crime and immorality which flow from that unnecessary evil.[33]

This was strong meat for the rural Southerners, but worse was to come. The new program would also abolish the right to own land; under Adams’s plan the state would be the only entity allowed to possess the title to Irish land, and those who were working the country’s farms, no matter how small the holding, would enjoy only “custodial ownership.” That touched a nerve in rural Ireland, for among other things it meant that inheritance rights would be lost. If Adams had his way, the family farm, the mainstay of rural Ireland, would be no more. It was a formulation designed to strike fear and anger into the bulk of republicans outside Belfast and Derry, and it did.

The fight back came in October 1979 at a special Sinn Fein conference held in Athlone. That year’s Ard Fheis had been postponed and was not scheduled to meet until the following January, and so a stopgap weekend gathering involving two hundred or so of the faithful, mostly from the South, was held instead. Someone had decided to leak the story of the internal turmoil to the press, and a story was published in the Dublin tabloid the Sunday World under the byline of the left-wing journalist Eamonn McCann. The story claimed that federalism was to be ditched and that the Provisionals were about to move sharply to the left under the influence of Marxist sympathizers. The story, which was uncomfortably accurate, also talked about the waning influence of O Conaill and O Bradaigh. The reaction of many of the delegates, who included the party’s thirty councillors in the Republic, was an angry one. According to one report of the meeting, some would have staged a walkout had Adams not intervened during the Sunday morning session to deny the report.[34] The identity of the deep throat was never established despite an inquiry ordered by the Ard Comhairle, although the Adams camp let it be known that it suspected the ousted An Phoblacht editor, Gerry O’Hare.

The Sunday World story put Adams on the defensive. Within ten days both the Army Council and the Sinn Fein leadership had been forced to issue statements denying the story, claiming that it was the result of British and Irish government efforts to raise a Red scare against them.

THE MARXIST ALLEGATION persuaded Adams to give an interview to the author, then a junior reporter with the Dublin weekly magazine Hibernia, to underline in person the IRA and Sinn Fein denial. It was the author’s first meeting with Adams and took place in a council house in the West Belfast housing estate of Andersonstown over a large tray of tea, sandwiches, and cake provided by an obviously adoring middle-aged hostess. Dressed in sports jacket and light trousers, sporting fashionably long hair and a somewhat unkempt beard, Adams still puffed on a pipe in those days. Anyone unaware of his real identity could easily have mistaken him for a visiting lecturer at a New England women’s college or a left-wing sociology professor at a red-brick English university. He was charming and impressively attentive and, displaying even then the consideration of the veteran political operator, stayed long enough to clear away the tray and wash the dishes when the interview concluded. Our matronly hostess beamed with pleasure and adoration when finally he left.

If he had his tongue stuck in his cheek during our conversation when he dismissed the left-wing allegations, it did not show. “First of all there’s one thing which should be said categorically,” he declared. “There is no Marxist influence within Sinn Fein; it simply isn’t a Marxist organisation. I know of no-one in Sinn Fein who is a Marxist or who would be influenced by Marxism.”[35] The Hibernia interview was reproduced in full in An Phoblach–Republican News, and the following two issues carried more reassuring stories, one describing a pilgrimage Adams had made to the home of the decidedly non-Marxist hero of the 1916 Rising, Patrick Pearse, and another applauding the ultraconservative head of Irish Northern Aid in New York, seventy-two-year-old Michael Flannery.

Emboldened by all this, Adams’s opponents blocked the “gray document,” and in a compromise deal a special subcommittee was set up by the Ard Comhairle to marry the radical program with the original Eire Nua in time for the postponed 1979 Ard Fheis. While Adams had been forced to draw in his horns, the compromise document did contain concessions to his hard-line approach. For instance, while the original Eire Nua program envisaged that private enterprise would still have a role to play in the new Ireland, the compromise paper said it would have “no role to play” at all in key industries, and that small local businesses would be permitted only “provided no exploitation occurs.”[36] Adams also got included a proposal to wage an “economic resistance campaign” that would commit Sinn Fein to work with radical trade unionists. Custodial ownership of land survived but became the clumsier and somewhat contradictory concept of “family or co-operative custodial ownership.” Even so, many of the delegates regarded the compromise program as an attack on small farmers, and the veteran and respected Leitrim republican John Joe McGirl had to intervene to assure the Ard Fheis that this was not the same agenda as that espoused by Goulding. “Ten years ago I parted ways with people whose policies I disagreed with,” he declared. “My politics have not changed and I support this document.”[37] The document was jointly proposed by O Bradaigh and Adams and was passed, but by a margin that suggested serious divisions among the rank and file: 65 percent were for, 30 percent, against, and 5 percent undecided. “That was too close for comfort,” recalled O Bradaigh.[38]

The year 1980 ended on a positive note for the Adams camp. By November it shared control of the Sinn Fein ruling executive with the old guard, and the Northerners guerrilla tactics had forced one key opponent, Sean O Bradaigh, brother of Ruairi, to quit the leadership in protest against the radicalism of the compromise Eire Nua document. His shadow as SF publicity director, Danny Morrison, took over his job. Morrison became the confrontational, belligerent edge of the opposition. “Adams made the snowballs but Danny threw them,” commented one of his victims. The Adams grip on Sinn Fein had tightened immeasurably.

THERE THEN FOLLOWED one of those events that no one had anticipated or could have anticipated. For the previous four years republican prisoners had been staging a protest designed to preserve special-category or political status. Frustrated by the failure of the protest and angry at their treatment as common criminals, IRA and INLA prisoners in the Long Kesh jail, now renamed the Maze prison, finally decided to bring matters to a head. The protest had begun with inmates refusing to don prison uniform, insisting that they wear only their bed blanket instead, and then they had refused to slop out their cells every morning, instead smearing the cell walls with excreta and pouring urine under their doors. As Catholic Church intermediaries tried but failed to broker a settlement with the British and the violence of the prison staff against inmates intensified, the prisoners finally decided in the autumn of 1980 to embark on a hunger strike in a bid to win recognition that their imprisonment was politically motivated.

That protest, which ended in confusion and defeat a week before Christmas, and the second, fatal fast, which began three months later, had the effect of temporarily uniting the Provisionals and papering over the widening cracks. The anti-Adams camp called a halt to the undeclared warfare, as one of their number recalled: “Ruairi and Dave avoided stand-up rows with Adams in order to prevent a split, especially during and after the hunger strikes when what we had was a very united and very strong movement.”[39] They could do nothing else.

The feud was on hold, but the way the hunger strikes were run actually served to strengthen the Adams leadership, not least because the profile of the Northerners was heightened considerably, while that of Southerners like O Bradaigh and O Conaill was reduced. The campaign in support of the hunger strikers was largely directed from the North. Adams and colleagues like Danny Morrison were regarded very much as spokesmen for the protesting prisoners, and when republicans began winning elections, first as H Block prison candidates and then as Sinn Fein members, they were seen as the architects of that success, not O Bradaigh and O Conaill.

There was another, possibly more potent factor. As IRA coffins started to come out of the jail, the argument that federalism was a sop to the forces ultimately seen as responsible for the prison conflict became more and more appealing to grassroots activists. The pressure mounted on O Conaill and O Bradaigh. By the time the 1981 Ard Fheis met in the autumn, the two veterans had lost all control of the Sinn Fein leadership. A motion proposed by the Ard Comhairle and by the party organizations in Belfast and Dublin called for the Sinn Fein constitution to be changed so that the party’s aim was no longer a “federal” state but a “democratic socialist republic” instead. The motion, a head-on tilt at Eire Nua, was passed but just failed to win the two-thirds majority necessary to change Sinn Fein policy. The following year, by which time Adams and Martin McGuinness had been elected as abstaining members of a new Northern Assembly, the motion got the backing needed to alter the constitution. The O Conaill–O Bradaigh camp had lost the war with Adams. O Conaill quit as vice-president of Sinn Fein, a post he had shared uncomfortably with Adams, and O Bradaigh would step down from the party’s leadership a year later, in November 1983, and be replaced by the West Belfast man.

The old guard had fought on in the hope that Adams and his colleagues would grow weary of the battle, but they badly underestimated their reserves of patience. Adams quite simply outlasted them, and each time they promised to bring matters to a head—for instance, by threatening to resign en masse, which happened more than once—Adams would temporarily retreat, only to regroup for another debilitating round a few weeks or months later. It was the old guard who tired first, not Adams. They also forgot just how ruthless he and the other Northerners could be.

SIX MONTHS OR SO before the changes at the top of Sinn Fein (some date it at around April or May 1983), a tall, willowy, dark-haired woman in her midthirties boarded a flight at Dublin airport bound for Toronto, in the Canadian province of Ontario, and said farewell to Ireland forever. It would have been surprising had any of her fellow passengers recognized her or been aware of the part she had played in the drama unfolding inside Sinn Fein. Christin Elias was not a figure who had been much in the public eye despite her prominence in republican politics, but the extraordinary circumstances of her exile and Gerry Adams’s rise to the summit of Sinn Fein were inextricably linked, if well hidden and only rarely discussed, events.

Christin Elias was by common admission an enigmatic figure who appeared on the Provisionals’ stage in the mid-1970s without any obvious tie to Ireland or its political ferment. About the only thing that fellow Republicans knew about her was that she hailed from somewhere in Eastern Europe; some believed that she was Hungarian, others that she was Russian or Ukrainian. Another version was that her mother was Irish and her father a Lithuanian who had been a high-ranking official in the Communist Party of the Soviet Union. In fact she was born in Canada of a Canadian-Irish mother of Methodist stock and a Bulgarian father. She had been educated in Sofia and spoke Bulgarian fluently, a skill she learned from her émigré father. She and her mother traveled a lot and had ended up more or less by chance in Dublin in 1974, where, attracted by the conflict then raging in the North, she offered her home in Churchtown in South Dublin to the IRA as a “safe house” where meetings were held and fugitives took refuge. She worked in an engineering plant by day, but at night was an enthusiastic Sinn Fein worker; both she and her mother joined the party’s Cumann in Churchtown. She soon adopted the Irish version of her name—Christin ni Elias meant Christin, daughter of Elias—and moved gradually toward the leadership of Sinn Fein.

An enthusiastic supporter of Eire Nua’s federalist program, she befriended O Bradaigh, O Conaill, and other members of the Ard Comhairle and generally made herself indispensable. “She was a very clever, efficient lady, who ate work and was totally committed to Eire Nua,” remembered one of them. “She had studied behind the Iron Curtain and was the one who kept on pointing out the left-wing stuff of Adams’s people, predicting what would happen next.”[40] By 1978 she had risen to membership of the Ard Comhairle and was Sinn Fein’s national education officer, charged with, among other things, promoting Eire Nua among the rank and file. When the hunger strikes started in 1980, she almost single-handedly ran the national committee formed to publicize the prisoners’ cause. One activist remembers that she often met the bills for printing hunger strike posters and leaflets out of her own pocket.[41]

It was perhaps inevitable that such a strong-willed woman would clash with Gerry Adams, but when she did the violence of the collision startled everyone. One witness vividly recalled the occasion:

In the months after the [July 1979] Army Council message dismissing Eire Nua, every Ard Comhairle meeting was getting more and more painful, the divisions were intense. We wanted to know that if the policy wasn’t to be Eire Nua what was it to be? So at this particular Ard Comhairle meeting Adams pulled a document out of his jacket pocket, but it was only two pages long, typed on each side. Behal laughed, everybody tittered as he started to read from it. When he finished, we thought it was just an introduction, it was so short. O Conaill said,“Are you sure that’s all?” and O Bradaigh asked when we would see the documents behind it all.

Christin sat back, drew on her cigarette and said, “Well, Gerry, isn’t this a rather slim document to produce after two years of turmoil? This is not even a foundation, we’d have a job to build on that. Surely this is a rather threadbare document?” Adams lost the rag. “What do you expect,” he exploded. “I had only twenty-four hours to produce the document!” We were all sitting there with our mouths open. Christin then replied, “In that case, Gerry, it was a very poor presentation.” Adams hit the table with rage, shouting that she was putting words in his mouth. From that day on her cards were marked.[42]

Other witnesses agree that Elias had made a dangerous enemy that day. “You could see the spark in Adams’s eyes; he had been hurt by her,” said one.[43] “Boy was he mad, and all his lieutenants were mad! I pulled at her sleeve because I knew how vicious they could be,” remembered another.[44]

It was around the time of this famous confrontation that Christin ni Elias was approached by an official attached to the British embassy in Dublin who asked her out on a date. His interest was ostensibly romantic, and he asked if he could see her on a regular basis. It was an implausible story, and despite his assurances that he was not from MI5, she was suspicious and feared that he was an agent for one or other branch of British intelligence. She went straight to a senior member of the Sinn Fein leadership, who passed her on to the IRA. “They established an ongoing situation,” explained one source familiar with the story. “Belfast people in the GHQ Intelligence Department handled the operation, and they told her to meet the guy and they briefed her on what to say.”[45] The relationship began in August 1979, and the pair would meet about once a month, usually for lunch at venues in Dublin or Belfast: “The IRA instructed her to report back on the meetings, and when she met them she gave them a written report, all typed out and in triplicate, which was typical of her.”[46]

Much to Christin ni Elias’s frustration there seemed to be no end to the affair. The IRA wanted her to continue meeting the official, but she was, as she told one Sinn Fein friend, “totally sick of the arrangement, it went on so long.” After more than two years, by the end of 1981 there was still no sign of the IRA operation coming to an end. “She was always hoping that she would be ‘interrupted’, that the IRA would abduct him,” recalled the friend.[47]

In early 1982 the IRA attitude to Christin ni Elias suddenly turned hostile, and it was not long before the reason was a subject of republican gossip. “The word got out that she was a British agent, a whispering campaign was started, and attempts were made to stop her being elected to the Ard Comhairle,” said one source.

Eventually in July that year, it was announced at the Ard Comhairle that there would have to be a court of inquiry into her, that she was suspect. It went ahead, but the truth was that it was a foregone conclusion, that the Army had already judged her and found her guilty.

The biggest charge against her was that she had sabotaged one of their operations. The background was that she had been told by the IRA to have lunch with the official in a certain Belfast hotel on December 22 or 23, 1981, and she agreed. But what Christin didn’t realize was that as it was Christmastime the hotel was full of office parties, so when they turned up there were no free tables and they had to go elsewhere. The IRA arrived to find no British official in the restaurant, and they immediately accused her of sabotaging their operation.

I told her she had been lucky, that they were probably going to kill both of them and accuse her of being a spy and meeting her handler. She paled. After that they came and searched her house. They held her blindfold for three days, all of them men, and they even accompanied her when she went to the bathroom. They questioned her all the time, and I believe they were looking for the blacks [copies] of the intelligence reports she had typed up for the IRA.[48]

Ruairi O Bradaigh arrived in the midst of the affair and was held at gunpoint until the IRA took ni Elias away, eventually abandoning her several miles from Dublin in the middle of the countryside.

ni Elias was expelled from Sinn Fein in November 1982, at almost exactly the same time as the party’s annual conference finally ditched her beloved Eire Nua policy. Those who were her political friends had little doubt that the timing of the purge against her was intimately connected to the internal battle being waged over federalism. It seemed the only explanation for the IRA’s dragging out the intelligence operation so long and then deciding to move when they did. “The incident really tore the curtain of trust,” commented a friend of ni Elias. “We felt they used the incident to get at her and to scare off everyone else. She was too articulate, too politically skilled, too astute to be allowed to survive.”[49] Another commented, “The message was quite simple. If they could get at her, they could get at anyone. No one was immune.”

The atmosphere on the Ard Comhairle soured after the incident. Not long afterward the Ard Comhairle asked for a special meeting with the IRA to discuss Eire Nua, and the Army Council sent along two of its members, Chief of Staff Martin McGuinness and his deputy Ivor Bell. “The whole meeting was dominated by what had happened to Christin and their litany of complaints about the 1975 cease-fire,” recalled a participant. “The two Army men did all the talking. The meeting was full of bad vibes, the tenor was that if you criticized the opponents of Eire Nua and the O Bradaigh–O Conaill leadership you were anti-Army. Things started to get bad on a personal level after that. Before, everyone would go for a drink together in Conway’s pub after Ard Comhairle meetings, but after that they would go to Conway’s and we would go to Mooney’s up by Parnell Square.”[50]

The reality was that the Christin ni Elias affair had struck fear into Adams’s critics. The Ard Comhairle was told that the three members of the Sinn Fein court of inquiry into the allegations had unanimously found against her, but one of the tribunal, Kevin Agnew, a solicitor from Maghera, County Derry, wrote privately to a party official, saying that this was not true; he had voted for her, and as a lawyer he was horrified at what had happened. But he beseeched the official to burn the letter after it had been read in case it fell into the hands of the Adams camp. Her friends were appalled at what had happened, but no one wanted to stand up and publicly make an issue of it; they were too frightened. When the author, then Northern editor of the Irish Times, revealed in September 1982 that ni Elias had been dismissed from Sinn Fein on security grounds, there was a rush to deny it. The Ard Comhairle, which still included some of her closest political friends, called the article “a piece of scurrilous journalism.” Those who had been dismayed at the way the IRA had behaved chose to stay silent.[51] A few months later Christin ni Elias left quietly with her ailing mother for Toronto, where she has lived ever since. Aside from a few Christmas cards and letters she has had little or no contact with her friends in Ireland since the trauma of 1982. “She was our Dreyfus,” confessed one of them.[52]

6. A Long, Hot Summer

By the start of the 1980s, Adams’s struggle to oust the O Conaill–O Bradaigh leadership had been virtually won. The IRA Army Council was at this stage completely under his control and had rejected Eire Nua, and it would just be a matter of time before the last vestiges of the old leadership had been swept out of Sinn Fein. The blueprint drawn up and refined in Cage 11 of Long Kesh had been almost fully implemented, thanks in no small measure to the patience, skill, and determination of its architect.

The IRA was, however, no nearer its goal of expelling the British from Northern Ireland than it had been in 1969 when its war had begun. While it was true to say that the Adams plan had rescued the organization from virtually certain defeat, and an ignominious one at that, the IRA still lacked the strength and resources, both political and military, to make the British change their Northern Ireland policy in any significant way. The IRA’s violence had fallen off dramatically from the heady days of the early 1970s when victory over the British did truly seem possible, dropping from an average of some 160 killings a year to fewer than 80, a decline of exactly half. Military stalemate reigned, with nothing on the horizon to suggest that major change was possible. The IRA could not be beaten, that seemed certain, but neither could the British.

It was in these circumstances that Gerry Adams led the republican movement into one of the most extraordinary and complete political volte-face in its long history in a bid to break this stalemate, executing a move that involved adopting some of the very same tactics and ideas whose espousal by the O Conaill–O Bradaigh leadership had made them such objects of hatred and scorn. The shift took many forms, but none was as dramatic or unequivocal as the decision to give the fighting of elections the same priority as the waging of war against Britain.

Those who knew Adams well during these years say that the episode not only demonstrated the utter pragmatism that governed his approach to politics but revealed a determination that neither he nor his generation of republicans would suffer the same sad fate as his father and his contemporaries. When their IRA project ended, in the case of Adams’s father in the late 1940s, they had retreated to their hearths demoralized and directionless, bereft of community support, to dream of what might have been and sustained only by the hope that a new generation would one day rise up, take on the torch, and succeed where they had failed. Old comrades would meet on Friday nights to drink and sing stirring ballads of war and sacrifice, of battles fought and lost—but rarely won. His father helped found the Felons Club, a social and drinking club on the Falls Road where membership was open only to those who had been imprisoned for the cause. There, old IRA comrades could gather and reminisce and help each other out when sickness, bad luck, or death struck. In some ways it almost seemed as if defeat suited them better than victory, for there was a sense in which Irish republicanism thrived on oppression and the isolated exclusivity that came with it. This had been the story of one generation of IRA men and women after another, and it was the atmsphere within which Adams was born and reared. Unlike his father and his generation, however, Adams would not settle for such paltry spoils.

One Cage 11 veteran remembered Adams once putting all this into words. “There was no way he was going to fade back into the obscurity of Ballymurphy. He hammered it home to all of us that something had to come out of all of this during our lifetimes, that no matter about previous campaigns this struggle was not going to be for nothing.”[1]

For this to be possible, ideas and policies once regarded as sacrosant could and would be discarded when necessary. The “move to the left,” which Adams had launched to isolate the old guard in Sinn Fein, was eventually reversed, as were other policies that characterized and even defined the Provisionals under his leadership in the 1970s and much of the 1980s. The opposition to federalism, once thought of as a core Adams value, faded and was forgotten, and so too was the opposition to electoral politics, to the idea that Sinn Fein should run in elections and thereby give establishment politics even that small level of recognition. The Adams camp had fiercely resisted electoral politics when its feud with O Bradaigh and O Conaill was raging, but once they had been vanquished, that too was abandoned and reversed.

Running in elections was a tactic on which republicans had historically taken a pragmatic stand and past IRA leadership often justified the practice as a way of advancing the struggle against Britain when military methods were no longer viable. Frequently, electoral success was used to vindicate later violence. As long ago as the 1870s, the Irish Republican Brotherhood (IRB) had authorized its leaders to run in elections and take seats in the Westminster parliament in London as part of a strategy to obstruct the business of government in Britain. In December 1918 Sinn Fein’s victory in the general election in Ireland paved the way for the IRA’s war of independence. After partition and the split with de Valera, anti–Fianna Fail republicans regularly contested elections in both jurisdictions, although always on an abstentionist basis. Mostly they were beaten, but on two celebrated occasions they scored remarkable victories. In 1955 Sinn Fein ran for several Northern seats in the Westminster general election and won in two, MidUlster and Fermanagh–South Tyrone, while in 1957, a year after the IRA had launched its Border Campaign, the party contested the Republic’s general election and ran in nineteen constituencies. They were successful in four: Fergal O’Hanlon won in Monaghan, J. J. Rice in South Kerry, John Joe McGirl in Sligo-Leitrim, and Ruairi O Bradaigh in Longford-Westmeath. There was nothing in republican ideology that forbade the organization from fighting elections. The only iron rule was that successful candidates could never take their seats in what were regarded as illegal and corrupt parliaments. In practice, as well, republicans ran only occasionally, when there was a particular advantage to be gained. Otherwise, as a matter of doctrine, priority was given to the strategy of armed struggle.

That pragmatic stand changed at the time of the 1970 split with Goulding. Those who left to form the Provisionals had a long list of grievances with the Goulding leadership, but high on that list was suspicion about its parliamentary intentions. Goulding’s move to leftist, agitational politics and his gradual erosion of the abstentionist principle were seen as parallel tracks leading in the same direction, toward participation in elections and the recognition of the partition parliaments in Dublin and Belfast. “Elections” became a dirty word.

Not all those who went with the Provisionals thought that in such simplistic terms. O Bradaigh had been an abstentionist member of the Irish parliament and knew the propaganda and political value of electoral success. He was also aware that, no matter how successful the IRA’s military campaign in the North was, the absence of elected representation was a serious weakness that the British would be foolish not to exploit. O Conaill likewise developed a more mature attitude to elections with the passage of time.

By the third or fourth year of the war in the North, around 1972–73, republican leaders knew that they could count their support in the tens of thousands and that if this could be properly harnessed, they could produce a respectable republican vote. The commonly held view at this time was that the IRA had at most the support of 1 or 2 percent of the North’s nationalist community, but republican leaders knew better. They knew, for instance, that they could not sustain the war effort at the levels it had reached in those years without considerable grassroots support. The days when republican gatherings could attract at most a few score hardy veterans were long gone. The war in the North had radicalized people by the thousand. Some had joined the IRA, but many more were ready to help in other ways. IRA leaders knew from annual events like the Bodenstown commemoration, for example, that if thousands of people could turn out to publicly show their support for the IRA, then there were probably many more who would vote for Sinn Fein in the privacy and safety of the polling booth. The republican movement was like an iceberg; the visible tip was the IRA’s violent campaign, but the rest was hidden beneath the surface.

But opposition to contesting elections was fierce in the Northern IRA. O Bradaigh and O Conaill wanted to fight two elections in the North during 1973, but Northern IRA leaders stopped them. The first was for local councils, and that opportunity arose when the British abolished the oath of allegiance that unionists had made mandatory for such contests. The pair argued that Sinn Fein should field candidates but not take seats. It would be a protest against internment, they said, that involved no loss of principle. Likewise when elections were announced for the new power-sharing Assembly, they urged that these should also be fought on an abstentionist basis so that the SDLP, Northern Ireland’s moderate nationalist party, would not have a clear run. But powerful forces on the Army Council, led by Chief of Staff Seamus Twomey, vetoed both proposals. An ally of O Bradaigh remembered the strength of Twomey’s opposition: “In early 1973 at a joint meeting of the Army and Sinn Fein, Sean O Bradaigh and Twomey had a serious row. Sean felt that we were allowing the SDLP to make all the running. Twomey’s response to that was: ‘Fuck you! You never fired a shot in anger all your life.’ The SDLP were allowed to get off the ground as a result of Twomey and after that we missed a beat in publicity terms. It was very noticeable that the traffic in journalists to Kevin Street [Sinn Fein’s Dublin office] diminished in the summer of 1973.”[2] When polling for Northern Ireland’s power-sharing assembly took place on June 28, 1973, an IRA unit from Gerry Adams’s old Second Battalion fired an RPG-7 rocket at a polling station in the Lower Falls Road area. It was as if the IRA could not contain its contempt for electoral politics. From then on, ballot boxes had to be taken from polling stations by military guard so that the IRA could not destroy them.

The pro-election camp was again outvoted by the IRA leadership when the British announced elections in 1975 to a new body, a constitutional convention, which was to replace the power-sharing Assembly brought down by striking Loyalists the year before. Thereafter opposition to fighting elections became part of the general Cage 11 critique of those members of the 1975 Army Council highest on their target list, like O Bradaigh, O Conaill, and McKee. During the 1975 truce the British secretary of state, Merlyn Rees, had legalized Sinn Fein in the hope that it would encourage republicans to enter electoral politics. This failed because a majority on the Army Council suspected that the British were really trying to suck the movement into constitutionalism. But Adams and his allies seized on this and linked O Bradaigh and O Conaill’s flirtation with elections to the same alleged failing that had brought the IRA into a near-fatal cease-fire in the first place. Talk of fighting elections became the next worst thing to discussing surrender.

The next opportunity to fight an election came after Ireland’s accession to the European Economic Community. The first direct contest for the European parliament was due to occur in 1978 but was then postponed for a year. O Bradaigh and his allies had prepared the ground to fight the election on a quasiparticipatory basis. “We thought we would have won one seat,” recalled an O Bradaigh ally. “Neil Blaney [a former County Donegal – based Fianna Fail minister] had contacted us to offer to stand down in Connacht-Ulster if we had stood while O Bradaigh and O Conaill would have stood in Munster.”[3] O Bradaigh spent two years persuading Sinn Fein and the IRA to agree to a proposal to take seats in Europe on what was called “a controlled basis,” that is, limited to using the offices and facilities in the Strasbourg parliament for propaganda purposes.

Adams led the opposition to this proposal, arguing that this was the thin end of the wedge. If Sinn Fein took seats in the European parliament, then the way would be opened, no matter what the proponents said, to taking seats in the Irish parliament or at Westminster, he said. A defining republican principle would be breached. He turned to the Revolutionary Council as a way if undermining O Bradaigh, and on the eve of the November 1978 Ard Fheis, IRA commanders by a single vote came out against contesting the Euro election. Even so, the first subsequent meeting of the Army Council backed the idea, but then the Adams camp once more intervened and a second meeting was called. “The Army Council was got at and the vote went four to three against,” recalled a prominent activist. “One person, Joe Cahill, was got at on the grounds that the move would open the way to constitutionalism, to taking seats in Leinster House and Stormont.”[4] Adams also spoke against, and the motion was lost. “It was one of the turning points,” commented the same source. “We went back into the doldrums. It was like we had been knifed.”[5]

The Sinn Fein–IRA squabbling opened the way for the former student civil rights leader Bernadette Devlin to contest the Northern Ireland Euro constituency, and on an issue the Adams group considered the exclusive property of republicans. The motive for her decision was the deteriorating situation in the H Blocks of Long Kesh, where IRA prisoners were into their third year of a blanket and no-wash protest, heading inexorably, it seemed, toward a hunger strike. With the assistance of women activists from the Relatives Action Committees, a support network set up by families of the protesting prisoners, Bernadette, as she was simply known in republican circles, raised the issue of intervening in Europe in early 1978. Her intervention split the Provos. While the Sinn Fein leadership in Dublin backed her and even offered financial support, the Northerners, led by Adams, fiercely opposed her.

The Belfast Sinn Fein leadership and the IRA’s Northern Command joined forces to disrupt Devlin’s campaign, using her reluctance to give “unequivocal support” for the IRA’s armed struggle as a stick to beat her. Even though she was running to publicize the cause of the IRA prisoners, the Adams leadership mobilized the H Block protesters against her. A statement issued in their name called on republicans to boycott her campaign and declared, “We know only too well that the only way to remove the H Blocks forever is to boot both them and the people who build H Blocks out of the country at the point of a gun…”[6] Privately Devlin was accused of using the prisoners’ plight to advance her personal political ambitions, while on the ground leading figures campaigned against her. The IRA’s chief of staff, Martin McGuinness, followed her around Derry, urging people to boycott the election with the slogan “Back the prisoners, back the war!”

Despite the public and often vindictive opposition of Adams and his supporters, Bernadette, running under her married name, Devlin-McAliskey, won just shy of 34,000 votes in the June 1979 contest, a performance that far exceeded expectations and amounted to a serious rebuff for the Adams camp. The lesson was clear—if Bernadette Devlin could get that result in the face of republican antagonism, what could she have gotten had the IRA and Sinn Fein thrown their weight behind her? It was evident that there was untapped and not insubstantial electoral support for republican candidates. Her vote was dwarfed by the 140,000 votes won by the SDLP leader, John Hume, but her performance nevertheless indicated a far higher level of sympathy for republican-type issues than conventional wisdom had imagined possible.

Adams has never admitted that his opposition to Bernadette Devlin’s European campaign was mistaken or that it was the spur for his own turnaround on electoral politics. In his autobiography he claimed that it was “around 1978–79,” that is, before the Euro poll, that it was decided “simply and in principle that there would be a positive attitude towards an electoral strategy by Sinn Fein.”[7] If a “positive attitude” toward elections had been taken at that time, there is scant evidence of it; in fact the evidence is that the Adams camp actually hardened its opposition to electoralism at this time. At Sinn Fein’s November 1980 Ard Fheis for instance, held only five days after the first hunger strike had started, Adams’s allies forced through a motion barring the party leadership from even considering running in the following May’s Northern local elections. Two of his close supporters, the IRA and Sinn Fein director of publicity Danny Morrison and IRA Executive member Jimmy Drumm delivered the telling speeches. The result was that when the election was held, two weeks after the death of Bobby Sands in May 1981, Sinn Fein could field no candidates and other, smaller groups like the Irish Republican Socialist Party, the Marxist People’s Democracy, and the Irish Independence Party, a greenish breakaway from the SDLP, cashed in on republican anger.

By the time of the next Ard Fheis, in November 1981, the situation had changed. The deaths of ten hunger strikers and the political turmoil that accompanied them had changed everything. Bobby Sands’s victory in the Fermanagh–South Tyrone by-election in April 1981 and the success of H Block candidates in the Republic’s general election earlier that year cleared the way for Adams and his supporters to openly argue for electoral politics. The scale of their flip-flop was huge. While O Bradaigh and O Conaill had argued only for occasional electoral interventions, Adams and his allies pushed for a permanent strategy of fighting elections, a strategy that would coexist, albeit uncomfortably, with the IRA’s armed struggle.

Once again it was done by first giving assurances to the republican grassroots about the leadership’s total commitment to the tactic of IRA violence. Adams’s ally Danny Morrison told the assembled faithful that the lessons of history “show that the only effective campaign is that of armed struggle.” He went on, “[T]here is no one and no argument that can convince republicans that Britain, as she often asserts, cannot eventually be broken.”[8] Having genuflected toward the altar of armed struggle, Morrison was delegated to lead the charge in the opposite direction and in favor of electoral politics a few months later at Sinn Fein’s annual Ard Fheis, this time arguing that military force on its own could not do the job of forcing Britain out of Ireland. It was as if the Bodenstown speech had never been given. “Who here,” he asked the delegates in what became one of Irish republicanism’s most famous speeches, “really believes we can win the war through the ballot box? But will anyone here object if with a ballot paper in this hand and an Armalite in this hand, we take power in Ireland?”[9]

The so-called Armalite and ballot box strategy had been born, but the deep contradictions inherent in the approach, the built-in tendency for one to always inhibit and damage the other, would soon set Adams and others in his confidence on an expedition to seek a way out of armed struggle altogether. Adams was almost at that point in his political journey reached years before by Cathal Goulding and others who had led the republican movement in his youth, the point at which seeking votes and fighting a war would become mutually exclusive options.

The trek toward electoral politics had been a slow and cautious one, but with hindsight it appears to have been predetermined by the political program put together in the cages of Long Kesh. The key moment was when Adams won the argument in favor of the IRA’s involvement in political agitation. Once that was accepted and the IRA set out to capture and hold public support, the logical dynamic pushed republicans more and more toward contesting elections as the most effective way of measuring and demonstrating that support. And the more the IRA sought public approval, the more it had to examine how its military tactics and its violent actions contributed to or inhibited the task. The Cage 11 agenda was sold to the IRA on the basis of its republican purity and military ruthlessness, but the truth was that it imposed political restraints on the IRA which became tighter with the passage of time and with Sinn Fein’s deeper involvement in electoral politics. Eventually the contradictions would be too great to sustain.

WHEN ADAMS and his allies moved into the leadership of Sinn Fein in the late 1970s, the party they found was by and large unsophisticated and conservative. Movement into political activity was blocked by people who were instinctively antipolitical in outlook and who were content to allow a secret and unaccountable IRA leadership, the seven-man Army Council, to decide their politics for them. If the Adams agenda was going to succeed, that had to change; new and young blood would have to be brought into the party, if only to dilute the influence of the old guard.

Recruitment to the Provisionals had come in two great waves in the 1970s. There were first of all the Belfast Sixty-niners, as they were called, those who had joined the IRA in the wake of the August 1969 burnings and riots. Another surge of support followed the internment operation of August 1971 and Bloody Sunday nearly six months later. The next wave would come with the 1981 hunger strikes, but this time with a difference. The previous surges had all been into the ranks of the IRA; this time Sinn Fein would be the major beneficiary.

Even before the prison deaths Adams had been throwing Sinn Fein open to activists whose roots lay not in the IRA or even in Defenderism but in radical, left-wing politics. The first manifestation of this was the decision to introduce Sinn Fein to feminist politics, a ploy that enjoyed the extra advantage of infuriating the O Bradaigh and O Conaill camps.

Until the late 1970s the IRA and Sinn Fein had what would now be termed a traditional attitude toward women. Although individual women, often on the coattails of their husbands, did rise to leadership positions—the assassinated Maire Drumm, wife of Jimmy Drumm and at one stage a vice-president of Sinn Fein, being a prominent example—the republican movement was male dominated. Adams had encouraged the recruitment of women into the mainstream IRA and in the late 1970s he encouraged them to join Sinn Fein. Not least of their qualities was that many of them detested the O Bradaigh–O Conaill leadership more than he did. Sinn Fein then adopted a women’s program and, much to the horror of the old guard, set up the women’s department, which approved the use of contraception and began to address, gingerly at first, the vexed and divisive issue of abortion, eventually putting the Provos closer to a pro-abortion stance than any other Irish political party.[10]

Bringing women’s issues into republican politics became part of the heave against the O Conaill–O Bradaigh leadership. It was intended to enrage the conservative old guard, and this it succeeded in doing, as one of their allies angrily recalled: “Radical feminism was married to republicanism, and for a long time I couldn’t work out whether we were a national liberation movement, a radical feminist group, or republican revolutionaries!”[11] But the move also had the effect of broadening Sinn Fein’s appeal to a new layer of political activists who might otherwise have found the Provisionals’ single-issue program of armed struggle in the North suffocating. These were just the sort of people who would be most open to electoral politics.

Adams also introduced republicans to broad-front politics, the tactic of making alliances with other groups and individuals on single issues, the essence of which implied a willingness to compromise core beliefs. The principle behind the tactic was that different groups would come together on the basis that, even though they could not agree on everything, there would be one or two issues on which they could find common cause. The benefit was that it brought greater political strength; the downside was that each group was required to bury the issue dearest to it. In the IRA and Sinn Fein’s case this meant that their insistence that everyone else was obliged to sign up to the armed struggle had to go. It was another important stage in the dilution of the influence of IRA militarism, and ironically it was the prisoners’ struggle in the H Blocks of the Maze prison to be treated as politically motivated prisoners that became the principal vehicle for another key and defining political turnabout.

BY OCTOBER 1980 the IRA and INLA prisoners on protest in the H Blocks of the Maze prison had reached the end of their tether. They had been refusing to wear prison unforms for the best part of four years in protest against the British attempt to treat them like criminals, and for over two years, clad only in a thin blanket, the prisoners lived, ate, and slept in their own dirt and stench. From time to time they were transferred to clean parts of the jail and guards moved in with steam hoses to decontaminate their cells, but soon their new accommodation was as foul and putrid as the one they had just vacated. The moves were often accompanied by violence, meted out by prison staff acting as if they were under orders to break the men’s morale. There had been nothing to compare with it in Irish history, and participating in the protest required special qualities and conviction on the part of the prisoners. The protesting prisoners numbered some three hundred, only around a half of those serving sentences in the Maze for IRA offenses, and their numbers sometimes grew but more often shrank as the protest showed no signs of achieving success.

The IRA and Sinn Fein leadership had, with reluctance, been persuaded to mount a campaign of street marches and demonstrations to advertise their cause, but it was clear that key figures were afraid that the prisoners’ plight would not attract the public support needed. The prison protest had the potential to bring about a great defeat, and those around Adams knew it.

Senior Catholic Church figures, notably the Irish primate, Cardinal Tomas O Fiaich, had intervened in a bid to persuade the British to relent, with no success. He dealt directly with Margaret Thatcher, the British prime minister, but had found her totally inflexible on the issue. Although human rights lawyers had taken a case to the European Commission on Human Rights, its outcome held out only the slimmest possibility of a settlement. The IRA had stepped up its campaign on behalf of the prisoners, singling out prison guards for assassination. But nothing, it seemed, would move the British.

As the prisoners’ leaders surveyed the situation in the early fall of 1980, they realized the time had come to embark on the ultimate protest, the hunger strike to death. Their commander, Brendan Hughes, the old Cage 11 ally of Gerry Adams who was still serving out his sentence for the Malone Road arms finds, asked each wing on the protesting IRA Blocks to send in lists of volunteers prepared to go on the protest. Scores of names came back, and from the list Hughes chose himself and six others, five from the IRA and one from the smaller INLA, to go on the fast. They began refusing food on October 27, 1980.

There is little doubt that the Adams leadership was opposed to that hunger strike. Popular support for the prisoners was still too low, and it was clear that Mrs. Thatcher would not only resist the prisoners’ demands, the key one of which was that they be allowed to wear their own clothes, but might relish the chance to do the IRA serious damage by allowing the protesters to die. Should the fast end in deaths but no concessions, Thatcher would have won a great victory.

The timetable agreed by Hughes and his prison colleagues meant that at some point six of the hunger strikers would have to decide whether the seventh should die. It was highly unlikely that all seven would reach a crisis point at exactly the same time, and inevitably the weakest prisoner would die first. It soon became clear that the British had rejected the option of forcibly feeding the men, on the grounds that this tactic, redolent of British behavior during the 1919–21 conflict, would only anger opinion south of the Border. The men would either die or somehow be persuaded to relent. Just before Christmas 1980 the crisis point was reached when the Newryborn IRA man Sean McKenna started to slip in and out of a coma. It was certain that he had only hours to live. Just at this psychological crux, a British government document appeared, presented to Brendan Hughes via a Catholic Church intermediary. It offered what seemed to be a compromise on the prisoners’ five demands, and the protest was called off. But even on a superficial reading the document was full of imprecision. Hughes and his five semiconscious comrades knew that rejecting the document meant that Sean McKenna would die. There was really no choice.

The ambiguity in the document was apparent to all who read it, and most who did concluded that the protest had failed. Despite that, the Belfast Sinn Fein leaders attempted to present the document as a victory, but their efforts were less than convincing. A celebration march held in West Belfast attracted only a paltry crowd. Morale at grassroots level was low and confusion widespread, not helped by the refusal of Adams’s allies in Belfast to show the British document to party members, claiming that their one and only copy had been sent to Dublin.

The British document was the result of secret negotiations between the IRA in Belfast and Michael Oatley, an officer in the British Secret Intelligence Service, MI6. Code-named Mountain Climber, Oatley liaised with two figures, one the Derry-based businessman Brendan Duddy and the other a member of the Redemptorist Order, Father Meagher, who ferried his messages to and from the Belfast IRA men. With Adams and his allies opposed to the hunger strike and fearful that it would end calamitously, the circumstances behind the appearance of the British document raised dark questions elsewhere in the Provisionals, suggesting that the Belfast leaders would go to any length to end the protest.

“The deal was done behind the backs of the national leadership by Northern Command,” claimed a well-placed spectator of events in Dublin.

Just as Sean McKenna was approaching his end, I heard there had been a message from the Belfast people, and I hotfooted it round to Parnell Square [Sinn Fein’s headquarters], where Joe Cahill opened the door. Eventually I was ushered in, and there they all were, Christin ni Elias, Ruairi O Bradaigh, Dave O’Connell, and Piaras O Duill. They said they had received a message from Belfast which said the British had conceded four and a half demands and that the only thing missing was an acknowledgment of political status. This was the first they had heard of it. The deal had been done behind their backs and behind the backs of the National H Blocks Committee; they didn’t know a thing about it.

Bernadette McAliskey and Piaras O Duill were dispatched to Belfast to meet Adams to find out what was going on and specifically whether or not the British paper was as generous as had been claimed. They were kept waiting from 3 A.M. until 8 A.M. before Adams and Morrison agreed to meet them. “The message the pair from Dublin had for them was to get their ass across the Border. But Morrison’s reply was ‘come and get us.’ There were tears in the National H Blocks office that day. If we had won, why were we crying? It was a deception which fired the second hunger strike.”[12]

THE FAILURE of the first hunger strike meant that a second protest was not only inevitable but would this time end in either victory or death for the IRA prisoners. Once again the pressure for the fast came from the prisoners and specifically from Hughes’s successor as IRA jail commander, Bobby Sands, a twenty-seven-year-old coach builder who had been drawn into the IRA when he and his family were expelled from their home in North Belfast by loyalist mobs. Known by his IRA colleagues as Geronimo, more for his resemblance to the Apache chieftain than for his hot temper, Sands was serving a fourteen-year-term for possession of weapons when fate propelled him into the history books.

A self-educated poet and a songwriter of some distinction, Sands had been in Cage 11 with Gerry Adams and was, by all accounts, a keen disciple of the Adams gospel. He was bright and a brilliant publicist who excelled at writing letters soliciting support for the prisoners’ cause from prominent personalities. He was popular with his prison colleagues, trusted, and, as events were to prove, brave as well. The new IRA commander designed the second protest in such a way that death, his own in particular, was almost guaranteed. The protest would be staggered. Sands started by himself, a lone protester whose fate, the decision to live or die, lay in his own hands. He began refusing food on March 1, 1981, just two-and-a-half months after the first hunger strike had collapsed. After him other prisoners, usually in groups of two, joined at two-or three-week intervals. If any hunger striker died, the moral pressure on those who followed to continue through to the end was huge. The fast also guaranteed that if there were deaths, the North would be pitched into a crisis every fortnight or so until the end. As an instrument for destabilizing political life in Ireland, it was beyond historical comparison.

Ten hunger strikers died agonizing deaths in jail, and sixty-eight people were killed on the streets of the North, before it all ended eight turbulent months later. Sands’s death was probably the most traumatic. He slipped away in the early hours of May 5, and within minutes the darkened streets of nationalist Belfast were echoing to the crash and thump of exploding gasoline bombs tossed by rioters and the thud of plastic bullet guns being fired by police and troops. Sands’s funeral three days later was the largest political demonstration probably in living memory, as tens of thousands of nationalists poured in from all over Ireland to pay homage to the dead IRA man, by now rapidly assuming an iconic status. It was as if people had difficulty believing that Mrs. Thatcher could have allowed him to die.

Sands’s death divided Irish society. Constitutional nationalists in the North, like the SDLP, as well the main parties south of the Border feared the potential for instability posed by the hunger strike deaths, while the Catholic Church wrestled with the moral questions raised by a protest that some theologians insisted was a form of suicide. International interest in the hunger strikes was intense. Media from all over the world poured into Belfast that summer, and Sands’s photograph, a flattering portrait showing a good-looking, long-haired youth, became a symbol of revolution. In Teheran a street was named after him, as was one in New York. The Irish diaspora was radicalized. Demonstrations took place in New Zealand, Australia, and Canada and in many American cities. In New York, Noraid supporters mounted a picket outside the British consulate on Third Avenue that lasted for years.

The months from May to September that year were unusually balmy by Irish standards, but a sense of crisis and of impending doom permeated life as one coffin after another was carried out of the jail to funerals that, although they drew smaller crowds than had turned out for Sands, were as emotionally intense as ever. The impact of the hunger strike deaths was felt throughout the island. That summer the author regularly and repeatedly drove around Northern Ireland, in and around the Border counties and down to Dublin, in the Republic, covering the protest for the Irish Times, and there was scarcely a crossroads in a nationalist area in either jurisdiction that was not draped with mourning black flags or did not have rows of telephone poles adorned with protesting placards. During that long, hot summer of 1981 as nationalist Ireland, North and South, vented its anger at Mrs. Thatcher’s government in London, a new phrase entered the island’s political lexicon—“nationalist alienation.”

Again the evidence is that the Adams leadership, fearing a second defeat, which would be impossible to dress up in any but the thinnest of garments, also opposed the second hunger strike. But that was only at the beginning. As events unfolded, the Sinn Fein and IRA leadership realized that the political potential from the prison protest was immeasurable. It was a realization that would propel the organization into serious electoral politics and herald a phase in the Provisional’s department that would usher in the peace process.

To the surprise of many observers, the hunger strike deaths did not lead to the enormous upsurge in IRA violence that many observers had predicted. The widespread expectation was that the organization would cause mayhem when their imprisoned colleagues began to die, but although there was considerable street violence and some continuing IRA violence, including one devastating land mine attack by the South Armagh Brigade that killed five British soldiers, the organization resisted the temptation to throw everything into war. The relatively restrained response to the prison protest surprised even IRA activists. To those with long memories, however, the summer of 1981 carried echoes of the summer of 1970 in Ballymurphy when, at Adams’s urging, the IRA was held back while increasingly vicious but well-organized rioting between Catholic civilians and the British army alienated thousands of people, drove scores into the ranks of the IRA, and transformed this slice of West Belfast into the strongest republican redoubt in the city. Something similar happened throughout nationalist Ireland during the summer of 1981.

In October the hunger strikes came to an end as more and more of the prisoners’ families brought their sons off the protest, usually when they had slipped into comas and were unable to stop them. The end also came not long after the appointment of a new, more accommodating British secretary of state, Jim Prior, who announced concessions on prisoners’ clothes and other issues. As the prison settled, Sinn Fein began to prepare for elections. In an interview with the author shortly after the protest ended, Gerry Adams announced that he would recommend that Sinn Fein embrace electoral politics when the party met for its November Ard Fheis.[13]

Writing in An Phoblacht – Republican News in 1984, three years after Bobby Sands began his hunger strike, Adams revealed that the main effect the hunger strikes had was to fast-forward the move toward electoralism. It made it “easier to argue for an electoral strategy within Republican ranks,” he wrote, effectively admitting that this had been his ambition and goal for some time.[14] Before the hunger strikes, one of Adams’s leading advisers had confided to the author that resistance from rank-and-file IRA activists meant that persuading people to fight elections would be a slow and lengthy business. Sinn Fein might be permitted by its conservative membership and by the IRA to fight elections at the council level, but the prospect of larger-scale contests, for seats at Westminster, was probably some years down the line.[15] The election of Bobby Sands in the Fermanagh–South Tyrone by-election of April 1981 transformed that scenario and made it possible, much sooner that anyone had imagined, for Sinn Fein to fully embrace electoral politics.

The idea of making some sort of electoral intervention on Sands’s behalf appears to have originated with Bernadette Devlin-McAliskey. If Sands was to stand for the Westminster parliament, there was only one seat that he could conceivably win, and that was Fermanagh–South Tyrone, which had been held by the independent nationalist Frank Maguire since 1974. There were other nationalist seats but both, West Belfast and Derry, were held by Gerry Fitt and John Hume, respectively, bitter political enemies of the IRA who would do the prisoners no favors. Maguire, on the other hand, was effectively a republican MP, although his official designation was Independent. A former IRA commander in Fermanagh, he had been interned during the 1956–62 Border Campaign and took a more or less abstentionist attitude to the Westminster parliament. He never got around to making his maiden speech, for example, and he rarely attended the House of Commons, although his vote did keep a minority Labour government in office at key moments.

Devlin-McAliskey conceived the idea of asking Maguire to temporarily relinquish the seat to allow an H Blocks candidate to run. Confident that he would respond positively, she made plans to visit him, and, if he agreed to go along with the proposal, Sinn Fein would be presented with a fait accompli. The idea of “borrowing” the seat was born. Just before her trip and only five days after Sands had begun his fast, Maguire suddenly died and the whole situation changed. The seat was now vacant, and a race began to fill it. Maguire’s family and the local Catholic church wanted his brother Noel to run in the by-election, and McAliskey let it be known that if that happened she would run against him on the H Block ticket. Some Sinn Fein leaders, Ruairi O Bradaigh in particular, were keen to run her in the seat, but O Conaill suggested running Sands instead, and Bernadette Devlin-McAlisky agreed to stand aside. Sinn Fein leaders, virtually the entire Ard Comhairle along with the Fermanagh republican Owen Carron, met in Clones, County Monaghan, to discuss the idea but initially rejected it. Locals feared that because Sands was an outsider in this most parochial of areas, nationalist voters would stay at home. The convention ended, but no one would leave. “People hung about talking, having little discussions and were clearly dissatisfied with the outcome,” remembered one delegate. “The outcome was a second convention, and the proposal went through. Adams had to be convinced. O Conaill was the main mover. He was convinced Sands could win. He had been IRA organizer there in the 1950s, he knew the area, knew the people, he’d even been shot there. He convinced everybody.”[16]

The drama was not quite over. The SDLP’s Austin Currie, a local man and a fierce opponent of the Provisionals, toyed with the notion of standing against Sands, but he relented after enormous pressure was applied by his SDLP colleagues to dissuade him. The SDLP feared that it would be blamed if Sands was beaten and then died, and no one in the SDLP, no matter how much he or she despised the Provos, wanted to carry that burden. Noel Maguire also flirted with the idea of running, but amid dark hints of threats against him he withdrew at the last moment. As the deadline for nominations neared, Adams hid in a nearby Catholic church ready to withdraw Sands if another nationalist candidate decided to run. It wasn’t necessary. Sands had a clear run against a local unionist farmer, Harry West. In the charged atmosphere of the time, nationalists swallowed whatever reservations they had about the IRA’s violence and elected Sands.

Caution was again the overriding consideration in Adams’s mind when the next opportunity arose to put hunger strikers before the electorate, this time in the Republic’s general election of June 1981, a month or so after Sands’s death. After much bickering between rival republicans, the decision was made to run nine prisoners in selected constituencies, four of them hunger strikers. The National H Blocks Committee had been divided over the issue. Some wanted to run nonprisoner candidates who would take their seats in the Irish parliament, where it was possible they could hold the balance of power and determine who made up the next government. But Adams vetoed the idea. No one on the committee could give him a guarantee that they would be able to hold the next Irish government to ransom. “He wouldn’t take the risk of any dissension within Sinn Fein about abstentionism,” recalled a member of the committee.[17] It was a missed opportunity. Two prisoners won seats to the Dail, the hunger striker Kieran Doherty in Cavan-Monaghan and Paddy Agnew in Louth, who actually topped the poll. The result took Ireland’s media completely by surprise and threw Irish politics into turmoil. The broader election had produced a virtual dead heat between Fianna Fail and its opponents, Fine Gael and Irish Labour, who wanted to form a coalition government. The two seats won by IRA prisoners came at the expense of Fianna Fail; had Doherty and Agnew been able to take their Dail seats, or if nonprisoner candidates had run, the pro–hunger striker candidates would have held the balance of power and the history of the prison protest and much subsequent Anglo-Irish history might well have been very different.

While caution had dictated Adams’s attitude to the first two hunger strike elections, his demeanor changed after Sands’s death when a by-election to replace the dead prisoner loomed. His principal concern was to ensure that Sinn Fein held on to the prize of the Fermanagh–South Tyrone seat, but there were problems. Following Sands’s election the British had changed the electoral law to prevent serving prisoners from seeking election, so nominating another prisoner was out of the question. That was not all. Nationalist voters in Fermanagh–South Tyrone had been won over to support Sands with the promise that the hunger striker would merely “borrow” their votes in order to publicize the prisoners’ cause. Once the hunger strike was over, the seat could again be contested freely by mainstream nationalists. Under British parliamentary rules the writ for any by-election had to be moved by another serving MP, but no one at Westminster wanted to see another contest held while the prison fast was still under way. In June the Welsh nationalist MP Daffyd Ellis Thomas phoned the National H Block’s Committee to say that he would move the writ for the by-election whenever they wanted, but the committee demurred. Sinn Fein was not quite so backward. Behind the committee’s back, Thomas was contacted and he agreed to set the by-election machinery in motion. Sinn Fein then imposed Owen Carron as a “Proxy Political Prisoners” candidate, and in August he romped home, easily beating the local ex-UDR officer Ken Maginnis and actually increasing Sands’s majority.

Carron’s victory was the perfect springboard for Sinn Fein’s entry as a party into electoral politics, standing in its own name and not that of hunger-striking prisoners. Carron’s victory came, however, at a price. Throughout the hunger strikes the Army Council had refused to intervene and order the prisoners to quit, even when it became clear that, despite election victories and widespread international publicity for the prisoners’ cause, Thatcher would not bend and that any more deaths would be futile. The Army Council had that power and had used it in the past. The former chief of staff Sean MacStiofain had been ordered to end his fast in 1972, for example. As the hunger strikes intensified, so did the demand from some that the IRA leadership should intervene to end the deaths. In early August 1981, after the seventh hunger striker had been carried out of the jail in a coffin, Adams agreed to go into the jail to speak to the hunger strikers. He had come under enormous pressure to do this from prisoners’ families and their Catholic Church adviser, Father Denis Faul, and eventually he relented, or so it seemed.

According to the accounts of those who were present, Adams met six of the hunger strikers in the prison hospital canteen and gave them a frank assessment of their situation. He told them that it was likely Mrs. Thatcher would let them die, and he went on to describe what was on offer from the Mountain Climber, Michael Oatley. But he left it up to the prisoners themselves to decide whether or not to end the protest. The IRA would support and understand them, he said, if they agreed to end the hunger strike either individually or collectively. But it was evident that neither he nor the IRA would order or even ask them to call a halt.[18] With the weight of Bobby Sands’s sacrifice and the agonizing deaths of other comrades bearing down upon them, the prisoners unsurprisingly decided to continue their protest. Their decision meant that when the second Fermanagh–South Tyrone byelection was held two weeks later, prisoners were still dying in the H blocks, and with nationalist emotions running high, Owen Carron’s victory was assured. The SDLP once again stood down in his favor, and with a clear run against the sole unionist candidate he won by an even greater margin than Bobby Sands. When Adams addressed the next Sinn Fein Ard Fheis in the winter of 1981 to urge that they fight elections, he had in Carron a real live MP standing beside him as proof of the potency of fighting elections. Adams had spurned the opportunity to order an end to the prison fast, but by doing so created the most favorable circumstances for Sinn Fein’s change in political direction.

As the hunger strike headed toward its conclusion, Sinn Fein threw away any pretense that the National H Blocks Committee was an independent organization. At a conference in Dundalk in September, chaired by Adams’s Army Council colleague Ivor Bell, block voting removed all but two of its non–Sinn Fein members. By this stage of the protest the National H Blocks Committee had created a national structure of support groups throughout Ireland that rivaled and even exceeded Sinn Fein’s, and once the protest ended, a concerted effort was made to absorb these local committees into the party.

Then, with not even a whimper, the eight-month-long hunger strike ended. On the morning of Saturday, October 3, a brief statement announced that the most traumatic prison protest in IRA history was over. The end came as a relief. One hunger striker’s family after another had intervened to take their sons off the protest as unconsciousness overwhelmed them, and it was clear that further fasts were pointless and even counterproductive. The stage had been reached where the motives of Provisional leaders like Gerry Adams for not ending the protests were being openly questioned by clerics like Father Faul, once a sympathetic friend of the IRA prisoners but now a bitter opponent of the Sinn Fein leadership. Ten men had died awful deaths inside the jail, many more perished outside; nationalist politics had been radicalized in a way not seen since the 1916 Rising when the British had executed its leaders, while Mrs. Thatcher had become a hate figure beyond parallel in Irish history. The stage had also been set for Sinn Fein to emerge as an electoral force in Irish politics, at first in the North but eventually in the South too. Quietly, meekly, and with almost no notice taken by the media, the National H Blocks Committee voted itself out of existence, and scores of its supporters moved over to Sinn Fein. “The fact was that after the hunger strikes the republican movement was swamped by new young members,” explained one former Sinn Fein leader. “Some H Block committees just became SF branches overnight.”[19] The decision by Sinn Fein to contest elections and to embrace this as a strategy was a formality.

THROUGHOUT the early part of the next year, 1982, the new British secretary of state, Jim Prior, spent his time constructing a new political initiative that he termed “rolling devolution.” Under Prior’s scheme the unionist and nationalist parties would be rewarded by a gradual and increasing return of powers to a local elected Assembly at Stormont if they could show they were able and willing to share power. Despite objections from both camps and lukewarm support from Margaret Thatcher, Prior pressed ahead with his scheme and announced elections for that October. In April, Sinn Fein declared that it would definitely field candidates on an abstentionist basis, and that rattled the SDLP, worried by the alienating impact on their supporters of the lengthy hunger strikes. After weeks and months of agonizing, the SDLP then decided that, while it would contest the election, it too would not take seats in Prior’s Assembly. The hunger strikes had made both nationalist parties in the North abstentionists.

Adams had great difficulty persuading the IRA to allow him to stand for election to the planned new Assembly. At an Army Council meeting in July 1982, he and McGuinness indicated their interest in running, Adams in West Belfast and the chief of staff in Derry. But the Army Council voted four to three against, on the grounds that neither man could be an elected representative and be able to carry out his IRA role at the same time. But Adams came back. At a second meeting in August, Joe Cahill, who had voted against, was absent on vacation, and Adams put the proposal once more. In the meantime Adams had won over one key figure, a County Donegal–based member of the Council, and the vote went in Adams’s favor, four to two. McGuinness and Adams were, however, forced to make a concession to the skeptics. Both men were allowed to run but on condition they give up their IRA roles, McGuinness as chief of staff and Adams as adjutant-general. McGuinness’s departure paved the way for Ivor Bell to become chief of staff.[20] Adams’s ally Brian Keenan, by now a prisoner in England, wrote from his cell in Leicester jail that he “emphatically” supported the move and endorsed the Army Council’s decision. Keenan wrote, “It is not enough for Republicans to say, with reference to the Army, actions speak louder than words. We must never forsake action but the final war to win will be the savage war of peace. To those of us who have struggled for years in a purely military capacity, it must be obvious that if we do not provide honest, recognisable political leadership on the ground, we will lose that war for peace.”[21] Not for the first or last time, Keenan sided with Adams at a critical juncture.

To again reassure the IRA, that November’s Sinn Fein Ard Fheis obliged election candidates to “be unambivalent in support for the armed struggle.” But in the run-up to Prior’s Assembly election there was a subtle shift in the rhetoric. Again An Phoblacht – Republican News was the first to reflect the change. In a front-page editorial the paper declared that, while “Irish freedom will only be achieved by armed struggle… it needs to be said loudly and unequivocally that freedom, unity and the creation of conditions by which we can proceed to the democratic, socialist republic will not be achieved by armed struggle alone and that armed struggle of a revolutionary nature cannot even be sustained without popular logistical back-up and support.”[22] A few months earlier armed struggle alone had been the only way forward. Now that would not be enough. The contradictions between the Armalite and the ballot box were beginning to be acknowledged. The hunger strikes had wrought many deep changes in Irish society, but it would be some time before the effects of this would become visible.

PART THREE
A Secret Process


Father Alec Reid giving the last rites to a British soldier killed by an angry mob after he had stumbled into an IRA funeral procession in 1988. Reid’s secret diplomacy, in conjunction with Gerry Adams, laid the foundations for the Irish peace process. (Trevor McBride, Belfast)

7. “Behind the Scenes”

It was Friday, October 22, 1982, just two days after Sinn Fein’s first outing as an electoral party in the contest for seats to Jim Prior’s new “rolling” Assembly in Belfast, and the Irish media and political establishment were just beginning to digest the implications of a stunning Sinn Fein performance. The party had won the support of 10 percent of the electorate, some 64,000 votes, and had secured five seats to the new parliament, one won by Adams in West Belfast, another by McGuinness in his native Derry. As horrified constitutional Irish nationalists and stunned British politicians grappled with the meaning of all this, in the rolling countryside that surrounds the historic cathedral city of Armagh, not far from the town of Markethill, members of the South Armagh IRA drove up beside fifty-four-year-old Thomas Cochrane and with one swipe of a wooden club sent him flying off his motorbike. Known as Tommy to his friends and family, Cochrane was a part-time sergeant in the local unit of the Ulster Defence Regiment (UDR), the mostly Protestant local militia that had replaced the B Specials way back at the outset of the Troubles. He was also an Orangeman, a member of the Lurgyross Orange Heroes Lodge, which met near his home in Bessbrook.

Cochrane was on his way to work at Glennane linen mill that morning when the IRA struck. The police later found his motorbike in a hedge. His crash helmet, and gloves and the club used to fell him were found nearby. There were also signs of a violent struggle. Although the facts surrounding his fate have never been definitively determined, Cochrane was almost certainly overpowered and then bundled into a car and driven deep into South Armagh, an area the British media had long since christened Bandit Country.

In its first statement the IRA said it had abducted Cochrane and was holding him for “interrogation because of his crimes against the Nationalist community.”[1] This was an unusual thing for the IRA to do; normally the IRA killed UDR members whenever and wherever they could. Targeting members of this locally recruited, mostly Protestant, and overwhelmingly part-time force for death was an integral part of IRA strategy; such killings kept ordinary unionists angry and in no mind to seek a political settlement with their nationalist neighbors. In the unfortunate case of Tommy Cochrane it seemed that the IRA had decided to exploit a chance to gather some possibly valuable information about the security forces. Cochrane’s motorbike journeys along a regular and predictable route provided the IRA with an ideal opportunity to kidnap and question him about his colleagues in the UDR and any other intelligence in his head. If such was the case, then the terrors that Cochrane went through can only be imagined. Members of the South Armagh IRA had a name for ill treating and even torturing those they questioned, but the truth was that they mostly didn’t need to; their terrifying reputation loosened most tongues almost immediately. Tommy Cochrane could not even have guessed at that, but he did know with stomach-churning certainty that he was going to die; the only question was when.

The knowledge that Cochrane was suffering an almost indescribable ordeal reverberated around Northern Ireland. Most Catholics knew instinctively that there would be retaliations for his kidnapping, and the fear in that community was almost palpable, heightened by a communal memory of the scores of co-religionists kidnapped and killed by loyalist paramilitaries in the terrible mid-1970s. In all likelihood the loyalists would take an eye for eye, and soon some innocent, uninvolved Catholic would die a terrible death in revenge. That afternoon, Cochrane’s wife, Lilly, and their grown-up son, Glen, made an impassioned appeal for his release. The plea was ignored by the IRA in South Armagh, but it was heard in West Belfast, by a Redemptorist priest based in the congregation’s huge Clonard Monastery, which sprouted amid the tiny terraced streets that formed the heart of the Falls Road.

That afternoon Tipperary-born Father Alec Reid decided he would journey the short distance from Clonard to talk to the president of Sinn Fein and the newly elected Assembly member for West Belfast, Gerry Adams, to argue for Cochrane’s freedom. An active mediator during some of the worst years of the Troubles, Reid had employed his negotiating skills to help end vicious republican feuds and had often intervened between the worlds of the IRA and the loyalist paramilitaries. He would say himself that he operated mostly at street level and had built up a unique relationship with the Adams leadership. “Al’s power in the peace process,” confided one fellow participant, “derived from the trust that he had built up with Adams. I know he believed that no one else, certainly no other priest, had more close and continual contact with him than he had, and because of that he was able to sense the opportunity and the time to move.”[2] But Father Reid, then in his late forties, was not a well man. He had suffered a nervous breakdown two years before and only now was beginning to function normally again. His mission to Adams, a man whom everyone knew to be a major power in the IRA, was his first significant expedition since his illness. His plea was to get the IRA to release Tommy Cochrane or at the very least to tell his family what his fate had been. The Redemptorist priest knew enough about the IRA’s affairs to know that Adams carried sufficient clout to do that.

Protestants reacted with horror and anger to the news of Cochrane’s abduction—horror because they could imagine with awful clarity what the UDR man was going through, and anger because it was their old enemy the IRA that had kidnapped him. Like their Catholic neighbors, Protestants knew there would be retaliation. Around midnight on that Friday night one of the most cold-blooded loyalist killers ever spawned by the Irish Troubles briefed his gang members at a drinking club deep in the Shankill Road in the heart of Protestant North Belfast. They were going to kidnap a Catholic and hold him for Cochrane’s release, he told them. A handsome man with a fondness for fast women and the high life, thirty-year-old Lennie Murphy was the leader of a gang of loyalist killers that had struck terror into Catholic Belfast in the mid-1970s. Known as the Shankill Butchers, Murphy’s gang had earned its terrifying name because of a series of almost ritual torture-murders during a sixteen-month period between 1975 and 1977 when they dispatched a number of randomly kidnapped Catholic victims with axes, cleavers, and butcher’s knives. In one killing, a man was almost decapitated, in another all but three of the victim’s teeth had been ripped out with pliers; three died when their throats were slit open. Murphy himself was regarded by the police as “a psychopath”[3] and was suspected of involvement in up to eighteen murders. He had killed with gun, poison, and knife; he consigned four Catholics to eternity in one single incident and wielded a sharp cleaver during the Butchers’ throat-slitting killings.

It was just after midnight that Friday when Joe Donegan finally left the Pound Loney drinking club in the Lower Falls Road. An unemployed carpenter, forty-eight-year-old Donegan lived with his wife and their seven children in the Brittain’s Parade district of Ballymurphy, but he had been born and spent most of his life in the Lower Falls. Only when the violence got just too bad did he decide to move his family the two miles or so out of the area to the relative safety of Ballymurphy, but he still spent a lot of time in his old haunts, where he would meet friends for a drink and a game of pool. When that night’s drinking was done, he made his way by foot onto the main Falls Road, meaning to hail one of the many black London-style taxis that ferried people to and from the city center. The black taxis had been imported to Belfast in the early days of the Troubles when riots and barricade burning meant that the regular Corporation bus services were suspended, sometimes for days at a time. The black taxis were operated mostly by former IRA prisoners, and soon the loyalist Shankill Road copied the idea. The UVF ran the service in most Protestant districts of the city.

Joe Donegan must have realized almost immediately what a terrible mistake he had made when he clambered into the back of the taxi that answered his hail. Within moments Lennie Murphy was battering him so viciously that by the time the taxi arrived at its destination on the Shankill Road, its back windows were spattered with Donegan’s blood.[4] Like Tommy Cochrane, Joe Donegan must have guessed that he was about to die, and his end too must have been unimaginably awful. The taxi had been stolen on the Shankill by Murphy’s gang with the intention of duping the first vulnerable Catholic they encountered. The ploy worked with terrifying simplicity. By two in the morning Donegan was dead. Dragged semiconscious into a house once used by Murphy as his home, he was tortured and beaten repeatedly by the loyalist leader and his associates; he was finally finished off with a series of blows to the head with a garden spade. These blows were so violent that the wooden shaft snapped.

Joe Donegan was dead, but Murphy kept up the cruel pretense that he was still alive. Over the weekend the UVF killer made a series of calls to nationalist politicians and Catholic clerics, threatening to kill him if the IRA didn’t release Tommy Cochrane. The fate of the UDR man was still not known. One account suggests that the IRA held him for several days and that he may still have been alive when Murphy made his calls.[5] But the truth remains unknown. The pretense ended, however, on the Monday morning when Joe Donegan’s battered corpse was found, wrapped in a blanket and dumped in an alley off the Shankill Road. Later that night the IRA phoned the nationalist newspaper Irish News to say that it had killed Cochrane and that his body would be returned “when security allows.”[6]

Father Reid’s effort to get Gerry Adams to arrange Tommy Cochrane’s freedom had failed, but it seemed that the Sinn Fein leader had tried to intervene in some way. At least that is what he said he had done. On the Sunday, Adams released a press statement claiming that “since Friday evening, Sinn Fein have attempted to get clarification on the condition of UDR Sgt Cochrane and of the IRA intentions towards him.” But this had been made impossible, he went on, by the heavy security presence on either side of the Border. As soon as it was possible, Sinn Fein would urge the IRA to make a public statement. Later, after it had admitted that the UDR man had been killed, the IRA attempted to persuade nationalists that it would have spared Cochrane’s life if the offer to exchange him for Joe Donegan had been genuine. In an off-the-record briefing the IRA told the Irish News that its GHQ in Dublin had ordered South Armagh to release the UDR man but that the message “was not received in time.”[7]

The weeks after the deaths of Tom Cochrane and Joe Donegan were the most violent of the year as the full significance of Sinn Fein’s election victories sank in. Despite more than a decade of killing, shooting, and bombing, the IRA’s political wing had won significant support among the Catholic population, and for many, not least loyalist paramilitaries, that was a message of despair, a signal that violence could pay off. A further 39 people were killed between then and Christmas, bringing the death toll for the twelve months to 112. There were to be many more years of bloodshed before the IRA declared its first cease-fire, but never again would the numbers of deaths reach the level of 1982.

Republicans reveled in the election successes, and when three policemen were killed in a huge IRA land mine explosion near Lurgan, County Armagh, a few days after the Assembly elections, An Phoblacht–Republican News hailed the deaths as evidence that the Armalite and ballot box really did complement each other despite what the skeptics might say. The police deaths set off another round of retaliations, this time carried out by specially trained RUC units in which six people, five Republican activists and a Catholic civilian, were shot dead in circumstances that poisoned relations between the nationalist community and the RUC for years afterward.

On November 16, the IRA wreaked revenge for Joe Donegan’s death when Lennie Murphy was cut down in a hail of automatic IRA gunfire as he drove up to his girlfriend’s house. The carefully planned assassination had been arranged by a double agent in the rival Ulster Defence Association, a notorious racketeer called Jim Craig, who was working for his own side and the Provisional IRA at the same time.[8] Then, on December 7, came one of those incidents that will be remembered as a sort of gruesome high-water mark of the Troubles. That night as crowds of young people flirted and danced at a disco in the Droppin’ Well public house near Ballykelly, not far from Derry, a bomb placed near the dance floor by the Irish National Liberation Army exploded without warning. The blast killed seventeen people, eleven of whom were off-duty soldiers based in the garrison town. The remaining six dead were civilians, five of them women.

In between all this slaughter there were, by the standards of the Northern Ireland of that day, the routine killings of policemen and off-duty UDR soldiers by republicans and loyalist killings of uninvolved Catholics. On the same day that Lennie Murphy died, his UVF colleagues in East Belfast singled out a sixty-four-year-old Catholic shopkeeper for death. A lone gunman walked into Patrick Murphy’s greengrocers and opened fire with a handgun, fatally wounding the father of six three times. He was the fourth victim of violence that day. Four days later the UVF in East Belfast claimed another Catholic life when twenty-five-year-old Michael Fay was abducted as he made his way to the hospital to visit his young daughter. He was taken to a nearby loyalist housing estate and shot in the head. The UVF claimed he had been killed in retaliation for Lennie Murphy’s death.

The bloody violence of the autumn of 1982 was to leave its mark on Northern Ireland in unexpected ways. The killing of Patrick Murphy and Michael Fay had roused a group of liberal Protestant clergymen to action. Some forty of them, including three former Presbyterian moderators, decided to attend their funerals as a mark of solidarity. Their spokesman was another Presbyterian cleric, the Reverend Kenneth Newell, a former Orange chaplain turned ecumenist whose community in Fitzroy Avenue in South Belfast had the year before forged links with Father Reid’s Redemptorist colleagues in Clonard. Asked to explain what had motivated the group, Newell had answered, “We want to show that people still have the ability to scream out at murder.”[9]

The Protestant initiative was to be a short-lived one that fell victim to Northern Ireland’s unrelenting sectarian politics. But the trip to Gerry Adams made by Father Alec Reid was, by contrast, to have much more long-lasting consequences. A few years later the Redemptorist cleric was to tell friends that the kidnapping of Tommy Cochrane was the event that had persuaded him to play a more active and effective role in trying to end the violence. As a result, he would say, he decided to take up peace work again.[10] From that decision flowed the often highly secret and protracted dialogue and interaction with Gerry Adams that ultimately led to the Provisional IRA’s decision to quit armed struggle. In a very real sense the IRA’s decision to kidnap Tommy Cochrane was the moment at which the peace process could be said to have been born.

TO SAY THAT Father Alec Reid is the unrecognized inspiration of the Irish peace process would be an understatement. Long known as a confidant of the Sinn Fein leader, Reid is accorded in most accounts the role of message carrier for Adams during the odyssey to peace, but the full story reveals him to be a much more substantial figure, who initiated, devised, and nurtured many of the ideological innovations that made Gerry Adams’s journey possible. Passionate about his project, secretive, trustworthy, but at times gullible and naïve, Reid persisted at times when others in the British and Irish governments were close to giving up—his reward was the IRA cease-fires and the political settlement that eventually came more than a decade after his bid to save Tommy Cochrane’s life. To those who supported his efforts, he is the unsung and largely unrewarded hero of the peace process; to those who do not, he is the crafty Wolsey to Adams’s Henry VIII.

To the British civil servants who had secret dealings with Reid in later years, the Catholic priest seemed a curious choice for a go-between for Adams. “When I first met him, I thought he was a strange intermediary, with his black leather jacket and, if I’m right, I recall his nose was dripping,” remembered one.[11] For his work as a mediator, Reid dispensed with the standard Redemptorist garb—a black cotton gown circled by a thick leather belt, garlanded with heavy rosary beads—and dressed in jeans and bomber jacket like many of the people, especially republicans, he was dealing with.

But it was not this that made him such a valuable intermediary for Adams. His years of work intervening between squabbling republicans or trying to calm IRA-loyalist tensions had created a bond of trust between him and the West Belfast republican leader. Posted to Belfast in 1967, a year after Gusty Spence’s UVF had spilled the first blood of the Troubles, Reid emerged as a serious peacemaker at the time of the 1975 feud between the Provisional and Official IRAs and first got close to Adams in 1977 when again he mediated between the two groups during the bloody dispute that saw Billy McKee meet his nemesis. The relationship blossomed, and soon the word was out among Belfast Provos that they should regard Reid as Gerry Adams’s friend. “You could say anything you liked to the Provos, but if you criticized Alec Reid their backs went up,” recalled a member of the rival INLA in the city.[12]

Reid was also discreet to the point of reticence and avoided the media assiduously, both qualities that Adams valued. When, in early 1989, the author heard rumors of Reid’s part in secret peace talks and wrote to him asking for an interview, he replied, politely declining. “… I trust that what I have already mentioned about the relationship which every pastor must observe towards the media will explain why it would not be feasible for me to accept [your request for a meeting]. The pastor must be above suspicion in these matters and, for that reason, over the past twenty years or so, I have continually refused to appear on the media or to give interviews in circumstances that might give rise to such suspicion or the appearance of it.”[13]

After the IRA declared its 1994 cease-fire, Reid went into virtual hiding. When the author tried to contact him at Clonard Monastery, a secretary would say only that he was not in Belfast and that she did not know where he was. He had left a simple message on her typewriter that read, “Father Reid is not and will not be speaking to members of the press or media.”[14] Whatever clandestine comings and goings the Redemptorist priest got involved in on his behalf, Adams could be sure that Father Reid would keep them secret.

Illness had laid Reid low in 1980 at a crucial moment in the IRA’s prison protest. When delicate mediation between the Catholic Church and the Thatcher government broke down, Reid blamed himself for the hunger strike that followed, according to those who knew him then. He believed, they said, that he had wrongly raised the prisoners’ expectations of a settlement and that this had dangerously exacerbated their disappointment and anger, possibly propelling them into a death fast. Reid’s brief from his superiors in the Redemptorist Congregation allowed him to engage in peace and reconciliation work, and it was natural, given the gravity of the prison protest, that he would find himself visiting IRA prisoners, especially their leaders in the years 1979 and 1980.

He struck a sympathetic figure in the eyes of the republican protesters. Generally nationalist in his outlook, he rejected the IRA’s violent methods but believed it was important to understand why it was that people joined the IRA and supported the use of violence. He believed it was vital that he talk to republicans, and in those days very few Catholic priests did. The IRA prisoners liked and respected him. “He was seen as a gentle, hopeful, optimistic man, continually giving us comfort,” recalled one former protesting prisoner.[15] When asked by Sinn Fein before he started his own fast to the death which priests he trusted and was most friendly with, Bobby Sands gave Reid’s name as one of only three he could list. “Not much use that, is it?” he asked acidly in a smuggled communiqué to Danny Morrison just before he went on his hunger strike.[16]

The prisoners may have trusted Reid, but the authorities suspected him of ferrying messages to and from the IRA leadership. “The screws would never search [Father] Toner, who brought in tobacco and stuff, but on two occasions Reid was strip-searched,” said the same ex-prisoner.[17] Although his search did not entail the humiliating examination of the anal passage, either by hand or mirror, routinely meted out to IRA inmates, it was nevertheless an embarrassing and insulting experience. The priest bore his ordeal without protest.

The IRA prisoners respected Reid, but they were sometimes frustrated by his naïveté, a criticism the British would share later when his joint diplomacy with Gerry Adams began to get serious. “He told us once,” remembered the former prisoner,

that he’d been to the American consulate in Belfast to try to drum up support [for the H Block protesters]. I asked him who was there, and he said the ambassador, officials, and an English guy. I said who was the Englishman, and he said someone called Oldfield. “My God!” I said, “that was Maurice Oldfield of MI6.” It turned out that they had asked him a lot of questions about Brendan Hughes and others in the prison leadership. They were trying to build up a profile, and he was innocently helping them.[18]

All the priests who visited the jail or who got involved in the protest were given nicknames by the IRA prisoners. Father Denis Faul, who said Mass every Sunday in the prison, was initially admired by the prisoners and his nickname, Denis the Menace, indicated a certain endearment that turned to bitter hostility when he opposed the hunger strike and the Adams leadership in late 1981. Father Tom Toner, the official Catholic chaplain, was Index, after the F(inancial) T(imes) Index. Cardinal O Fiaich was Saggart Mor, literally “big priest” in Irish. Father Alec Reid became known as Behind the Scenes. “He was always telling us not to worry, things were going on behind the scenes, and it would all turn out okay,” recalled the same IRA inmate.[19] He said it so often that the prisoners, half cynically, renamed him.

Reid put enormous faith in secret talks that had been opened with Thatcher by Cardinal O Fiaich and the bishop of Derry, Edward Daly. The threat of a hunger strike in March 1980 prompted the dialogue in which O Fiaich and Daly argued with the British that a small concession on scrapping the prison uniform, just one of the IRA’s five demands, could avert a great disaster. Of all people in Ireland, the republican-leaning Catholic primate and his colleague from Derry knew just how destabilizing and emotional an IRA hunger strike could be. In late October 1980 the crisis came to a head. The prisoners announced their plans for a hunger strike, and O Fiaich and Daly frantically pressed the British for the concession on prison garb. The British responded with a piece of smoke and mirrors that greatly embittered the clerics. The prison uniform would be scrapped, Margaret Thatcher’s Northern Ireland secretary, Humphrey Atkins, told them, and be replaced by “civilian-style clothes.” It did not take much probing to discover that “civilian-style clothes” were just the old prison uniform renamed and redesigned. Disappointed and disillusioned, the clerics withdrew. Four days later the first hunger strike began.

The prisoners remember Reid as being extremely optimistic about O Fiaich’s efforts. The cardinal had visited the H Blocks in July 1978 and was so upset by the sight of prisoners living in their own excreta that he issued an angry statement comparing the conditions in the H Blocks to the sewers of Calcutta. The prisoners now had a powerful friend, and after the Calcutta remarks Reid assured the prisoners that O Fiaich would see Thatcher and “could sort it out.” So it was that the failure of the O Fiaich–Daly mission hit him hard. He fell ill and suffered a nervous breakdown. Disappointed, guilty, and fearful, Reid decided to quit his prison diplomacy, but before he left he made one last attempt to stop the protest, as the prisoner’s leader, Brendan Hughes, recalled: “When the hunger strike was planned, he begged me not to start it, and when I said no, he asked to speak to someone else to try to persuade them. So I sent Sands to see him, and Bobby was as strong as I was.”[20]

IT WOULD BE IMPOSSIBLE to understand the relationship that was built up between Gerry Adams and Father Alec Reid without first appreciating the extraordinary influence that the Redemptorist Congregation exercised on the Catholic population of West Belfast during the Sinn Fein president’s childhood. When Reid visited Adams to plead for Tommy Cochrane’s life in October 1982, he came not as just another Catholic cleric but as a member of a community that for nearly a hundred years had shaped and molded the views and attitudes of Adams’s family and those of the rest of his community.

Founded in 1732 by Alphonsus Liguori, the lawyer son of wealthy aristocrats from Naples, the Redemptorists, whose official title is the Congregation of the Most Holy Redeemer, are the Jesuits for the poor and working classes. Bound by vows of poverty, chastity, and obedience and steeped in the Irish Marian tradition, they are renowned for a “hellfire and damnation” style of preaching that is designed to strike fear of death and an eternity spent in the fiery pits of hell into the hearts of waverers. They are often invited to conduct intensive missions in parishes where the church authorities believe the congregation has become lax. “We provide a 64,000-mile service for the faithful,” a member of the order once told the author.[21]

The Redemptorists set up their first base in Ireland in 1853 but did not come to Belfast until 1896, when a surge in the city’s Catholic population, the result of rapid industrial growth and an influx of linen workers from rural areas set alarm bells ringing amongst the Irish hierarchy. Concerned that his burgeoning flock, numbering some 100,000 by the turn of the century, would fall prey to the droves of assiduous Protestant evangelists who roamed Belfast in those days, the then bishop of Down and Connor, Dr. Henry Henry, asked the Redemptorists to set up shop, and they readily agreed. They chose to establish their base in the area around Clonard off the Falls Road in West Belfast, a crowded district of small terraced streets where many of the migrant Catholic linen workers had been housed.

The growth in Belfast’s Catholic population drew an increasingly hostile Protestant reaction, and from the mid-nineteenth century onward, animosity to and intolerance of the newcomers grew. A series of vicious citywide riots gradually imposed a sectarian and political segregation on the map of Belfast. Protestants in the adjacent Shankill Road area particularly resented the arrival of the Redemptorists in Clonard. During the construction of the first church building, there were threats of violence. “This threat became very real in the winter of 1899,” the congregation’s official history recorded, “and for several days and nights, local men patrolled the area to defend the Monastery should it be attacked.”[22] These tensions were replayed seventy years later during the violent August of 1969, when loyalist mobs invaded the Clonard area and burned down Bombay Street, only yards from the monastery buildings. It was said their real target was the monastery church, from whose tower, according to Protestant mythology, IRA snipers could fire down on the Shankill. No one knew or understood, and perhaps even sympathized with, the Defenderist roots of the Provos better than the Redemptorists of Clonard.

The West Belfast that Gerry Adams grew up in was steeped in the influence of the Redemptorists. The most direct influence on the future IRA leader, his friends, and family was exerted through the confraternities that the congregation had run for many years. There were three “confos,” as Adams and his contemporaries called them, one for “ladies,” one for men, and another for boys. A society of lay Catholics, usually led by a Redemptorist priest, the confos would meet separately every week to say prayers, sing hymns, and hear a stirring sermon on Catholic doctrine from one of the Clonard priests. The confraternities were the Redemptorists’ way of ensuring that the faithful stayed faithful. The societies had flourished in West Belfast. By 1957, when Adams was old enough to join, the male confraternities had hit record numbers; nearly 9,300 West Belfast males were on the books that year.

Adams himself has nothing but fond memories of his days as a confrater. “One night a week,” he wrote in his autobiography,

we used to go for our religious instruction to confraternity down in Clonard Monastery and if we left early we could spend the bus fare on sweets. We cut down the Springfield Road and joined hundreds of other boys in the chapel. To me Clonard was a wondrous place with high, high ceilings and a huge high altar. The altar boys wore long, red soutanes and white gowns. The priest’s incense spiralled upwards through the shafts of sunlight which came slanting down from stained-glass windows at the very top. It wasn’t a long service. Father McLaughlin, who was in charge of the confraternity, got up and made a joke or preached a sermon and then we sang a few hymns. I didn’t mind it all; in fact I found parts of it good fun.[23]

The Redemptorists adapted well to the conditions of West Belfast and had little difficulty relating to the often violent world that some of their flock inhabited. A not untypical example of this coexistence of muscular Catholicism and physical-force republicanism was Gerry Adams’s maternal uncle Alfie Hannaway. An active republican from 1936, Alfie was first an organizer for the IRA boy scouts, na Fianna Eireann, and was responsible for recruiting Tom Williams, the legendary IRA man hanged in 1942. Alfie Hannaway was later promoted to the staff of the IRA’s Second Belfast Battalion, and in the early 1940s, as the IRA’s “English campaign” was fading, the Stormont government interned him. But Uncle Alfie wore another hat, which he donned weekly on behalf of the Redemptorists, as an obituary recorded:

His strong faith which was nurtured alongside his Republican beliefs led him to a strong association with the Clonard Monastery and a lifelong friendship with the Redemptorist priests. A member of both the boys’ and men’s confraternities for over 60 years, he earned the nickname “The Bishop” among his work mates, and when he was seen approaching the comment came that there must be another retreat in the offing and he was looking for people to enrol.[24]

The association between the Redemptorists of Clonard and the republicans could be seen in other ways. The Redemptorists’ celebrated annual novenas in West Belfast, nine intensive days of prayer, sermons, and services intended to refresh and strengthen religious belief, attract tens of thousands of local Catholics and transform the Clonard area into a huge religious carnival site. The novenas are held each June at around the time of the republican Bodenstown commemoration. The more spiritual republicans, like Alfie Hannaway, would spend the month of June first renewing their Catholic convictions in Clonard and then they would travel to County Kildare to reinvigorate their political faith at Wolfe Tone’s graveside.

Gerry Adams may not have been as pious as Uncle Alfie Hannaway, but he too forged a close relationship with the Redemptorists, which worked to his benefit when he had to convince skeptics of his peaceful bona fides, as the New York–based insurance mogul Bill Flynn recalled when he met Adams at Clonard in 1991: “I’m a Catholic,” he told the writer Conor O’Clery. “Adams is a Catholic, a communicant. I would see these priests and brothers, living there, trying to help people, under poor conditions. To see them, and their respect and friendliness for Adams, I knew he had to be a man of peace.”[25]

FATHER REID would later say to others who participated in his peace enterprise that the only chance of getting the IRA to the point of ending its campaign lay in putting together a process that had the imprimatur of the Catholic Church firmly stamped upon it. “Al would say over and over again that the church was the only body which could pull something like this off and that if she didn’t do it no one ever would,” recalled one confidant.[26] Reid advanced a number of reasons why the church should get substantially involved. The first was that the IRA trusted Catholic intermediaries in a way it did not trust others. There was an encouraging record. Priests, many of them Redemptorists, had acted as messengers or go-betweens during the prison hunger strikes as well as in feuds between rival republican groups. Adams had himself been party to a lot of this activity and trusted the individuals concerned, especially Reid. The Redemptorist was not modest about his influence over the Adams’s leadership: “He told us, and I think we came to believe him, that he had more close and continual contact with them than any other priest and possibly more than any person outside the IRA,” said one government contact.[27]

Reid also contended that the church would bring a special flavor to the effort arising from its extraordinary, centuries-old influence over Irish society. This, he would say, would bestow a moral authority and stature on the enterprise that would both legitimize it and add to the pressure on republicans to respond positively. Reid’s third point followed on from this. He argued that the church’s role in providing a neutral setting for the process would give cover to any other political parties who were talking to Adams and his colleagues. “The point he made, and John Hume grasped this particularly well, was that if there was a leak, responsibility could be shifted onto the church,” recalled an Irish participant.[28]

Had news of Reid’s 1982 overtures to Adams leaked out at the time, there is little doubt that in the prevailing climate the Irish political and media establishment would have been aghast at the revelation of any contact between the church and the Provos. But the news would have shocked many republicans as well. The hunger strike scars were still fresh when Reid approached Adams and the memory of the role played by the church still rankled with rank-and-file Provos. During the prison protest, republicans had bitterly accused the Catholic hierarchy of betraying the hunger strikers and of kowtowing to the Thatcher government. The church’s stance, they claimed, had been supine and had encouraged Thatcher to take an unbending line on the prison protest, contributing in no small measure to the ten hunger strike deaths. While the reality may have been much more complex, the belief was strongly held. When the protest was finally abandoned, the IRA prisoners issued a public statement attacking the church with unprecedented acrimony. Accusing the church of being “intricately immersed in the field of politics and deceit,” the prisoners went on, “We contend that their position has at all times been established by political consideration rather than the Christian values of truth and justice. Therefore their stance has been extremely immoral and misleading.”[29]

It was ironic therefore that it was to the head of the Catholic Church in Ireland, Cardinal Tomas O Fiaich, and through him the Vatican that Alec Reid, and, indirectly, Adams, turned for assistance in advancing his peace project. O Fiaich responded warmly to the suggestion from Reid that they hold talks with Gerry Adams. “Reid was convinced, and in turn persuaded Tom O Fiaich that the church should get involved in devising an alternative to the IRA’s violence, that her resources, influence, independence, authority, and the lines of communication open to her all meant that the church should encourage and even initiate talks with Adams,” recalled one peace process confidant.[30] Once Reid was able to tell the cardinal that Adams wanted to find a political alternative to violence and that he could deliver the IRA, O Fiaich agreed to take part. From then on, Reid could accurately describe his initiative as a church enterprise.

IT WAS REID’S and Adams’s good fortune that at this time the occupant of Ara Coeli (House of Heaven), the Irish primate’s official residence in Armagh, was Tomas O Fiaich. Had it been anybody else, the Reid initiative would perhaps have fallen at the first hurdle. Born in Cullyhanna, one of the most republican villages in South Armagh, O Fiaich was a staunch Irish nationalist, although one strongly opposed to the IRA’s armed struggle. An academic who for many years had headed St. Patrick’s College in Maynooth, County Kildare, the Irish church’s principal seminary, O Fiaich was appointed archbishop of Armagh in October 1977, in succession to the late Cardinal Conway, the Belfast-born conservative who had led the church during the turbulent years of civil rights protests and the birth of the Provisionals. Within two months O Fiaich was embroiled in controversy when, in an interview with the Dublin-based Irish Press newspaper, he showed his political colors. “I believe the British should withdraw from Ireland,” he said. “I think it is the only thing that will get things moving.”[31] The comment stirred the DUP leader, the Reverend Ian Paisley, to dub O Fiaich “the IRA’s bishop from Crossmaglen.”[32] Although Paisley got O Fiaich’s birthplace wrong, the label, in unionist eyes, was accurate and it stuck. O Fiaich was eventually elevated to the Irish primacy but only after a reportedly concerted British diplomatic effort to persuade the Vatican to choose someone else.

O Fiaich’s decision to back Reid’s enterprise was bound to animate controversy in the Irish church, not least because it came at a time when political pressure to isolate Adams and Sinn Fein was growing. The origin of much of this pressure elsewhere in Ireland as well as Britain lay not in London but in Dublin. After Adams’s success in the 1982 Assembly election, the British let it be known that he and his four colleagues would have full access to Northern Ireland Office ministers to discuss constituency matters.[33] Adams actually got to meet one of them, Housing Minister David Mitchell, with whom he raised the subject of poor living conditions in his native Ballymurphy. But it was to be one-shot event. The incoming coalition government in Dublin, headed by Garret FitzGerald, had followed these events with growing dismay. FitzGerald believed passionately that the Provos should be isolated and that to talk to them would only encourage IRA violence. He had angrily condemned the British for engaging in past cease-fire talks with the IRA, on one famous occasion going so far as to recruit U.S. Secretary of State Henry Kissinger to his cause, and so he successfully lobbied the NI secretary, Jim Prior, to reverse the decision. Sinn Fein was banished from respectable company and stayed that way for over a decade.

The principal supporter of this view in the Catholic Church was the second or third most important bishop in the country. Dr. Cahal Daly, who occupied the see of Down and Connor, centered in Belfast, was technically responsible for the spiritual welfare of Gerry Adams and most of the West Belfast IRA. He was, though, as violently opposed to what Adams represented as O Fiaich was critical of British policy in Ireland. Unlike Fitz Gerald, Daly rooted his objections in Catholic theology and moral teaching— namely that the IRA’s armed struggle was a sinful activity and thus abhorrent—but the effect was the same. In one typical onslaught he called the IRA “an evil and barbaric organisation,”[34] which should be shunned until such time as its leaders accepted the church’s moral teachings on the use of violence. He regularly urged Catholic voters not to support Sinn Fein.

Cahal Daly was a formidable opponent of the IRA. Born in County Antrim in 1917, he was ordained in June 1941 and spent most of his early years in the priesthood as an academic. An accomplished linguist and scholar who held separate doctorates in theology and philosophy, Daly had taught Scholastic philosophy at Queen’s University, Belfast, for twenty years. When he was elevated to the diocese of Ardagh and Clonmacnois in County Longford, the Catholic Bishop’s Conference was to discover that this thin ascetic figure was possessed of one of the ablest minds in Irish Catholicism. The other bishops soon asked him to compose their pastoral letters, thus giving Daly the opportunity to shape Catholic attitudes to a whole range of social and moral issues during the 1970s and 1980s. “He was streets ahead of the other bishops in intellectual ability,” recorded a contemporary.[35] While the Provisionals returned Daly’s hostility with interest—“They cordially detested him,” recalled one of Adams’s closest advisers[36]—the widespread respect for his intellectual abilities, in Rome and farther afield, meant that his opposition could not be ignored.

Before he embarked on his initiative, Alec Reid had been careful to secure the approval of the Redemptorist Congregation, both in Ireland and in Rome. Cardinal O Fiaich and Bishop Edward Daly of Derry, an often stringent critic of the IRA who nevertheless argued for dialogue with its leaders, agreed to write him letters of comfort, stating that his proposed talks with Adams had their approval. Armed with these letters, Reid won over his superiors. But he needed more. Constitutionally the Redemptorists were independent of Bishop Cahal Daly, even though the Clonard Monastery and their retreat house, St. Gerard’s in North Belfast, both later used as venues for peace talks, lay within Daly’s diocese of Down and Connor. The Irish Redemp torists answered to their superiors in Dublin and Rome, not to Down and Connor, but nevertheless the situation was awkward. After all, they were proposing church involvement in a process of which their local bishop heartily disapproved. In the councils of the Catholic Bishops’ Conference, the Southern bishops had left the issue to their Northern colleagues to debate and decide, but if support for the enterprise was to be found in Rome, Reid would have to have to answer Daly’s argument. Reid needed a heavyweight to counter Cahal Daly.

He found him in the shape of a fellow Southern Irish Redemptorist, Father Sean O’Riordan, from Tralee in County Kerry. Just a year older than Cahal Daly, O’Riordan matched the Belfast bishop in academic prowess, in Reid’s view. The professor of moral and pastoral theology at the Redemptorist-run Alphonsian Academy in Rome, O’Riordan also held a double doctorate, in theology and in history. A liberal theologian, he was a scholar of world renown who together with other academics in the Alphonsian Academy had crafted key documents for the Second Vatican Council.

By 1983 Reid had recruited O Fiaich to his cause but his attempts to persuade Cahal Daly to join his enterprise foundered.[37] All this brought the division within Irish Catholicism out into the open. In December 1983 Adams had written to Daly asking for a meeting. “During the last 12 months,” he wrote, “a number of people [had] suggested to me that you and I should have a meeting and that discussions between us would be of benefit to the people of West Belfast.”[38] Adams said that the unnamed intermediaries—meaning Alec Reid—had discussed with Daly the possibility of a meeting, and he had expected to hear from him. In the absence of an approach from the bishop, he was making himself “available” if Daly was interested.

Daly replied and marshaled the moral argument against dialogue. Adams’s membership of Sinn Fein and that party’s “unambiguous support” for the IRA’s violence was “completely contrary” to the church’s moral teaching, he wrote. If there was a change of heart on Adams’s part, he would be glad to meet, but otherwise, no, the meeting could not happen.[39] In January 1984 Adams responded, “Whilst accepting your desire for confidentiality I cannot, in honesty, accept any other conditions, no more than I would expect you to accept conditions set by me. We all have a duty to seek ways to resolve the present conflict. This can only be done by means of dialogue and by an honest and frank exchange of views.”[40]

Daly’s reservations about the enterprise were not just theological. He also distrusted the Provos and feared they would deliberately leak news of their meeting to cause him embarrassment and weaken his influence with unionists. When Adams made their correspondence public in April 1984, it seemed his doubts had been well founded. “He was not hostile to what was happening, nor did he ever try to stop it, but he wanted nothing to do with it himself,” explained a source close to Daly.[41] Seven years later, when Cahal Daly succeeded Tomas O Fiaich to the archdiocese of Armagh and was able to learn much more about the years of secret diplomacy, he became an enthusiastic supporter of the process. But in 1983 and 1984 his views were fixed. Nonetheless, the correspondence with Adams had set out the theological case against dialogue; the Redemptorists were obliged to respond, and they looked to O’Riordan for an adequate reply.

The opportunity came a year later, in January 1985, when again Adams and Daly clashed publicly. In a speech Daly had castigated the IRA’s campaign as immoral and counterproductive, and in response Adams published an open letter asking the bishop what his alternative to violence was. Adams posed a series of questions to Daly, and it was to these that O’Riordan addressed himself, as one figure familiar with the debate recalled. “Adams had asked Daly that if he was advocating a peaceful struggle, then was he ready to spell out an alternative, and O’Riordan basically said that Adams was justified in asking the question,” the source said. “He unequivocally backed Reid’s opinion that the church should talk to anyone and everyone and had a duty to help in the process of devising an alternative to armed struggle. He also welcomed a remark by Adams that he himself would be pleased to consider any alternative to armed struggle, and that was an important thing to say.”[42] O’Riordan tape-recorded his analysis in Rome and sent it to Clonard, where it was transcribed. Whenever Reid needed to recruit politicians or other influential figures to his enterprise, O’Riordan’s document was at hand to strengthen his case.

In January 1985 the Irish peace process was just over two years old, and by this stage the Vatican was being kept fully informed by Cardinal O Fiaich of the progress of the dialogue with Gerry Adams. One channel of communication was the papal nuncio in Ireland, Archbishop Gaetano Alibrandi, a figure whose sympathy for Irish republicanism was a constant source of friction between the government in London, which resented Alibrandi’s perceived interference in Northern Ireland’s affairs, and the Vatican. Valuable though Alibrandi’s support was, the Redemptorists and O Fiaich had more direct access to the Vatican through a contact who was often at Pope John Paul II’s elbow.[43] Archbishop Justin Rigali had been talent-spotted early on in his church career as a potential Vatican highflier and by 1985 had spent virtually all his life as a priest in Rome, immersed in the high politics of the church. By that stage he was well placed to shape papal thinking and Vatican policy.

Born into a devout Catholic family in Los Angeles in 1935 and ordained in 1961, Justin Rigali was, at just thirty-five years of age, appointed director of the English-language section of the Vatican’s Secretariat of State. The job description meant he liaised on behalf of the papacy with the church in Britain, Ireland, the Antipodes, and North America, all powerful and wealthy outposts of Catholicism. His elevation brought him right into the inner councils of the papacy. He became English-language translator for Pope Paul VI during his trips abroad, and when John Paul II ascended the throne of Saint Peter, it was Rigali who organized and accompanied the new pontiff on his many international trips, including his visit to Ireland in September 1979. It was during this journey that Rigali and Cardinal O Fiaich became friends, and as his influence with John Paul II grew, so the Clonard Redemptorists could be more certain of a friendly reception for Alec Reid’s diplomacy. Rigali, like John Paul II a theological conservative and a strong supporter of the pro-life lobby, rose irresistibly through the Vatican hierarchy, each move strengthening his influence in papal circles. During the years 1985 to 1990 he sat on no fewer than seven Vatican commissions; in December 1989 he was named secretary of the Congregation of Bishops and a year later secretary of the College of Cardinals, one of the most powerful posts in Rome. In 1994, as a reward for his years of loyal service and as a possible precursor to the red cardinal’s hat, John Paul II made him archbishop of St. Louis, Missouri. The esteem in which the pope held Rigali was illustrated in January 1999 when he chose St. Louis as the only venue for the U.S. leg of his trip to the Americas.

Rigali gave the fledgling Irish peace process vital access to the papacy, but, no less important, he also opened doors to the politically powerful U.S. church, especially to those in the hierarchy of Irish-American ancestry. “Rigali was a key man in influencing people like Cardinal O’Connor and Cardinal Law,” recalled one church source. “The Americans were involved long before Clinton came on the scene.”[44] Cardinals John O’Connor of New York and Bernard Law of Boston both received their red hats at around the same time, in 1984 and 1985, respectively, just as Alec Reid’s enterprise was picking up speed. Both men were to play crucial parts in persuading U.S. political opinion of Gerry Adams’s bona fides and greatly assisted his efforts to woo and then win over important figures in corporate Irish-America.

BY 1985 AND 1986 small glimpses of the secret diplomacy were occasionally visible, and with hindsight it is possible to discern the direction in which the talks were heading at this time. Three elements went together to make up the embryonic peace strategy. One was a proposal to forge an alliance of Irish nationalists which would replace the IRA’s violence as the cutting edge of the republican struggle. The second was the idea that a conference should be held to which all shades of opinion in Ireland would be invited and which would hammer out a political settlement. The third was a declaration of neutrality from the British, a statement that said they had no desire to impose a settlement on anyone in Northern Ireland and would be content to accept whatever political representatives could agree in negotiation. Taken together, the peace strategy outlined a way in which Adams could accept the principle of consent in relation to the political future of Northern Ireland while being able to say that none of the core republican doctrines had been abandoned or compromised.

The tip of the iceberg soon became visible. Gerry Adams first broached the idea of pan-nationalism at Sinn Fein’s November 1984 Ard Fheis when he called for “a firm, united and unambiguous demand from all Irish Nationalist parties” for an end to the unionist veto. This he said would create a new situation in which future arrangements could be worked out in “business-like negotiations” with all the parties in Ireland.[45] A few months later, in February 1985, during the course of a radio current affairs program, Adams asked the SDLP leader, John Hume, for talks to establish “a united nationalist approach” to the British.[46] In March he again called for “a united and dogmatic Irish nationalist approach” to the North.[47] These were all clues to the direction of the still-secret diplomacy. Cardinal O Fiaich meanwhile was treading similar ground. In January 1984 the Catholic primate urged the convening of a political forum either in Belfast or in Dublin where “representatives of all sides” could meet to discuss the future of the North.[48] Three times in 1985 he called on the British to indicate that they would not be staying in Northern Ireland forever and in the meantime to use their good offices to “try to bring Protestants and Catholics together.”[49] It is possible to discern in his remarks the germ of what soon became recognizably the Adams peace strategy.

It was all very tentative stuff, and only those who were privy to what was going on behind the scenes could be aware of its real significance. The rest of Ireland, distracted by the ferocity of the IRA’s campaign, dismissed the comments either as irrelevant or as deliberate distractions. The idea that the SDLP, Fianna Fail, or any other constitutional nationalist party in Ireland would want to link up with the Provisionals at that time or that the British would somehow agree to such far-fetched proposals was simply unimaginable.

It was also an impossible concept for most Provisional activists and supporters to grasp, at grassroots or even at leadership level. The common IRA/Sinn Fein view of the SDLP and Fianna Fail, the hatred at the core of their relationship, had been expressed in a bitter statement by the IRA prisoners at the end of the 1981 hunger strike—the same statement that had so bitterly excoriated the Catholic Church. Attitudes had changed little since then. Fianna Fail and the other Southern parties, the prisoners had said, were “accessories to the murder” of the ten hunger strikers for failing to confront Thatcher, while the SDLP was merely “an amalgamation of middle class Redmonites, devoid of principle, direction and courage.”[50] The enmity between the Provos and the Irish establishment was deep and seemingly unbridgeable.

Even if Gerry Adams had wanted to confide in his senior colleagues, it is highly questionable that he would have received a sympathetic hearing. Most IRA activists at this point believed unwaveringly in the armed struggle and tolerated electoral politics only because it seemed to offer, as Adams and others had told them it would, a way of increasing logistical support for the IRA. Had they thought that the political path down which Adams had taken them would lead into negotiations that threatened to dilute dearly held republican beliefs, most would have seen it as treachery.

So it was that the diplomacy of Alec Reid and Gerry Adams was kept a tightly guarded secret even from the IRA Army Council. Although Adams was slowly to win over key members of the council to his strategy with the passage of time and did confide in the small group of advisers around him in Belfast, his enterprise with Reid was never discussed or formally approved by the body charged with deciding IRA policy and strategy.[51]

What the reaction of IRA leaders would have been had they been fully aware of the ideological territory being traversed by the still-secret peace process can only be guessed at. The logic of the entire initiative was, of course, to produce an alternative to the IRA’s armed struggle. The goal of Reid’s diplomacy, the reason for its existence, was to achieve a cease-fire, and not long after it was launched, that possibility was being actively canvassed, as one senior Protestant cleric remembered:

It was around 1983 or 1984 that I became aware through Alec Reid that Adams was asking three questions, a sort of testing-of-the-water exercise in my view. The first question was “What would the parameters of a political settlement be if there was an IRA cease-fire?” Second, “Would the unionists and the British just pocket the cease-fire and return to the trenches?” and third, “How influential would people like me be with unionists? How proactive was I prepared to be?”[52]

THE REID-ADAMS INITIATIVE had gotten off to a moderately encouraging start. Adams himself was clearly willing to discuss an alternative to the IRA’s violence, and to contemplate huge ideological shifts. There was no shortage of Catholic Church figures, at very senior levels, eager to assist him in his difficult journey. But the real problems lay elesewhere, in persuading and cajoling and otherwise moving the Provisional support base in the same direction. By the mid-1980s, however, other events were combining to make that task a little less daunting than it appeared.

The Provisionals’ foray into electoral politics had been a triumph at the start, but Sinn Fein soon discovered that this new weapon was a two-edged sword. As long as electoral support for Sinn Fein was growing, as it was in the first two years, everything was fine, but once the vote started to slip and falter, this would be bound to be seen as a verdict on the IRA’s armed struggle. Sinn Fein’s electoral strategy made each poll a popularity or unpopularity contest for the IRA. The election of Adams to the West Belfast Westminster seat in the British general election of 1983 was the high point of the strategy. A total of nearly 103,000 people voted for Sinn Fein in that poll, almost 43 percent of the nationalist electorate, and the party came close to a second seat when Danny Morrison lost by just 80 or so votes in Mid-Ulster to the Paisleyite cleric Willie McCrea. But after that, it was a gentle but unstoppable slide downward. The Sinn Fein vote was being pinched in two directions. The novelty wore off for many Sinn Fein voters, and it became more and more difficult to cajole them into the polling booths, while the IRA’s continuing violence put a natural limit on Sinn Fein’s appeal. There was a broad swath of Catholic voters who just would not vote for Sinn Fein while the IRA killed and bombed. In the 1984 Euro election the SDLP leader, John Hume, easily outpolled Danny Morrison, and the SF share fell to 37 percent of the nationalist vote. A slight rise to 39 percent was recorded in the council elections the following year, but then two results, a series of Westminster by-elections in 1986 and the full British general election in 1987, saw the SF vote settle at around the 35 percent mark. As the SF vote dipped, Adams and his allies slowly but surely put more and more of the blame on botched IRA operations, which, they said, sapped support for Sinn Fein. And as the pressure grew on the IRA to refine, dilute, and ameliorate its campaign, it became easier to sell the idea of a political alternative.

Another key reason for the fall in the Sinn Fein vote was the Anglo-Irish Agreement of November 1985. The agreement, signed by Margaret Thatcher and Garret FitzGerald, gave the Irish Republic a formal if only advisory say in Northern Ireland affairs, which Dublin’s diplomats skillfully used to smooth the rough edges of Britain’s security policy, thereby reducing the grievances that nourished Catholic support for the IRA. The product of two tortuous years of diplomacy, the Hillsborough agreement was regarded North of the Border as a triumph for the SDLP’s John Hume and the politics of persuasion. A vocal unionist reaction against the agreement, which saw a burst of loyalist street politics and a temporary withdrawal by unionist parties from constitutional politics, helped persuade nationalists that they had won some sort of victory. It injected new life into the SDLP and stabilized nationalist politics. The Irish political establishment’s nightmare scenario, Sinn Fein outpolling the SDLP, evaporated. From then on it was clear that, short of another hunger strike or a repeat of Bloody Sunday, Sinn Fein’s vote would never rise much above the 80,000–90,000 mark again. The Anglo-Irish agreement had created a stalemate in the competition between the two nationalist parties, and this was to make it easier for Adams to argue among his close advisers that only a radical departure could break the logjam.

ELECTORAL SETBACKS raised questions about the direction Sinn Fein and the IRA could and should be taking, and that, in general, was of great assistance to the Reid-Adams project. But the IRA was still full of formidable men and women whose standing made them dangerous to have as opponents, who almost certainly would have stenuously resisted the infant peace process had they known of its existence. Fortunately for the Adams camp, some of the fiercest of these rivals were removed or otherwise fell at around the time that the peace process was beginning to develop its own sense of direction.

The first to go was Kevin Mallon, who had become director of operations after Brian Keenan’s arrest in 1979. Mallon, who also sat on the Army Council and was on the same ideological wavelength as O Conaill and O Bradaigh, was a powerful figure in the IRA whose record went back to the Border Campaign, during which he was tried and acquitted on the capital charge of murdering an RUC man. The IRA in Tyrone was extremely loyal to him, and many of its members there would have sided with him in any political dispute with the Adams camp. It was Adams’s great fortune that Mallon self-destructed and that he did not have to confront him.

Mallon’s downfall came over a number of botched kidnapping operations in the Republic that he had organized to help fill the IRA’s depleted coffers. One was the abduction of the racehorse Shergar, whose owner, the Aga Khan, refused to pay the ransom demanded by the IRA. As it was, Shergar was an uncontrollable, highly strung animal, and his kidnappers were forced to put him down. Worse was to follow in November 1983 when the IRA kidnapped the English-born supermarket executive Don Tidey in Dublin and demanded a seven-figure sum for his return. Three weeks later Tidey was rescued in a forest near Ballinamore, County Leitrim, but only after a fierce gun battle with the IRA in which an Irish army private and a Garda recruit were shot dead. The killings outraged opinion in the Republic, and for a while there was speculation that the government of Garret FitzGerald would actually proscribe Sinn Fein. In the following days, one senior republican after another emerged to express criticism of the kidnapping tactic, while Adams privately castigated “militarists” in the movement.[53] Mallon was finished.

The other upheaval in the Provisional IRA leadership was a much more serious affair for Adams, on a number of grounds: it was the first direct challenge to his political leadership of the Provisionals, it originated in his home base of West Belfast, and it was led by a figure whose regard in the IRA was beyond criticism. Ivor Bell had become chief of staff in the autumn of 1982, when Martin McGuinness stood aside to participate in the Assembly elections. Bell’s collaboration with Adams went back to the early 1970s, when both held staff positions in the Second Belfast Battalion and the Belfast Brigade. When the pair ended up in Long Kesh, they cooperated in the campaign against the Billy McKee leadership and together devised the plan to reorganize the IRA. The names Adams and Bell went inseparably together in the 1970s in the same way that those of Adams and McGuinness would twenty years later. Initially an enthusiastic supporter of electoral-ism—he had argued to run non-abstentionist candidates during the 1981 general election in the Republic—Bell developed reservations when he saw the negative impact on the IRA’s armed struggle, especially in Belfast, where IRA operations had been scaled down to facilitate Adams’s bid for the West Belfast seat and then for Morrison’s European contest.

In 1983 and 1984 the Belfast IRA, once the cockpit of the organization, had accounted for only eleven of the IRA’s ninety-five victims. Bell’s fears that the campaign was being run down and that resources that should have gone to the IRA were being diverted to Sinn Fein were shared by senior figures on Northern Command, including his partner, Anne Boyle, and the operations officer, Sean McIlvenna, and in the Belfast Brigade by its commander, Eddie Carmichael, while two other senior Belfast figures, Anto Murray and Danny McCann, sympathized with the criticism. Bell’s tenure as chief of staff came to an abrupt end in September 1983, when the Belfast Brigade adjutant, Robert “Beano” Lean, became the latest in a series of embarrassing and damaging IRA supergrasses to agree to testify against former comrades, and the damage he could do to the IRA was enormous. Although Lean later retracted, his evidence led to the arrest and brief imprisonment of Bell and some twenty-seven other senior figures. Under IRA rules Bell lost all rank in the organization, and Kevin McKenna, a key figure in the Tyrone IRA, took his place as chief of staff. On his release Bell got back onto the council, but his power had been eroded. He was given the task of liaising with Libyan intelligence and also joined the IRA Executive, from where he began plotting against the Adams strategy.

The Belfast Brigade meanwhile was intent on causing trouble for Sinn Fein. Angered by the amounts of money channeled to Sinn Fein from IRA coffers to finance election campaigns, the IRA leadership in the city carried out a number of killings and other actions intended to embarrass and damage Adams. Just before the June 1983 general election, the Belfast Brigade detonated a huge bomb outside Andersonstown RUC station, which devastated local homes right in the heart of the West Belfast constituency. The bombing angered many voters but, owing mostly to the fact that the opposition to Sinn Fein was split two ways, failed to deprive Adams of victory. In December 1983 the up-and-coming Ulster unionist politician Edgar Graham was shot dead outside the law department of Queen’s University, Belfast, where he taught. In April 1984 Mary Travers, the daughter of a leading Catholic magistrate, Tom Travers, was shot dead and her father seriously wounded when an IRA gunmen opened fire as they left Mass in an affluent area of South Belfast. Travers later claimed the gunman had also tried to shoot his wife in the head but that his weapon had misfired.[54] The author was in the home of one of Adams’s closest advisers not long after the shooting happened when Adams phoned from Donegal, where he was on vacation. It was clear from their exchange that the Sinn Fein president was dismayed to hear what had happened and seemingly did not know that the operation was in the pipeline. An attack such as that on Travers should have been cleared at the highest political level and Adams, or at least those close to him, should have been told about it. Two months later, in the midst of Danny Morrison’s Euro campaign, the Belfast IRA shot dead a minor criminal, Jimmy Campbell, as he sat drinking in a social club off the Falls Road. The punishment was out of all proportion to Campbell’s transgressions—he was just a petty thief—but the killing had been done to embarrass Sinn Fein, and it led to protests and pickets outside one West Belfast polling station.

It is not clear just when Bell and his colleagues began their attempt to overthrow Adams. The version put around by Adams’s allies afterward was that it started when Bell’s partner, Anne Boyle, was dismissed from Northern Command, allegedly for endangering the security of a Northern Command staff meeting, and that this led an angry Bell to set the plot in motion. Another says that the spark was the death in December 1984 of another Bell ally on Northern Command, Sean McIlvenna, a North Belfast IRA man who had been operating on the Border for some years. McIlvenna, thirty-three, was killed by a single bullet in the back when, by chance, the RUC came across him and other IRA men shortly after a land mine attack in County Armagh.[55] The dissidents complained that while Adams and his Sinn Fein coterie were able to buy bulletproof jackets to shield them from loyalist attack, the IRA could not afford to give men like McIlvenna the same protection, and this is why the police bullet had felled him.[56] Bell was convinced that a majority of IRA Volunteers were unhappy with the electoral strategy, and he had come close once before to successfully challenging it. A majority of the Council had indicated their sympathy for him when he was chief of staff, but he had been unseated just as he was about to move against Adams.

Whatever the truth, when Bell began to canvass fellow members of the Executive in late 1984 and early 1985 to call for an extraordinary Army Convention in a bid to abandon or severely limit Sinn Fein’s electoral strategy, it did not take long for Adams to hear about it. He moved swiftly and resolutely against the dissidents. Bell, Boyle, Carmichael, McCann, and Murray were court-martialed and expelled from the IRA, although two, McCann and Murray, were later readmitted. Charged with treachery, Bell faced a possible death sentence, but Adams interceded, no doubt aware of the adverse publicity and speculation that would follow the discovery of his corpse in a South Armagh laneway, the customary dumping ground for such victims. Nevertheless the death threat continued to hang over Bell. He and Anne Boyle have lived quietly in West Belfast ever since; both have stubbornly refused to talk about the events of that time, least of all to journalists. The Ivor Bell rebellion was over, and in the space of a few months Adams had seen Kevin Mallon and Ivor Bell, two potentially serious obstacles to his leadership—and to the infant peace process—removed.

By the end of 1985 the peace process was just over three years old, and already it was clear what the driving force behind it was. Sinn Fein’s success in the 1982 Assembly election had opened up intriguing possibilities. The results of that election and especially Adams’s own election to the West Belfast seat a year later meant that Sinn Fein was now a major political force on the island, confounding and dismaying its critics and enemies in both Ireland and Britain. The Sinn Fein achievement had alarmed supporters of the political status quo and had forced Britain and the Irish government to put together the Anglo-Irish agreement, the most ambitious attempt to stabilize Northern Ireland in ten years.

Sinn Fein’s improved circumstances meant that when, in the aftermath of Sergeant Cochrane’s kidnapping, Father Alec Reid visited Gerry Adams, he found the Sinn Fein and IRA leader in a much more receptive frame of mind than ever before to discuss options to violence. On any prior occasion such discussions would have taken place in a vacuum. Now, after the electoral successes, there was something real to talk about.

Success at the hustings fueled the peace process but so, paradoxically, did setbacks. Unwittingly, perhaps, the Sinn Fein and IRA grassroots had backed a strategy that almost precisely measured popular disenchantment with the IRA’s violence. In the days before Sinn Fein fought elections, the IRA could afford to disregard such considerations and take refuge in the thought that while the Irish people might not vote for republicanism, deep down they secretly supported it. Now there was no such fantasy in which to seek refuge.

As 1986 dawned, much of the groundwork in the peace process had been done and Adams’s own internal problems had been eased with the departure of key rivals. But a peace process with only one participant was worthless. The next key questions were about the British and Irish governments and the extent to which they would be prepared to deal with Adams. The answer to those questions would determine the success or failure of the whole enterprise.

8. Dealing with the Brits

The history of contacts between the IRA and the British government was, by 1986, a long but unhappy one. British ministers, including the then Northern Ireland secretary, William Whitelaw, had met a delegation of IRA leaders, including Gerry Adams and Martin McGuin ness, in London as far back as June 1972 during the first IRA cease-fire. In 1975 there were further contacts, although this time no British politicians were directly involved, and instead the talking was done between British officials and representatives of the Army Council. The talking in both cases ended in angry recrimination. In 1972 the cease-fire was abandoned within days of the contact amid IRA accusations of British bad faith, while in 1975 Adams and others alleged that the British had tricked the IRA into the cessation in order to buy time to devise the IRA’s defeat. Officially the Army Council was in 1986 so hostile to the notion of a cease-fire that it was committed never to talk to the British again, unless it was about their withdrawal from Ireland.

All this meant that when Gerry Adams decided to open a dialogue with the British about the burgeoning peace process, he chose Father Reid to act as his go-between and representative. The Redemptorist acted as a sort of clerical cutoff, providing cover to all involved, both republican and British, and it was a task he performed with diligence and discretion when, finally, the moment came for Adams to move. Precisely when the pair made contact with the British government is not known, but the evidence strongly suggests that sometime in 1986, courtesy of Father Reid, a highly secret line of communication was opened between the Northern Ireland secretary, Tom King, and the Sinn Fein leader.

In the intervening years both the British and the Sinn Fein leadership have, for reasons best known to themselves, gone to considerable lengths to persuade the outside world that the first contact between them did not take place until much later, in 1990, when British intelligence made contact with the IRA and commenced a dialogue with Martin McGuinness.[1] But both parties have been extremely economical with the truth about the full extent of their dealings with each other.

They each had very good reasons. Adams had not told his Army Council colleagues about his dialogue with Father Reid, much less the indirect contact opened up on his behalf with the British through the Redemptorist priest. Had they known just about the talks with Reid, Adams would have faced some hard questions from other IRA leaders. But knowledge that Adams was talking to the British government, especially one led by the woman most Republicans blamed for the 1981 hunger strike deaths, and was doing so without their sanction, would have set off alarm bells in the Army Council, possibly with serious consequences for Adams. The government of Margaret Thatcher had equally compelling reasons to keep a tight lid on the story. Mrs. Thatcher’s hatred and contempt for the IRA was renowned and although she had allowed MI6 officers to speak to the IRA during the hunger strikes, by 1986 there was another reason to keep her government’s contact with Adams, albeit filtered through Father Reid, a tightly guarded secret. Two years earlier, in October 1984, a 20-pound IRA bomb exploded in a bathroom of the Grand Hotel, in the south-eastern English coastal resort of Brighton, during the annual Conservative Party conference. The bomb exploded just before 3 a.m. and came very close to killing Thatcher and other members of her Cabinet. Five people were killed, including Sir Anthony Berry MP, while Thatcher’s hard-line Trade and Industry Secretary, Norman Tebbit, was badly injured in the blast. Mrs. Thatcher herself had a narrow escape; a bathroom she had been in a few minutes before the explosion was destroyed. In a statement the IRA said: “Today, we were unlucky, but remember, we only have to be lucky once— you will have to be lucky always.” In fact this statement was mostly bluster; the IRA never again attempted to kill Mrs. Thatcher, while senior figures, including Adams, were said by colleagues to be privately thankful that the effort had failed in its primary aim. Even so, the Thatcher government was taking an extraordinary political risk by opening up communication with Adams. Common sense suggests that someone in British officialdom with sufficient influence and knowledge of the background may have advised that the risk was worth taking.

As can now be revealed for the first time, confirmation that Gerry Adams was in indirect conversations with the British government as early as 1986, or at least 1987, has been given in interviews with the author by two former British secretaries of state for Northern Ireland who conducted these dealings with the Sinn Fein leader. One was Tom King, who came to Belfast in September 1985, just before the Hillsborough agreement was signed, and the other was his successor, Peter Brooke, secretary of state between 1989 and 1992, a figure whose term of office is most identified with the emergence of the first public clues about the secret diplomacy that lay behind the peace process. Both men have since been elevated to the British House of Lords.

A scion of the British establishment, Brooke came to Northern Ireland with what seemed like impeccable unionist credentials. His father had been a Tory home secretary in the early 1960s when the unionist and British Conservative parties were as one, and his family roots in Ireland went back 250 years to the heyday of the Anglo-Irish ascendancy. Although welcomed at first by unionists, he was viewed with some unease and doubt by the end of his term, since his sometimes less than forthright approach to the Provisionals unsettled them. Their alarm bells were triggered very early on in his ministerial term when he took a soft line on the defining question of how the British would respond to an IRA cease-fire. A charming and often amusing man, Brooke was well liked by the media and became extremely popular with nationalists, not least because unionists distrusted him.

In an interview given to a press agency in November 1989 to mark his first one hundred days in office, Brooke conceded that the British and the IRA were in a military stalemate. He then made remarks that came so unexpectedly that they confounded observers at the time. Now, in the knowledge of the secret Reid-Adams diplomacy, they make perfect sense. “There has to be a possibility that at some stage debate might start within the terrorist community,” he said. “Now, if that were to occur, then you would move towards a point, if in fact the terrorists were to decide that the moment had come when they wished to withdraw from their activities, then I think the government would need to be imaginative in those circumstances as to how that process should be managed.”[2]

In an interview at his office in Westminster, many years later, Brooke explained to the author what lay behind his words. “That 1989 interview was based on a mixture of what I had heard from John Hume, MI5, and military intelligence and finally what I had heard about Reid,” he said. “I had heard about Reid during a briefing I got from Tom King. What Reid was saying to us was that there was an opportunity here to end it, and one of the aims of what I said in the interview was to communicate my attitude on talks to Sinn Fein.”[3]

What Peter Brooke learned about the Reid-Adams diplomacy suggests that the dialogue had begun not long after Tom King had become secretary of state. “I… saw intelligence [on Reid-Adams],” he told the author, “going back as far as 1986, intelligence which underlay my November 1989 interview.”[4]

Tom King also confirmed the existence of the Reid-Adams pipeline to the author during interviews conducted at Westminster in 2000. A former British army officer and gentleman farmer-businessman turned MP for one of England’s more affluent shire constituencies, King was fortunate to have been made a minister. A close friend of Michael Heseltine, a major critic of Thatcher’s right-wing ideology, King was on the wrong end of the Tory Party spectrum for many of Thatcher’s allies, but when he arrived in Belfast he made efforts to get in the unionists’ good books. He hit the headlines when a few weeks after the Hillsborough pact had been signed, he said the Irish prime minister, Garret FitzGerald, had “accepted that for all practical purposes and into perpetuity there will never be a united Ireland.”[5] The remark ruptured relations with Dublin but delighted unionists.

King could not say whether or not the Redemptorist pipeline to Adams was in place before he arrived in Belfast, but he did confirm that it was working while he was there. “I became aware of it,” he said. “I may have asked questions, but I think I knew there was a Clonard connection.”[6] King also confirmed that Margaret Thatcher was aware of the contact with Adams and had approved it.

King said that he was first active in the communication “at the turn” of 1986–87; in other words, that the dialogue did begin sometime in 1986.[7] This accords with Peter Brooke’s memory of the intelligence briefing he received, which also suggested that the pipeline was opened sometime in 1986—at the very least. The matter is important not just because it fundamentally challenges other accounts of the peace process, which have the first contact between republicans and the British happening much later, but because it suggests that the British must have had an invaluable insight into the mind and intentions of the republican leader on the eve of what proved to be a turbulent and violent period in Northern Ireland, encompassing the smuggling of huge amounts of Libyan arms to Ireland and an intensification of IRA violence. The knowledge that Adams was at least prepared to talk about peace at this time was priceless information.

Other evidence suggests that the Adams-King dialogue continued until late 1987 or mid-1988 when spectacular bursts of IRA violence—the postponed “Tet offensive”—caused the British to cool the diplomacy. Either way, it is clear beyond contradiction that secret talks between the British and Adams began much earlier than any of the participants have so far admitted, a disclosure that demands a fundamental reassessment of the genesis and origins of the Irish peace process.

This disclosure also challenges another long-held belief about the peace process—namely, that it was the Irish government in Dublin with which Reid and Adams first made contact and that this was followed, in an admittedly lengthy and tortuous manner, by attempts from Dublin to coax the British to join the process. That is clearly not true. It is now evident that at the very least the British were contacted at the same time as Dublin, and that, far from having to be cajoled by Dublin into joining the process, the British, very possibly, were engaged first. What also emerges from this account is that the peace process was well advanced by the time the Eksund set sail from the Mediterranean and into the trap laid between them by British intelligence and the well-informed IRA traitor. Had the vessel and its deadly cargo gotten through and the IRA been able to launch its “Tet offensive,” then a well-developed peace process would have been its first casualty. Whoever betrayed the Eksund saved the process, whether wittingly or otherwise.

It is also clear that the Reid-Adams dialogue with Tom King played an enormously important part in the evolution of the peace process even if the Provisional representation was confined to Gerry Adams and his close advisers and did not, at that stage, embrace the rest of the IRA leadership. Both Brooke and King believed, they told the author, that Adams was acting with the approval and knowledge of the Army Council. They were wrong, but they were not to know that. Adams was engaged in an enterprise of which the Army Council knew nothing, and had it been privy to these events it is likely that it would have heartily and angrily disapproved. The Reid-Adams initiative was a hugely dangerous exercise for the Sinn Fein leader.

A crucial moment in the secret contact came early on when King received a lengthy letter from Adams, which posed a number of questions about Britain’s policy in Northern Ireland and its attitude toward the Provisionals. King has confirmed that this happened and indeed believes there may have been two letters from Adams. “I had one or two letters but I can’t remember the detail, just the opening, ‘A Chara,’” he said.[8] “A Chara” or “friend” is the traditional Irish opening to a letter and the equivalent of “Dear so-and-so.”

In an earlier interview King said that he had received a series of questions from the Reid contact, and it is apparent that this was a reference to the Adams letter. “We received questions,” he said, “asking us, ‘Is this right?’ ‘Is that right?’ ‘How can that possibly be?’ ‘Surely there are strategic and economic reasons to hang on in Northern Ireland?’”[9]

There were six questions in Adams’s letter. Although Tom King says he has forgotten what they were, another well-informed participant in the peace process does remember. They were the following:

1. What is the nature of the British government’s interest in Ireland?

2. What is the British government’s attitude toward self-determination, and what will it do to ensure that there is no veto exercised which would militate against the exercise of self-determination?

3. Will the British government play a positive role in persuading unionists toward self-determination?

4. Will the British government publicly state its position, assuming the appropriate replies to these questions?

5. Will the British government accept that the republican movement will represent itself in any negotiations?

6. In the context of dialogue free from interference, will the British government publicly state its intention to withdraw from Ireland and give a date by which such withdrawal will be complete?[10]

Taken at face, the questions contained nothing that any of Gerry Adams’s IRA colleagues could have quarreled with. They were all seemingly embedded in traditional IRA dogma, all apparently predicated upon the analysis that Britain’s presence in Northern Ireland prevented the Irish people as a whole from freely deciding their own future. But the questions opened the way for the British to offer Adams an exit from armed struggle by redefining what was meant by concepts such as “self-determination” and “withdrawal.”

Tom King has declined to say whether he or someone else wrote the document that was sent in reply to Adams, but that there was a reply is not in doubt. It came in an undated, unsigned statement, twelve paragraphs and nearly a thousand words in length, that responded to four of Adams’s six questions in the order in which they were asked.[11]

The British answers set the agenda for the rest of the peace process and created a template for negotiations that after delays and many failures of nerve were to lead some ten or so years later to the Good Friday Agreement of April 1998. Most of the concepts and much of the language that characterized the process during the long years of its gestation and development were contained in the secret British reply to Gerry Adams’s letter to Tom King. If the peace process has a philosophical fountainhead, this is it.

The British reply, now revealed for the first time, is possibly one of the key, historic documents of the peace process. It reads in full:

The Provisional IRA say they are fighting a war of independence to win freedom and self-determination for the people of Ireland. They maintain that force must be used to persuade the British government to acknowledge such a freedom because it has a colonial self-interest in remaining in Northern Ireland and in exercising authority there. This interpretation of the political situation in Northern Ireland is based upon the false assumption that Britain has some interest of her own in remaining there and is therefore engaged in some way in defending those interests by force of military strength.

Britain of course has an interest in Northern Ireland which is to respond with a warm goodwill and friendship to the needs of the people of Northern Ireland as a whole. (We respect and admire the contribution which they have made to the union in peace and war.) But let me be very clear! In the second half of the 20th century no matter what has been the position in the past the British government has no political, military, strategic, or economic interest in staying in Ireland or in the exercise of authority there that could transcend respect for the wishes of the majority in Northern Ireland.

The political and security situation as it has now developed in Northern Ireland is due to the historical, political, religious and cultural divisions which separate the people of the nationalist tradition from the people of the unionist tradition in Ireland. These divisions are at the root of the conflict there and not any self-interested dominion policies of the British government.

The central issue in the conflict therefore is not to persuade the British government to decide on the question of self-determination in Ireland but to bridge the divisions between the people of both traditions there in a way that will enable them to decide it freely and democratically for themselves. The political consensus and agreement which would bridge these traditional divisions and so open the way to a peaceful resolution of the questions of Irish self-determination can only be achieved through the processes of political debate and dialogue between the peoples of both traditions.

Should they through their political representatives agree to engage in such dialogue the British government would do everything possible to accommodate and facilitate it. It would, for example, provide the necessary framework and forum which could take the form of an ongoing conference or convention composed of the official representatives of the people of both traditions. It would confine its own role in such a dialogue to that of facilitator and so would not officially take part in it or, much less, seek in any way to dictate to it.

This means the British government is prepared to withdraw from the central area of historical, political, religious and cultural conflict and from the central forum of political debate that would seek to resolve it so that the parties to the conflict, namely the people of the nationalist tradition and the people of the unionist tradition can engage freely, independently and democratically in the political dialogue and in agreement-making which would bridge their divisions.

Should this dialogue result in agreements based on the consent of the people of the two traditions which would bridge the political divisions between them and to define the forms of new political structures which would embody, institute and organise such agreements the British government would respond with the necessary legislation.

We hope that this statement will clarify our present policies in Northern Ireland. We also hope that it will prompt the Provisional IRA to review their present policies and to see that the use of violent tactics against “colonial interests” which in fact do not exist is a mistake and a tragedy beyond words not only for the people of Ireland generally but also for themselves.

Questions about Irish political self-determination can only be answered by the people of the nationalist and unionist traditions in democratic conference among themselves. Britain can assist them but she cannot answer for them because the problem is not that the British government will not agree to Irish self-determination but that the people of the two divided traditions do not at present agree amongst themselves on how to exercise it.

It is our fervent hope that what we have said in our statement will convince the Provisional IRA and the Sinn Fein party that the political divisions between the people of the two traditions in Ireland which are at the heart of the conflict cannot be bridged much less healed by force. Their use of force is, in fact, holding back the day of justice and peace for the people of both traditions because the political debate which would bring that day forward cannot take place freely and democratically while they or indeed any significant section of either tradition are using the argument of the gun.

We accept that the republican tradition within the nationalist community should be represented at the peace conference table but we cannot and will never accept that the Provisional IRA and the Sinn Fein party should be represented there while they continue to use the tactics of violence. Apart from other considerations they themselves would never really expect to sit down with representatives of the unionist tradition to confer and decide about the political future while unionists feel as they do that Republicans are holding a gun to their heads.

The strategies of peacemaking are the strategies of political dialogue, political persuasion and political force. These alone can create the political agreements whatever their developing or final form which would breach the divisions between the people of the nationalist tradition and the people of the unionist tradition in Ireland. Only on such a bridge can the conflict be ended and the foundation of a just and lasting peace for both traditions be soundly and progressively established.[12]

The secret British document made three significant and important statements to Adams. It said first that, contrary to IRA dogma, Britain had no interests in Northern Ireland, no selfish reason for remaining there, and was indifferent about the nature of any settlement as long as it was not imposed by force and did not offend the principle of consent. Britain was also prepared to allow Sinn Fein to have a part in the forging of any settlement, but that would be conditional on an IRA cease-fire. That was not a departure from policy. Britain had always made it clear that IRA violence was the obstacle to Sinn Fein’s taking part in normal political life. What made this offer interesting was that Britain also offered to convene the sort of conference that Irish nationalists had long asked for and promised that it would not attempt to impose its will upon its deliberations. Sinn Fein could attend, providing the IRA ended its violence, while Britain also gave a pledge to implement any settlement in legislation, a hint that the 1920 Government of Ireland Act, which upheld the partitionist settlement, could be significantly amended and even replaced.

The most intriguing and momentous part of the British reply came with the novel redefinition of what British withdrawal would actually mean. The traditional republican interpretation of this was straightforward. The British would physically leave, their troopships would sail from Belfast harbor, and their politicans and officials would decamp while Irishmen, Protestant and Catholic, got on with the business of constructing a new government for the whole of Ireland. Irish and British constitutional politicians had long rejected this scenario as unworkable, largely on the grounds that Protestants would resist it probably to the point of civil war. But now the British came up with an ingenious reworking of the idea. Britain would not physically withdraw but it would withdraw politically, that is, it would promise not to interfere or intervene in the decision-making process—the all-party negotiations in other words, aimed at securing a final settlement. Irishmen would be allowed to construct the deal while the British acted as impartial facilitators.

Two of Adams’s questions went unanswered, and many of the ensuing problems faced by the peace process would be caused by this. The British did not respond to Adams’s request to make their position public—that would not happen for two or three years, until Peter Brooke gave his neutrality speech. Nor would the British name a date for the end of the process, the point at which they would finally leave Ireland. At the time of this writing the answer has still not been given.

Nevertheless, it is possible to see in the secret British letter the blueprint for the peace process as it unfolded in the coming years. Armed with this redefinition of British disengagement and a statement of neutrality, Adams could happily continue the secret diplomacy with Reid while the road map showed that all-party negotiations, an IRA cease-fire, and adherence to the principle of peaceful consent were the way forward. This is precisely what did happen, although it would be nearly eleven years before the destination was reached. The argument that the peace process was a precooked deal and that much of what happened in the process was about managing the various forces involved, not least the IRA and Sinn Fein rank and file, finds its origin in this document.

AT THE START of their dialogue Tom King had responded diligently to the Reid-Adams pipeline, but other diplomacy was under way as well, albeit more tentative. Sometime in early 1987, and separately from the more considered response to Adams’s letter, King authorized a statement to be delivered to the Sinn Fein leader saying that there was no strategic or economic interest in staying in Northern Ireland as far as Britain was concerned, nothing “that would override the democratic will of the people of Northern Ireland.”[13] According to King this was in response to an overture from the SDLP leader, John Hume, rather than the Reid-Adams pipeline, and King’s memory is that it happened in early 1987. “It was before Loughgall [the SAS ambush that killed eight IRA members] because after Loughgall things slowed down, and we were annoyed by this,” he recalled.[14]

Tentative peace talks may have been opened with Gerry Adams, but the latter’s colleagues in the IRA continued with their armed struggle regardless. Unaware of the secret diplomacy, certainly ignorant of the correspondence between Tom King and Adams, the Army Council authorized the launch of the much delayed and reduced “Tet offensive.” It was this decision, later on in 1987, that brought the dialogue between King and Reid-Adams to its first crisis.

At the end of August 1987, three Irish people—two men and a woman, all with Dublin addresses—were arrested in the grounds of King’s estate near Bridgeport, Wiltshire, and later charged with conspiracy to murder the Northern Ireland secretary. Paramilitary-style equipment, notes of King’s car registration numbers, and a large cash sum were found in their campsite nearby. The three were initially convicted but later cleared on appeal. Nevertheless, King was convinced that all the while he was sending messages to and receiving them from Gerry Adams, the IRA had been plotting to kill him, and he was furious. A few months later, in March 1988, the attentuated “Tet offensive” was launched with an attempt to bomb the Mediterranean colony of Gibraltar, and that was followed by a wave of violence culminating that September in an effort to kill the head of the Northern Ireland civil service, Ken Bloomfield, in a bomb attack on his home outside Belfast.

The alleged IRA surveillance of his home caused an incandescent King to cut the cord to Alec Reid. “He told Adams to get stuffed,” recalled one informed source. “He took Bridgeport as a very personal affront, and I also suspect that he was getting a lot of support from Mrs. Thatcher for cutting off the talks.”[15] In autumn 1988 the Thatcher government responded to the IRA upsurge with a tough package of security measures, which included a ban on radio and TV interviews with members of Sinn Fein, a change in the law that eroded the suspect’s right to silence during police interviews, and a mandatory oath of nonviolence for all candidates in local elections in Northern Ireland. What brought matters to this point was the stark difference between the desire for peace of which Reid and Adams talked and the reality of the IRA’s violence, a difference the British might have better understood had they realized just how secret the Reid-Adams diplomacy was.

From then on and for years ahead, there were two distinct British views of Adams’s and the Provos’ motives, as one participant with a ringside seat remembered:

At the start Tom King saw this as something that had to be pursued, that it was an historical chance. He saw it as the IRA going through their dying throes, and they were looking for a way out. It was only after what happened at Bridgeport that he concluded they were spoofing. The Foreign Office was very positive all the way through. I can’t remember a time when they were negative. In the Northern Ireland Office it depended very much on who was there, but the prevalent view was that they were doing something new but for the same purpose. The difficulty was that it was so hard to take Reid seriously; he was so passionate about things [but] with no evidence of change.[16]

The letter from Adams to King that produced the remarkable British reply was undated—as is the secret reply itself—but public comments by the Sinn Fein leader suggest that the ideas contained in the British correspondence were circulating inside Adams’s head toward the end of 1986 and the beginning of 1987. In a December 1986 interview with the Irish Times, for example, he stated, “I say the British still have an interest in this country. If they don’t, let them say it. What they should be saying, quite simply, is that they have no interest in Ireland and they’re going to leave. That would have the effect very quickly of bringing everyone—and people might say that includes us—to their senses and starting to work something out.”[17] That point was dealt with in the secret British answer.

Three months later there was more of the same. In a speech to the Oxford Union, Adams called “for the establishment of an all-Ireland constitutional conference consisting of elected representatives of the Irish people that would draw up a new constitution for an independent Ireland and organise a new system of government open to all significant organisations of political opinion in Ireland.”[18] The British reply also dealt with that matter. With hindsight it is difficult not to conclude that these comments were an intriguing public glimpse of a still very secret diplomacy.

One unresolved question concerns the identity of the British civil servants who were dealing directly with Father Alec Reid, those who very probably drafted the secret answer to the Adams letter. The conduit between Tom King and those who were talking to Alec Reid was Ian Burns, then the deputy under secretary of state at the Northern Ireland Office, who retired in the early 1990s.[19] According to informed sources, Burns was a consistently optimistic influence on NIO thinking during this period, and to some of his colleagues this suggests he was privy to much of Adams’s thinking. There is also evidence that he was the conduit to King for highly confidential information coming from a second source in or near to the Pro visionals, this time in Derry. The contact civil servant in that case was Mark Elliott, a Foreign Office official seconded to the NIO who was in charge of the British side of the joint Irish-British secretariat, which had been set up as part of the 1985 Anglo-Irish Agreement. Elliott made regular monthly trips to Derry, and his subsequent reports were considered so sensitive that for security reasons he wrote them out by hand and delivered them directly to Burns.[20]

Elliott’s role had its roots in the bureaucratic structure erected by the British when the outbreak of violence in August 1969 and the deployment of British troops forced a more direct involvement by London in the day-today running of Northern Ireland’s affairs. To keep an eye on the increasingly volatile situation, the British sent over a senior civil servant to be its official representative, a sort of ambassadorial figure, to liaise with the unionist government. Over the years the office became the regional headquarters of British intelligence, and it was eventually located in a secluded villa called Laneside, in the pleasant seaside town of Hollywood, east of Belfast. It was from Laneside that British officials like Michael Oatley of MI6 made discreet contact with the IRA in 1974 and 1975 on the eve of the cease-fire, and it was there that secret meetings with IRA leaders were held. Laneside was the nerve center of British policy in Northern Ireland.

Eventually Laneside was closed down. Some of the office’s functions were transferred to the Political Affairs Bureau (PAB) at the Northern Ireland Office, and the interest of Whitehall was signified by the presence of a Foreign Office official at its head. In later years the job was given to Mark Elliott. The intelligence section of Laneside went its own way and was later housed in a new office block in the Stormont complex along with MI5.

Neither Tom King nor Peter Brooke can say, will say, or can remember who on their staff dealt with Father Alec Reid. “I don’t know who was dealing with Reid,” said King bluntly; “probably the political relations side.”[21] But the evidence is that the Political Affairs Bureau in the NIO was kept firmly in the dark about the secret communications with Gerry Adams, as one former British official explained: “I never had the slightest wind of it, no hint of it all. I can’t believe that the way the office was organized it could have been possible. No one would have thought of doing it that way, because it would have got out almost immediately,” he said.[22] This official, like other of his former colleagues interviewed by the author, is convinced that the intelligence services must have handled the contacts.

We were very much part of the ordinary office, and it’s conceivable that one person could have been up to it, but I just find it impossible to believe that. I think it was done through other channels. And frankly I suspect if you were doing something like that, you would have left it to someone who perhaps had some training and background. They are the sort of people who are trained to do that sort of thing, and they are probably better placed for that sort of endeavor. What you also have to remember is that when this started (a) it was not stated government policy, and (b) it easily could have gone very, very pear-shaped, and I would have thought whichever minister or ministers [were] involved would have wanted some cover.[23]

Some aspects of the diplomacy suggest that it embraced a strong “need to know” element of the sort that usually indicates an intelligence role rather than the conventional duties of a mainstream government department. Both King and Brooke denied all knowledge of the secret British reply to Gerry Adams and said they had never read the document until shown a copy by the author. Had they known about it at the time, the political risk to them would have been enormous. The document was explosive, its content an absolute contradiction of stated British policy toward the IRA in the mid- to late 1980s, which was to shun and isolate both it and Sinn Fein. Had the document or any of its contents leaked, the sitting secretary of state, either King or Brooke, would have been forced to resign.

The piece of evidence that suggests that both King and Brooke may have been kept in the dark about the secret reply to Adams is that one of them, Peter Brooke, had no knowledge at all of Adams’s original letter to Tom King until this writer told him. He had not been told about it during his initial briefing as NI secretary from civil servants, and Tom King had not disclosed it to him either, even though, according to Brooke, King had briefed him on the Reid-Adams conduit. It seemed that someone, somewhere in the NIO or possibly in the intelligence apparatus had decided it would be better if Brooke remained ignorant about the correspondence. When the author told Brooke about the King letter during lunch in Westminster, he was astonished and frankly incredulous. It was the first he had heard of the letter.[24] But the existence of the Adams correspondence is undisputed by King himself, even if he disavows all knowledge of the reply. The British politicians, it appears, were only partly aware of the full picture.

There is another clue that strongly suggests intelligence involvement. It also comes from Peter Brooke’s memory of events. Previous accounts of the peace process have told how in late 1990 a senior MI5 officer called John Deverill met Brooke with a proposal to initiate secret talks with the IRA leadership and that when Brooke gave the go-ahead, back-channel negotiations began.

In an interview with the author in June 2000, Brooke explained that what really happened was that Deverill came to see him not to get authority to start talks with the IRA but to inform him that a different intelligence officer would be taking over day-to-day handling of what appeared to be an already ongoing operation.

What was put to me in relation to political cover—and I was being told because of a change in personnel—was that a channel existed, represented by an individual in Derry, there was a vehicle or conduit under which British intelligence services acting on behalf of the British government was in a position to say things to him by which Sinn Fein could test if they wished what we meant by some action by us. The circumstances suggest that the dialogue with the IRA had been going on for some time.[25]

If British intelligence was conducting covert conversations with the IRA behind the backs of British ministers, then, equally, it could have been sending secret letters to Gerry Adams, and figures like Peter Brooke and Tom King would have known nothing about them. Although this is necessarily conjecture, there is no doubt that there was a British letter to Adams and that it laid the foundations for the Good Friday Agreement. Someone in the British administration wrote it.

By the time the secret Reid-Adams pipeline to the British was opened, the Redemptorists and British intelligence were not exactly strangers to each other. During the 1980 hunger strike a Redemptorist priest, Father Brendan Meagher from the congregation’s Dundalk House, had acted as a messenger between Michael Oatley of MI6 and leading Provisionals, Gerry Adams, Martin McGuinness, and other IRA figures in Belfast, as well as the leader of the hunger-striking prisoners, Brendan Hughes. According to one account, members of the Provisional leadership who wanted a settlement to the protest[26] had initiated the contact, and some republicans believe that had Father Alec Reid been fit and healthy, he would have performed the role played by Meagher. Code-named Angel, Meagher met Oatley, code-named Mountain Climber, at Belfast airport to receive a copy of last-minute British proposals to settle the prison dispute.[27] Meagher took the document to the prison and then to a house on the Falls Road to show to Adams, McGuinness, and Danny Morrison.

Five years later, in 1986 and 1987, another Redemptorist priest, this time Father Alec Reid, would shuttle between Gerry Adams and the British government. But this time his mission was much more historic than settling a hunger strike. This time it was about ending the IRA’s ancient war with Britain.

9. “Stepping Stones”

Few twentieth-century Irish politicians have aroused such powerful passions as Charles Haughey and none whose political legacy has been so fiercely argued over, even during their lifetime. If Ireland had a Richard Nixon figure, then it was Haughey. Once banished into the political wilderness, tarnished with allegations of gunrunning, Haughey made an extraordinary comeback, rising to become leader of Ireland’s largest political party, Fianna Fail, and on three occasions Irish prime minister, or taoiseach. Like Nixon’s, his years in office were dogged by scandal and controversy until finally he was driven out of office and hounded by disclosures and allegations of corruption and venality. But just as the passage of time and reflection caused Nixon’s period to be reassessed, so Haughey’s role in the Northern Ireland peace process, a story never before fully told until now, may well place his own stewardship in a different light.

Charles Haughey had never met Gerry Adams, but the men were not entirely strangers either. They first started talking to each other, albeit indirectly, during the 1981 hunger strikes when Haughey was taoiseach and trying to negotiate a settlement of the prison protest with the British prime minister, Margaret Thatcher. The go-between was an old friend and confidant, Padraig O hAnnrachain (O’Hanrahan), whose Fianna Fail credentials were impeccable. A County Clare man, O hAnnrachain had been recruited by de Valera in the 1950s and became the great man’s private secretary. In the days of Lemass and Lynch, he headed the Government Information Service, and was responsible for ensuring that the taoiseach got the best media coverage possible. Discreet and affable, O hAnnrachain, who died in 1988, was trusted by Haughey to handle the secret phone calls to and from Gerry Adams during the turbulent spring and summer of 1981. Had Adams wished, he could have destroyed Haughey with a judicious leak about their dialogue, but the contact was never revealed and trust between the men grew.

Five years later Haughey was no longer ensconced in Government Buildings in Dublin, but he was still the leader of Fianna Fail, despite the best efforts of rivals to unseat him. In August 1986, a few months before he was once again elected taoiseach, Father Alec Reid came to see him at Kinsealy, the splendid Georgian estate owned by the Haughey family on the northern outskirts of Dublin. It was the start of a dialogue that would culminate just nine months later in an extraordinary offer of an IRA cease-fire from Gerry Adams and later in the creation of a strategy that enabled the republican leader to coax, direct, and otherwise persuade his followers to take a path that would lead to the ending of the Provisionals’ long war against the Northern Ireland state.

BY THE TIME the Provisional IRA Army Council ordered the second cease-fire of the peace process, in July 1997, a decade or so after Father Reid’s journey to Kinsealy, it seemed, as one erstwhile loyal fan put it, that “the pathological anti-Haugheyites” had won the argument about Haughey’s place in Irish life.[1] The former Fianna Fail taoiseach had been mired in one of the worst corruption scandals ever to hit Irish politics, and it seemed that it would be this rather than the achievements of office that would shape history’s judgment of him.

It was the Provisional IRA that was indirectly responsible for Haughey’s plight. Back in October 1981 the IRA was strapped for cash, and so the Army Council authorized the kidnapping of Ben Dunne, one of Ireland’s wealthiest and most flamboyant supermarket tycoons. It was a risky operation, for it was sure to attract the anger of the Irish political establishment, but despite that the IRA went ahead. The task was given to the IRA in South Armagh, and it went about the kidnapping with typical flair and careful planning. Just north of the Border at Jonesboro the IRA staged a fake car accident, and when Dunne stopped to give aid, armed men pulled him from his car. He was held for a week and then released into the hands of a local journalist. Both Gardai and the Dunne family denied reports that some £300,000 was paid to the IRA for Dunne’s life, but the suspicion persisted.

Whatever the truth, the experience scarred Dunne and in no small measure contributed to a serious cocaine habit, which finally surfaced eleven years later in a bizarre incident at the Grand Cypress hotel in Orlando, Florida, when the supermarket boss threatened to jump from the seventeenth floor. Other members of his family were appalled and angered, in particular his sister Margaret Heffernan, who ordered an investigation into the financial affairs of their supermarket chain, Dunnes Stores. The probe re vealed secret payments to Charles Haughey and led directly to the government of the day setting up a tribunal to inquire into the former taoiseach’s past financial dealings.

Further investigation by the tribunal established that Haughey’s expensive lifestyle, which included not just his ownership of Kinsealy but also an island retreat off the County Kerry coast, had been bankrolled by a small group of wealthy Irish businessmen. The tribunal calculated that the businessmen had donated £8.5 million in the seventeen years since 1979, when Haughey first became taoiseach. Ben Dunne, who gave some £2 million, turned out to be the single most generous donor. Although no evidence was ever produced at the tribunal to suggest that Haughey had returned favors to the businessmen while in office, the suspicion lingered and was strong enough for many in Irish public life, the media in particular, to consign him forever to the rogues’ gallery of Irish politics.

It was difficult, in watching the pursuit of Haughey, not to be aware that many of his enemies fell upon him with the sort of glee and venom that characterizes the patient hunter who at long last has cornered an elusive prey. For years significant sections of political life and the media in the Republic had held Haughey responsible for the creation and birth of the Provisional IRA, and there were some who had long hoped for the day when he would be brought low. It is, though, one of the great ironies of the peace process that no single Irish politician did more than Haughey to start the Provos on the path that eventually resulted in the cease-fires.

His contribution was twofold. The first was that many of the political ideas on Northern Ireland developed during his time as Fianna Fail leader were incorporated almost wholesale into the strategy worked out by Adams and Reid. In many important ways the strategy was Haughey’s, not theirs. The second contribution was that by the time he left office in 1992, the theology of the peace process had been fully worked out, and the completion of the enterprise was then only a matter of internal management and negotiation. Whatever about Haughey’s role as midwife to the Provos, there is no doubt that he was in there at the beginning of their end.

The story begins at the start of the Troubles. When the violence flared in the summer of 1969, forty-four-year-old Haughey was the ambitious minister for finance in a badly divided Fianna Fail government led by the former Cork hurling star Jack Lynch. The street violence in the North would make those divisions worse and finally propel the Southern state into what one experienced commentator later described as “arguably the most serious political conflict since the Civil War.”[2] Lynch’s cabinet was riven by personality differences and by competing ambitions, but these coincided with a deep ideological fault line. Lynch had been won over by the influential Irish civil servant Dr. T. K. Whitaker to a much more moderate policy on Irish reunification than normally was Fianna Fail policy. It was one that stressed the need for unionist consent to change,[3] a view echoed later by constitutional nationalists in the North and in particular by John Hume, the leader of the SDLP. The approach implied that partition was not the sole responsibility of the British and that an attitude of confrontation with the government in London or the unionists was unlikely to be productive. Whitaker preached the need for slow, gradual, peaceful change, a breaking down of barriers between unionist and nationalist, Protestant and Catholic, and Lynch agreed.

Not everyone in Fianna Fail accepted the Whitaker-Lynch analysis. For large sections of the party the violence of 1969 signaled an opportunity to complete the unfinished business of 1919–21 and to make right the wrong that had been done to Irish sovereignty by the imposition of partition. The principal spokesman for this viewpoint in Lynch’s cabinet and the Fianna Fail party was Neil Blaney, the minister for agriculture, whose base in County Donegal was adjacent to riot-torn Derry. It was perhaps no accident that Blaney hailed from a part of Ireland that had suffered most from the partition settlement, for the Border meant that Derry lost its natural hinterland while Donegal was stranded between two states and often felt abandoned.

Haughey seemed an unlikely member of the Blaney camp, but he did have excellent family credentials. He was born in Castlebar, County Mayo, in 1925, although his parents were both Northerners who had left Northern Ireland not long after partition. They hailed from the County Derry village of Swatragh, and his father, Sean, joined the local unit of the IRA at the outbreak of hostilities, becoming adjutant of the Second Derry Battalion. Like most Northern IRA men, he sided with Collins after the Treaty, but he appears not to have actually fought in the subsequent civil war.

By 1922 he had risen to be Brigade OC of the Second Northern Division, but then the family moved to Mayo, where Haughey Sr. took a regular commission in the new Irish army. The young Charles Haughey did not follow his father’s footsteps into the IRA, but he did the next best thing and joined Fianna Fail. An accountant and lawyer by training, Haughey was first elected to the Dail in 1957 and four years later was promoted to the post of minister for justice. Within a month he had dealt the final blow to the IRA’s dying Border Campaign, when he introduced special military courts to try republican activists. The secretary to his department, Peter Berry, later credited him with having made the move that “broke the back of the organisation.”[4] That and the fact that only once before the crisis of 1969 had Haughey mentioned Northern Ireland in a speech[5] led many to conclude that he held no noticeably strong views on the national question. Haughey had made a bid for the Fianna Fail leadership when his father-in-law, Sean Lemass, retired as taoiseach but withdrew when Lynch was adopted as the compromise candidate acceptable to most tendencies in the party. But such was the impact of the violence of August 1969 in the Republic that the leadership question seemed ready to be reopened.

The violence forced the Lynch government to make some sort of response; the television scenes of blazing Catholic houses and refugees flooding across the Border had roused public opinion in the Republic. Lynch managed to resist efforts to send the Irish army across the Border but agreed to set up a special cabinet committee to relieve “the distress” suffered by Northern nationalists, whose control he placed in the hands of his enemies, Blaney, Haughey, and another cabinet hawk, Kevin Boland. As minister for finance, Haughey was charged with administering a fund of £100,000 to spend on the North, a considerable sum in 1969–70. Haughey and Blaney also liaised with Captain James Kelly, an officer in Irish military intelligence who was mixing with Northern republicans and nationalists on behalf of the government.

From these circumstances came two extraordinary allegations. One was that Haughey, Blaney, and others, including a senior Belfast IRA man, John Kelly, had tried to import weapons from Europe to supply defense committees in Belfast. The allegation was all the more staggering given the size and quality of the arsenal supposedly involved: 200 submachine guns, 84 light machine guns, 50 general-purpose machine guns, 50 rifles, 200 pistols, 200 grenades, and 250,000 rounds of ammunition, enough to equip a battalion of the British army.[6] Since the defense committees were under the sway of either the Official IRA or, in more cases, the new Provisional IRA, it meant the weapons were headed for republicans who would not hesitate to use them, and not necessarily in a defensive mode.

Haughey, who along with Blaney was later sacked by Lynch when the affair became public, was charged with conspiracy to import weapons. The subsequent hearing, known simply as the Arms Trial, was one of the most sensational in Irish history, but it ended with Haughey’s acquittal, while the charge against Blaney was dropped at an earlier stage. Although spared shame and imprisonment, Haughey was banished to the Fianna Fail equivalent of a political Siberia. John Kelly was similarly acquitted.

The other allegation was that, along with Captain Kelly and Blaney, Haughey had conspired to foment a split in the IRA with a twofold aim: to neutralize or weaken the politically radical and increasingly violent Gould-ing leadership in Dublin, while creating an instrument in the North that would be more amenable to Fianna Fail control. Cabinet papers of the day that have recently been published acquit Haughey of this particular charge; they reveal that this was a policy agreed upon by all Lynch’s ministers in April that year, long before the August riots. The papers show that the Department of Justice had recommended a policy of dividing the IRA’s rural conservatives from the urban radicals and that the cabinet endorsed this. Even so, the working out of the policy put Haughey and Blaney at the center of the scheme, almost as if it was their private, freelance plan.

In the wake of the Arms Trial, Haughey was dispatched to the political wilderness, but he turned this into an opportunity. He cultivated the grassroots and discovered that the rank and file responded well to a tough republican rhetoric. In 1977 Fianna Fail won a thumping majority in the general election, unseating the strongly anti-IRA Fine Gael–Labour coalition. Lynch was obliged to include Haughey in his cabinet, making him minister for health and social welfare. Within two years Lynch was gone, forced to quit by a series of by-election reverses, and Haughey narrowly won a hard-fought leadership contest with the retired taoiseach’s preferred successor, George Colley.

The day after his election as Fianna Fail leader, on December 7, 1979, Haughey signaled a new assertive direction regarding Northern Ireland matters, an end to Jack Lynch’s tractable approach to London. Not only did the new Irish leader flag a more confrontational line with the British, but he hinted that the way forward was for the two governments to bypass the unionists altogether and deal directly themselves with a shared problem. The government’s line, he announced, would revert to the traditional Fianna Fail program that called on the British to encourage Irish unity and to declare their commitment to implement “an ordered withdrawal” from the North.

His first Fianna Fail Ard Fheis, in February 1980, gave an opportunity to signal the changes in a way the ordinary grassroots would more readily understand. Haughey was piped into the Royal Dublin Society’s auditorium, not far from the British embassy in affluent Dublin 4, by a trade union band to the air of “A Nation Once Again,” the hymn of nationalist Ireland. He followed that with a speech that for the first time rejected the traditional goal shared by the British and the Irish governments of securing an internal settlement. Northern Ireland, he told the wildly cheering delegates, had “as a political entity, failed and… a new beginning was needed.”[7]

ACCOMPANIED BY cabinet colleagues, Haughey traveled to Downing Street for talks with the new British prime minister, Margaret Thatcher, in May 1980. This teapot diplomacy, so named after the silver-plated gift that Haughey brought as a present for Thatcher, launched the Anglo-Irish approach to Northern Ireland. Although it was to culminate five years later in the Hillsborough agreement, the diplomacy also contained the germs of ideas that would re-emerge during the peace process dressed in Sinn Fein garments. In a Dail debate after the Downing Street summit, for example, Haughey obliquely introduced the idea of an all-party conference if the British were to make a declaration expressing their interest in ultimate Irish unity. “If that interest were declared,” he told TDs, “we could then start working together, the Government here, the British Government and Irishmen of every tradition in the North towards a solution which will guarantee permanent peace and stability.”[8]

The proposal for an all-Ireland conference was hardened up between 1983 and 1984, during sittings of the New Ireland Forum in Dublin, a body set up by constitutional nationalists to meet what was widely perceived to be the burgeoning threat from Sinn Fein’s growing electoral strength in the North and its alarming potential to cause political instability in the South.

In his contribution to the forum report, Haughey made clear that in his view the way to achieve unity would be for Britain to respond positively to proposals for an all-party conference at which unionists and nationalists would thrash out the institutional details and constitution for the whole island. Sinn Fein could attend, but only if the IRA ended its violence and the party renounced violence. This was one of the central proposals contained in the later secret British correspondence with Adams. It had been borrowed wholesale from the Fianna Fail leader and made British policy, at least in the private diplomacy with the Sinn Fein leader.

But that was not all. During one private, unreported session of the forum in October 1983 Haughey produced a formulation on the crucial issue of unionist consent whose ambiguity was later borrowed almost entirely by Gerry Adams. Its significance was that it opened the door to a reinterpretation of the consent principle that would allow Adams to accept the unionist right to self-determination and thus to accept, if only in a de facto sense, the existence of the Northern Ireland state. In his address to the forum, Haughey said, “As regards the veto—the constitutional guarantee [to unionists]—we should make a clear distinction on consent. Consent is only applicable to arrangements on a new Ireland. But consent by the Unionists to British action to find a solution is not required.”

By the end of 1984 Haughey had articulated three of the principal features of what would eventually become the Sinn Fein peace strategy. His denunciation of Northern Ireland as a “failed entity,” his call on the British to declare their interest in seeing Ireland united, and his proposal for an all-Ireland constitutional conference were all vital ingredients in the mix.

But it was Haughey’s assertive style with the British that gave his politics the color that Adams and Reid needed. In 1982, for instance, he had refused to endorse Britain’s military expedition against the Argentinean invaders of the Malvinas, or Falkland Islands, and when the SS Belgrano was sunk he reasserted Irish neutrality at the United Nations, much to the fury of Mrs. Thatcher and the delight of Irish republicans, of the Provisional as well as the Fianna Fail variety.

His opposition to an internal Northern Ireland settlement was given tangible expression that same year when he appointed the SDLP deputy leader, Seamus Mallon, to the Irish Senate, even though British law barred Mallon from sitting in the locally elected Stormont Assembly at the same time. An embarrassed British administration was forced to expel Mallon from its own parliament. He publicly accused the Duke of Norfolk of being a British spy and regularly lambasted the Fine Gael leader Garret FitzGerald for a weak-kneed attitude to the Thatcher government, on one occasion accusing him of colluding and secretly collaborating with the British on security matters.

A year after the Hillsborough agreement had been signed, Haughey was the first mainstream politician to complain that the deal had not only failed in its declared objective of improving the lot of Northern Nationalists but that “the position of Nationalists in the North has, in fact, seriously worsened.” Whatever the private thoughts of Gerry Adams and his closest colleagues about the Hillsborough deal, this was an echo of what they too were saying in public.

By December 1986, two months before Haughey was to take power for the fourth and last time, Adams was able to tell Fianna Fail’s newspaper, the Irish Press, that he regarded the taoiseach-to-be as “a genuine Nationalist.”[9] Haughey’s role during the 1981 hunger strike—an accessory along with Fine Gael’s Garret FitzGerald “to the legalised murder of ten true and committed Irishmen,”[10] according to the IRA prisoners’ statement at the end of the protest—was forgotten.

Haughey’s forceful brand of Irish nationalism appalled significant sections of Irish political society, but it was precisely this quality that allowed Adams and Father Reid to contemplate advancing the fledgling peace process in the mid-to late 1980s. Haughey was the right man, in the right place, at the right time. Had any other recent Fianna Fail leader, from Jack Lynch onward, been taoiseach at this time, it is likely the project nursed by Adams and Reid would have been stillborn.

HAUGHEY’S LEADERSHIP of the Irish government and his forceful brand of nationalist politics made possible the next seminal moment in the Irish peace process, one kept a tightly guarded secret until now by all the participants, not least because once again, as with the secret British diplomacy, it signaled not just the private desire of Gerry Adams for an end to the IRA’s violence but his willingness to accept a settlement that fell short of what most republicans would find acceptable, from the Army Council to the Volunteers on the streets of Belfast, Derry, and South Armagh.

So it was that sometime in the second week of May 1987, the editor of the Irish Press, Tim Pat Coogan, delivered a lengthy written message from Father Alec Reid to Charles Haughey, containing Gerry Adams’s terms for an IRA cease-fire. It was a remarkable document whose message said not just that an opportunity existed to call an IRA cease-fire but that if the enterprise was handled properly, the proposal brought by the Redemptorist priest on behalf of the Sinn Fein leader could mean taking the gun out of nationalist politics forever.

As Reid explained the initiative, it became clear that he was communicating not just on his behalf or on Adams’s but on behalf of the Irish Catholic Church as well. The enterprise, he explained, had been endorsed by the Redemptorist order in Ireland and by other senior church figures, most notably Cardinal O Fiaich, who would give political cover to Haughey if the dialogue were ever publicized. It had been developed after lengthy thought and debate, Reid said, which involved church figures talking widely to key figures in the nationalist and unionist communities. The message the church had received, he said, was the same from everyone. As the body with the required resources, influence, and access to power centers, it should and would try to get this peace initiative under way.

Since the church’s aim was primarily to bring the IRA’s violence to an end, it had made a priority of approaching Gerry Adams, Reid explained. Adams also had told them that the church could play an important role in finding the ways and means of ending the armed struggle but that there was no chance of persuading the IRA to do this unless there was unrestricted dialogue with fellow nationalists aimed at formulating a strong political alternative to the armed struggle involving all Ireland’s nationalist parties. Only in that way, Adams had told them, could the peace initiative succeed.

This was the kernel of the Reid-Adams initiative: the creation of a pan-nationalist axis whose political clout would be greater than the IRA’s and sufficiently strong to persuade the hard men of the organization to lay down their weapons. For this reason Haughey was being approached in his capacity as Fianna Fail leader, Reid explained, not just because he was taoiseach. The dialogue, if it ever happened, would initially be between Sinn Fein and Fianna Fail, with the church acting as a neutral host and facilitator. Even the start of such a dialogue, Reid suggested, would be sufficient to influence the course of the armed struggle, and once it was under way, he believed, the IRA would respond and cooperate positively.

Sinn Fein, he said, had consistently told him that republicans would fully cooperate with the church in the search for what became known as “the alternative method” to armed struggle. They, or rather Gerry Adams, had also given a number of pledges, should the dialogue start. Adams would not insist on any preconditions either officially or unofficially and promised that all their dealings would be treated in the strictest confidence; there would be no question of Sinn Fein’s leaking embarrassing details for political advantage. The peace initiative would be given the highest priority by the Sinn Fein leader, the necessary resources and energy would be devoted to it, and Sinn Fein promised to “engage constructively”[11] with anyone the church decided should be involved in the dialogue.

Reid went on to tell the Fianna Fail leader that Sinn Fein would accept any settlement that came out of a properly structured process, even if it fell short of its stated aim of achieving a socialist Ireland. The goal of the IRA’s armed struggle, he had been told, was not the creation of a thirty-two-county socialist republic; rather, it was to establish the right of all the Irish people to decide their own political future in free negotiations, the so-called right of national self-determination. The goal of socialism was a political one that Sinn Fein would pursue by political methods only.

This was an important and revealing concession. The move to the left during the 1970s had been constructed by Adams and his allies in Cage 11 as part of a strategy to capture the movement. By 1987, however, the need to make an alliance with Fianna Fail dictated a compensating move in the other direction—socialism would be discarded in the interests of advancing this new Adams strategy. The underlying message to Haughey and others involved in the process was that pragmatism would determine the flow of events, not dogged adherence to political ideology.

Adams did have certain bottom lines in negotiation, however, Reid explained. The most crucial was that nationalists and unionists should be able to negotiate their political and constitutional future free of dictation from the British, and as long as this principle was honored, Sinn Fein would be content with whatever shape the agreed settlement took.

It was at this point that the Reid-Adams enterprise dovetailed with the secret British correspondence with the Sinn Fein leader. The British had told Adams that while physical withdrawal from Northern Ireland was out of the question, they would promise not to interfere or dictate the terms of any settlement reached by unionists and nationalists; they would withdraw politically, in other words. It was all very Jesuitical, but by the late spring of 1987 Adams was approaching Haughey with the very same idea and sought his cooperation in making this a public and formally acknowledged policy of the British government.

Adams wanted the British to implement this promise, Reid said, in a practical fashion by declaring in some convincing way their willingness to set aside section 75 of the Government of Ireland Act (1920), the legislation that formalized the partition settlement. This section gave the British parliament supreme authority over Northern Ireland, in effect saying that Britain could veto the wishes of the people of Northern Ireland even if their representatives reached political agreement. As long as a majority favored staying in the union, this was not an issue, but theoretically it could be used to thwart a vote in favor of Irish unity.

This was to be the revised and reinterpreted version of British withdrawal, the centerpiece of the Reid-Adams initiative: a declaration of constitutional disinterest and a promise not to dictate or influence the outcome of the all-party talks. If the British withdrew the legal claim of veto over the people of Northern Ireland, in other words, then Sinn Fein would be content to view this as the culmination and realization of the IRA’s goals, for which, by that time, it had waged some seventeen years of killing, shooting, and bombing.

In practice, Reid said, Sinn Fein would accept a promise from the British to repeal the offending section of the 1920 act after agreement had been reached in negotiations between unionists and nationalists, not before. The traditional republican model for British withdrawal foresaw the all-party conference happening once the British had at least declared their intention of leaving Northern Ireland. In Adams’s new version, the conference would happen first, and then the British would make good their declaration.

The Sinn Fein leader had added an important qualification to all this, Reid told Haughey. The British did not have to withdraw immediately or even in the foreseeable future. In fact Sinn Fein wanted the British to remain, and would insist that they did, in order to oversee the practical details of implementing any agreed settlement.

Then came the central message from Reid to Haughey: if the British agreed to do this, the IRA would declare a cease-fire. The IRA’s long war would come to an end, and in a much more definitive fashion than any of the short-lived and doomed cease-fires of the 1970s.

Once representatives of nationalist and unionist opinion were able to meet freely without a British veto hanging over their heads, then, as far as Sinn Fein was concerned, all options for a settlement of the national question would be on the table. Sinn Fein would at that point abide by negotiating arrangements that would guarantee unionists their heritage and culture in accordance with their right of consent and the right of consent of the Irish people as a whole. This was Gerry Adams’s way of recognizing the integrity of the Northern Ireland state and of opening the way to Sinn Fein eventually helping to govern it.

IT WAS AN EXTRAORDINARY offer. The Reid-Adams principle married the separate and conflicting concepts of all-Ireland and Northern Ireland–based self-determination, and in practice that would mean accepting the outcome of separate referenda held in the two divided parts of Ireland, albeit on the same day, which is precisely what happened in the weeks following the signing of the Good Friday Agreement in April 1998, eleven years after the astonishing offer to Haughey was made. Traditional republican theology held that self-determination had to mean the Irish people voting as one unit; in the message from Father Reid, Adams was saying that he would accept two separate votes and, crucially, that they would have equal value. If the North voted differently from the South, then he would accept this. It was setting established republican dogma on its head, and substituting heresy for orthodoxy.

Although senior republicans like Gerry Adams continue to deny to this day that they have conceded the right of a majority in the North to maintain the link with Britain, in practice the unionist majority,[12] the subtle, ambiguous formula devised by Father Reid and the Sinn Fein president meant that they had. Sinn Fein, under Adams, was a little like a team of soccer players who strongly object to the offside rule but agree, nevertheless, to play in the cup final.

There was an important sense in which the proposal was based on fantasy. It assumed that the British would want to frustrate the will of the people of Northern Ireland, even if they did vote themselves into a united and independent Ireland. It was clear from a variety of public statements over the years of the Troubles that they would do no such thing. They had said as much in 1973, before the Sunningdale agreement, and again in 1985 when the principle of consent, both for the union and against it, had been written into law when the Hillsborough Agreement had been signed by Margaret Thatcher and Garret FitzGerald. But the great achievement of Adams and Reid was that they had fashioned a wording that enabled Sinn Fein to join the ranks of constitutional nationalism while preserving the outward aspects of the party’s traditional uncompromising brand of republicanism. As an instrument for keeping the Pro visionals united and the rank and file unaware of the true implications of the peace process, it was almost perfect.

All this was a preamble to the core proposal that Reid had brought from Belfast. It was for what the rest of the world would soon call a pan-nationalist front, an alliance of Irish nationalist parties, North and South, that would try to agree a common policy on the North. What this meant, said Reid, was that the three principal parties—Fianna Fail, the SDLP, and Sinn Fein—must agree at a minimum to press the British to remove their right of veto from the Government of Ireland Act.

Reid outlined two ways in which the enterprise could be advanced. Either the church could host face-to-face talks between Sinn Fein and Fianna Fail, whose confidentiality the church would underwrite, or the church could mediate between Fianna Fail and Sinn Fein in a bid to devise ways in which direct dialogue between them could begin.

There was one immediate problem with the Reid-Adams proposal. It was not entirely clear in what circumstances the IRA would declare a cease-fire. Reid seemed to outline two scenarios for an end to armed struggle—one when the British agreed to remove their veto from the Government of Ireland Act, the other when the pan-nationalist alliance was formed. It was clear that the chronology outlined by Reid-Adams envisaged the latter happening first, but just when would the IRA cease-fire start?[13] It was a loose thread in an otherwise finely woven garment.

This, in its essentials, was the blueprint of the Irish peace process, and with variations and amendments, and not necessarily in the order originally envisaged, it was the plan that was eventually implemented.

It would be quite wrong to say that with this message from Alec Reid the peace process was full-fledged, but its shape had been sketched out, a road map of sorts indicated, and the parameters of the historic but still secret ideological compromise that Gerry Adams was prepared to make clearly identified. There was a great deal of distrust to overcome, much of it caused by the IRA’s post-Eksund campaign of violence. Like the British, the Haughey government noted the gulf between what Gerry Adams and Alec Reid preached and what the IRA practiced.

At the time, according to one informed source, it never occurred to Haughey or his advisers to ask Adams if he had the backing of the Army Council for the initiative. Like the British, they assumed Adams would never take such a dangerous course unless he had the informed support of his leadership colleagues. Had they known that the process was, at this stage, as much a secret from the IRA leadership as it was to everyone else, the Reid enterprise might well have suffered an early and precipitate ending. But as it was, the question was never asked.

The process was full of difficulties, not least the need to overcome Irish and British government doubts about the IRA’s real intentions. One great obstacle was Margaret Thatcher. Few of those involved believed that movement along the lines sketched by Reid-Adams would be possible as long as she was British prime minister, as one of Haughey’s key aides, Martin Mansergh, was to write later: “I was always fairly clear that there was little hope of an end to belligerence in Ireland while she remained British prime minister.”[14] He and everyone else would have to wait until November 1990 before she left the political stage. But the important point about Father Reid’s overture to Haughey was that it contained an unmistakably strong signal that Adams knew where he wanted to go, and even if the how was still a little hazy, the ending of armed struggle by the IRA was clearly now on the agenda.

Faced with two options by Father Reid—one of direct dialogue with Adams, the other indirect dialogue facilitated by the Redemptorist order— Haughey chose the safer and, as he had in 1981, chose to talk to Adams indirectly. With the history of the Arms Trial hanging over him, Haughey calculated that the risk of exposure was too great and that the slightest leak could destroy him. He kept the Reid approach secret from cabinet colleagues and particularly from his coalition partners, the Progressive Democrats, who were led by his deadly enemy, Des O’Malley, a determined foe from the Arms Trial days.

Haughey’s decision caused the first crisis in the enterprise. Adams made a desperate plea to Haughey for face-to-face meetings, but to no avail, as one well-informed source told the author. “Adams was looking for [face-to-face talks] on the basis that if he was going to go to the IRA—and he said he would go to individual active-service units if necessary—he would have to be able to tell them that he had looked Haughey in the eye and that Haughey had assured him that this would be the policy of the Irish government.”[15] Cardinal O Fiaich even offered a room at Maynooth College, Ireland’s principal seminary, where the two men could meet discreetly. But the proposal was too risky, and Haughey declined. The prospect of a speedy IRA cease-fire receded.

Known as An Sagairt, the Irish word for priest, in the code language worked out by Adams and the Redemptorist priest, Reid had come to see Haughey armed with a detailed, worked-out strategy, which he called “A Concrete Proposal for a Political Strategy for Justice and Peace.”[16] Spread over three documents, “A Concrete Proposal” drew together all the various strands of the secret discussions between the Reid group, Sinn Fein, and the British, as well as the ideas developed by Haughey.

One paper set out the six fundamental principles that would underpin the strategy. Two of these crucially redefined the principle of Irish national self-determination to embrace the need for unionist consent. The idea behind Reid’s proposal was that if the project went ahead, then the three nationalist parties would agree to subscribe to the principle. The real significance of this is that it meant that Sinn Fein and the IRA would agree to abide by something that republicans had traditionally abjured and waged war to resist. Another key principle encapsulated the new definition of British withdrawal. By agreeing to this, republicans would formally abandon the IRA goal of ejecting Britain from Northern Ireland by force.

A second document suggested twelve “stepping stones,”[17] as they were called, toward an agreement among Irish nationalists on a peace strategy.

These were as follows:

• An agreement in principle that there should be a joint Nationalist strategy

• An agreement in principle that peace can only come with the free, independent and democratic consent of the Irish people

• An agreement in principle that the aim of the strategy should be to design and create a New Ireland with a new Constitution

• An agreement in principle that the Irish people as a whole should design this new Constitution in unfettered dialogue amongst themselves

• An agreement that the Irish people consist of two traditions, Unionist and Nationalist

• An agreement in principle that consent must be two-fold in nature, requiring agreement from both Unionists and Nationalists

• An agreement in principle that this two-fold consent can only be achieved by political dialogue

• An agreement in principle that the framework for dialogue would be a Constitutional Conference that would sit on a semi-permanent basis until final agreement about Ireland’s future had been reached. Membership of the Conference would be through direct election and it was hoped that both Unionist and Nationalist parties would attend. The British, the Irish government, Sinn Fein, other Irish Nationalists and the Unionists would agree the arrangements for the Conference.

• An agreement in principle that the British would withdraw from the central decision-making process in Northern Ireland

• An agreement in principle amongst Sinn Fein and the other Nationalist parties that they would agree to try to persuade the British to make a declaration containing five points:

(i) That they will set aside the 1920 Government of Ireland Act when agreement is reached in the Constitutional Conference;

(ii) That they will say they have no selfish interest in remaining in Ireland

(iii) That they will facilitate the Constitutional Conference;

(iv) That they will not interfere in or dictate to the Conference;

(v) That if there is an agreement then the British will implement it in law.

• An agreement in principle that an advisory committee representing the leadership of the main Nationalist parties be set up to examine how best to implement and propagandise this strategy and win support for it in Ireland and abroad

• An agreement in principle, although couched in more discreet language, that Sinn Fein would try to win approval for the strategy from the IRA.[18]

All twelve “stepping stones” would in time be incorporated into the peace process, though with variations and modifications. The British declaration, for instance, became a British-Irish joint declaration after it grew clear that the British would make no promises about the fate of the Government of Ireland Act until the Irish agreed to amend articles 2 and 3 of Eamon de Valera’s 1937 constitution, which formally expressed the Republic’s territorial claim over Northern Ireland.

The leaders’ advisory committee would, in later documents, become first a nationalist “Convention,” which would advise on the steps necessary to achieve democratic self-determination, and then finally the Forum for Peace and Reconciliation, set up in Dublin after the 1994 IRA cease-fire. An entirely nationalist body, the forum was designed to introduce Sinn Fein to the civilities of constitutional politics.

The problem of how Charles Haughey should proceed with the dialogue with Gerry Adams yet keep him at arm’s length was a difficult one, but the solution was eventually found in the structure of the strategy itself. Reid and Adams wanted to involve the main Northern nationalist party, the SDLP, and in particular its leader, John Hume, in the enterprise, and it was through Hume that Haughey’s difficulty was overcome.

At the same time that Reid had sent his lengthy message to Haughey, in May 1987, he dispatched a similar document to the Derry office of the SDLP leader. Hume responded cautiously. He had done one favor for Reid before, according to Tom King, when he had pressed the then NI secretary to give a private assurance of Britain’s neutrality to Adams, but the Reid letter was the first formal request to Hume to involve himself fully in the process.

Hume met Reid several times, first in June 1987 and then in August and again in September, before agreeing in December 1987 to meet Adams. Like the Fianna Fail leader, Hume was cautious about supping too close to the Provos. As all this was happening, Haughey was putting together his own alternative to the face-to-face dialogue demanded by Adams. The proposal was simple: he would suspend his direct involvement and instead nominate John Hume to represent him and to act as a sort of go-between for the Irish government. This would solve the problem of having to meet Adams directly while preserving the initiative. Adams protested angrily, but he had little choice. He agreed, on one condition, to which Haughey readily assented: the communication line between him and Haughey through Father Reid would remain active, while Haughey would promise to reinvolve himself if and when necessary.[19] There was another condition that Haughey insisted upon. The dialogue between him, Reid, and Adams was to be kept hidden from the SDLP for fear that it would be leaked and Haughey ruined. Hume was deliberately not told about the contacts and for long afterward believed that his subsequent meetings with Adams marked the start of the peace process. In retrospect this was a key moment, for it gave the peace process an identity it did not entirely deserve. Thereafter the process would be known as the Hume-Adams process, and with that the SDLP leader was set on the path that would eventually lead to a Nobel Peace Prize and international acclaim. The choice of Hume as Haughey’s representative had an unexpected bonus. The SDLP leader’s association with the Reid-Adams enterprise gave it a level of acceptability it otherwise would have lacked, especially in the Republic, where Hume’s standing was exceptionally high. Haughey began the Irish part of the peace process, but Hume gave it respectability.

It was at this point that Haughey brought in his talented Northern adviser, Martin Mansergh, a former diplomat in the Department of Foreign Affairs whom Haughey had made his principal point of communication with Northern parties. There then began a partnership that was to last for more than a decade in which the Redemptorist priest would carry written and oral messages between Adams in Belfast and Mansergh in Dublin, who would in turn analyze and pass them on to his political master, Haughey at first and, in later years, Albert Reynolds and then Bertie Ahern.

Much to the anger of the unionists and the great surprise of the rest of the world, the first of a series of talks between Sinn Fein and the SDLP took place on January 11, 1988, at Clonard Monastery. It was the public working out of the secret agreement arrived at by the nationalist leaders. For two hours Hume and Adams discussed and agreed arrangements for wider negotiations involving delegations from each side. The meeting had been requested by “a third party interested in creating political dialogue,” the two men said in a veiled reference to Alec Reid. As always the Redemptorist’s role would remained cloaked in shadow.

In March the delegations met at the Redemptorists’ retreat house, St. Gerard’s on Belfast’s Antrim Road. Adams was accompanied by three key allies—Danny Morrison, a fellow Army Council member and director of publicity for both the IRA and Sinn Fein; Mitchel McLaughlin, a veteran Sinn Fein member from Derry; and Tom Hartley, a former Sinn Fein press officer from West Belfast and a driving force behind the bid to politicize the Provisionals. It was a carefully balanced delegation. Adams and Morrison represented the Army, but the image presented by McLaughlin and Hartley to the outside world and certainly to the SDLP was of figures who had spent their republican careers entirely in Sinn Fein with no or next to no involvement in IRA matters.

Hume came with his own top team: Seamus Mallon, his County Armagh-born deputy; the university lecturer Sean Farren, one of Hume’s closest confidants; and Austin Currie, a somewhat independently minded figure from County Tyrone who could trace his political genesis back to the old Nationalist Party. There were five more meetings that year, in March, May, June, July, and September. One meeting was a private head-to-head between Hume and Adams; otherwise the party delegations were involved. Haughey agreed to send Mansergh and a junior TD, Dermot Ahern, to low-level meetings with Sinn Fein, which took place in the Redemptorist house in Dundalk with Father Reid present. Two such meetings happened, one in May, the other in June.

With hindsight the curious feature of the talks is the extent to which they were conducted in public. In sharp contrast to the secret diplomacy that had occurred from 1982 and 1983 onward, each session of the 1988 Sinn Fein–SDLP talks was known about almost as soon as it had taken place, courtesy of detailed briefings to the media. After the first meeting, both Hume and Adams readily agreed to be interviewed by the author and other journalists. Adams, for instance, used the interviews to call speculation that a cease-fire was on the agenda “mischievous and erroneous” and denied that Haughey knew about the meeting beforehand or that Hume had come armed with what he termed “concrete proposals.”[20] In April documents that had been exchanged by the parties were leaked at a high level to the author and published in the Sunday Tribune.[21] In September, when the talks officially concluded, both sides made their respective papers public, and these were reproduced in full in the Irish Times.[22]

The structure of the talks raised questions about the reason for the publicity and whether more would have been achieved had the two parties kept their dialogue secret, as they very soon did. Not least of the advantages of maintaining secrecy was that both Hume and Adams would have avoided the very extensive unease that emerged in their respective parties as a result of the publicity surrounding the meetings.

The Sinn Fein–SDLP talks did, however, make one significant achievement; activists in both parties became accustomed to the notion that once deep and bitter enemies could sit around the same table and talk. The dialogue also managed to introduce into nationalist and republican political discourse many of the key concepts and language of the peace process. Seen as a conditioning process, the 1988 talks between the SDLP and Sinn Fein make a great deal of sense.

ALMOST AS SOON AS the public talks ended, the private dialogue between Adams and Hume started, and this continued in secret for five years, until one Saturday morning in April 1993, when Gerry Adams and Father Alec Reid met John Hume in his house in Derry, on the edge of the Bogside and in circumstances that almost guaranteed that someone would see them. They were spotted and the story was made public. A statement subsequently issued by the two men used the precise language of Reid’s “Concrete Proposal.” They were engaged, they said, “in a political dialogue aimed at investigating the possibility of developing an overall political strategy to establish peace and justice.”[23]

Although hopes were high that progress could still be made, the reality was that the year 1988 had seen the process falter. The promise of an IRA cease-fire held out in Father Reid’s first messages to Haughey and to Hume in the spring of 1987 had not been fulfilled. The intensity of the IRA’s military campaign was such as to deter the British from taking any more risks, while in nationalist Ireland the conviction grew, as Martin Mansergh was to say later, that the Provisionals still had some distance to travel. “Both the SDLP and Fianna Fail,” he wrote, “formed the view separately that northern republicans were not then ready to end their campaign, and that the primary aim of any continuing dialogue was to end their political isolation and build a broad front.”[24] Neither party was prepared to accept that outcome, he added.

ALTHOUGH CONSIDERABLE progress had been made in advancing the theology and in shoring up the political foundations of the peace process, the truth was that Adams and his supporters still had an enormously long way to go before they could persuade, maneuver, or cajole the IRA to accept a cessation. None of the mainstream nationalist parties were aware that the Army Council had been kept in the dark about much of the process prior to the talks with the SDLP or that the details of the Reid-Adams proposals were still a secret to all but Adams and his close advisers. It had taken Adams and his supporters in the Provisionals five years to overthrow and remove the O Bradaigh–O Conaill leadership and to move the Provisionals into electoral politics. It would take a lot longer to end the armed struggle.

The peace process did not quite end when the SDLP and Sinn Fein concluded their public dialogue. Not only did Hume and Adams continue to meet regularly and in secret—the first meeting, at the invitation of the Re demptorist superior in Ireland took place in the order’s Dublin house within weeks of the official ending of the dialogue—but in October an extraordinary event by the standards of 1988 took place at a private conference in the German town of Duisburg in the Ruhr valley.

Four of Northern Ireland’s senior politicians were there. Peter Robinson, the shrewd, ascetic deputy to the Reverend Ian Paisley, represented the hard-line DUP; Jack Allen, a gregarious publican from Derry, came from the largest unionist group, the Ulster Unionists; Austin Currie came from the SDLP; and Gordon Mawhinney, the deputy leader of Alliance, represented the moderate, middle-of-the-road liberals. The weekend conference had been organized by Irish and German churchmen in an attempt to get peace talks going back in Belfast. The meeting had the blessing of the party leaders, and hopes for a breakthrough were high.

What made the gathering doubly significant was the attendance of Father Alec Reid, who participated as the representative of Gerry Adams, albeit an unofficial one. All the participants knew beforehand that he was to be there as someone familiar with the views of Sinn Fein, and so did their party leaders. His presence kept the unionists in terror of discovery for months afterward. At that time, talking to Sinn Fein was possibly the most unforgivable sin in the book, and even being in the same room as a surrogate figure like Reid would have caused an outcry.

At that stage the prospect of participating in political talks did not concern Sinn Fein or Gerry Adams—the IRA’s raging war ruled that out. But Reid was there to give the unionists a highly important message, as one account of the weekend conference recorded:

Alec Reid spoke of the two traditions and how essential it was to enter into a dialogue with the right spirit. It was a Republican tradition to use violence but if we do [sic] enter dialogue he was certain that we would find that they were not so difficult. In his view the Republican Movement must be dealt with—it represents the views of the Nationalist people. They are pursuing the right of the Irish people to determine their own future and would pursue that right by political means if given reasonable opportunity.[25]

As a political event the Duisburg conference was a sideshow, but its significance was twofold: the Sinn Fein leadership was signaling a willingness to join the mainstream, while the readiness of unionists to accept Father Reid showed that they were not entirely appalled at the prospect. Everyone knew that if Sinn Fein wanted to become constitutional politicians, the IRA’s campaign had to end.

The peace process hadn’t stopped, but it had slowed down, and further movement was spasmodic. Hume and Adams met only four times in 1989, as the IRA’s post-Eksund “Tet offensive” worked its way through failure and one setback after another. In February 1989 Haughey made another bid to advance the “stepping stones” strategy when he told the Fianna Fail Ard Fheis that an end to the IRA’s violence could dramatically change things. Using language that could have been taken wholesale from the Reid document, Haughey said that such a development

would open up, as it did in the New Ireland Forum, the possibility for a broad consensus among Nationalists on how to achieve political stability based on justice. Our efforts, supported by a large majority of Irish people everywhere, could then be constructively directed to persuading our Unionist countrymen that their future lay with us in a partnership of equals and in convincing the British government that the future of Ireland could and should be left to all the Irish people to decide for themselves.[26]

This was the Reid-Adams formula for national self-determination, almost word for word.

Later that year Hume incorporated Reid’s twofold consent formula into a practical proposal, suggesting to his party conference in November that any political settlement agreed by the Northern parties and the two governments should be put to the whole Irish people in twin referenda, held in the separate jurisdictions on the same day. Such a proposal, which Hume had first floated way back in 1981, would mean, he said, that from a nationalist/republican viewpoint “for the first time the people of this island would have expressed self-determination on how we live together.”[27] It would also remove, at a stroke, the basis for the IRA’s opposition to the 1921 settlement, rooted as that was in the fact that the unionists had flouted the principle of all-Ireland self-determination.

It was around the same time that the Northern Ireland secretary, Peter Brooke, chose to make his neutrality speech, directing it at Gerry Adams and his allies from the heart of his constituency in Westminster. Such a public declaration had been the missing part of the secret British correspondence, but now Brooke was rectifying that omission. Offering Sinn Fein a full political role in Northern Ireland’s affairs if the IRA’s violence ended, Brooke tackled two of the issues raised in Alec Reid’s “stepping stones” document. “The British Government,” he said, “has no selfish strategic or economic interests in Northern Ireland; our role is to help, enable and encourage. Britain’s purpose, as I have sought to describe it, is not to occupy, oppress or exploit but to ensure democratic debate and free democratic choice.”[28] Reid had said the British would need to declare their neutrality and agree to sponsor a dialogue with which they would not interfere. Peter Brooke had delivered. Slowly the “stepping stones” were being laid.

ALTHOUGH THE PEACE PROCESS was moving forward, its speed was not much faster than a glacier’s, and nearly everyone agreed on the reason for that. As long as Margaret Thatcher was British prime minister, Adams could never deliver the IRA leadership or grassroots. IRA supporters would condemn even the suggestion of a deal with Thatcher as surrender. Nor could the participants be confident that a declaration of the sort outlined in the Reid-Adams proposal from Thatcher’s lips would carry any weight in the Provo heartlands. The hatred between them was just too deep. Suddenly, three weeks after Brooke’s “neutrality” speech, the logjam was removed and the glacier picked up speed. An internal Tory Party crisis forced Thatcher to resign, and with her went the last, great obstacle to an IRA cease-fire.

Thatcher’s departure revived the process. Her successor, John Major, was by and large an unknown quantity on Northern Ireland, but in its way this was an advantage. The fact that he carried no political baggage on the Northern Ireland issue meant that he could, if he wanted, be more flexible and even imaginative in his approach. There was also an important change in the bureaucracy. At the Northern Ireland Office, John Chilcot became the new permanent secretary. A former British Home Office official, Chilcot had little difficulty buying into the theory of the peace process and quickly grasped what Adams was attempting to achieve. He constructed his own pipeline into the process via John Hume, and through that contact Chilcot would also get to know Adams’s mind. Along with his talented deputy, Quentin Thomas, and the British cabinet secretary, Sir Robin Butler, the three officials were to make up the British negotiators during the tortuous road to the 1994 cease-fire.

Despite Thatcher’s fall and Brooke’s neutrality declaration, progress was still painfully slow. In Dublin the process had stalled. Haughey and Hume distrusted Adams’s motives, made suspicious by his failure to deliver on earlier promises of an IRA cease-fire, and the talking had slowed down. Neither man was aware at that stage that Adams’s foray into the peace diplomacy had been undertaken without the Army Council’s knowledge or approval and that his every step had to be carefully measured. By 1990 the Redemptorists wanted to withdraw Father Reid from the process amid fears about his health, but Haughey had urged them to persist and they relented.

Finally, in the early part of 1991, Haughey was spurred to action. He assembled his own team of Irish civil servants to spearhead a new initiative, choosing to extend knowledge of the peace process beyond his private office. Dermot Nally, an experienced diplomat whose involvement in Northern Ireland matters went back twenty years, headed the team. He was secretary to Haughey’s cabinet, the Irish equivalent of Sir Robin Butler. Despite his distaste for the Department of Foreign Affairs [DFA], Haughey decided to include two of its senior officials, Noel Dorr, who headed the DFA, and Sean O hUiginn, a senior official in the Anglo-Irish division whose remorseless logic would irritate the British and earn him the sobriquet “The Dark Prince.” O hUiginn, from Boho in County Mayo, had been the first Irish joint secretary to the Anglo-Irish secretariat. Opposite him, representing the British, sat Mark Elliott, the Foreign Office official whose trips to Derry were so secret only manuscript records were made of them.

In the autumn of 1991 Hume and Adams produced the first draft of a model joint British-Irish government declaration incorporating, inasmuch as they affected the British, Reid’s “stepping stones.” Mansergh and O hUiginn made changes, and when John Major traveled to Dublin for his first Anglo-Irish summit in December 1991, Charles Haughey handed him a copy of this document, known as “Draft 2.”

According to its terms the British would say that they had no “selfish, strategic, political or economic interest” in staying in Northern Ireland and that they wished only to see the Irish people live together “in unity and harmony.” Dublin would accept that Irish self-determination was subject to Reid’s twofold consent principle and agreed to set up a permanent convention that would advise on ways to exercise self-determination.[29] This would convene whether or not the British agreed to put their name to the declaration. Thus began the complex and often tortuous intergovernmental negotiations that would lead in a mere two years to the Downing Street Declaration.

HAUGHEY’S OFFICIALS had prepared a lengthy list of speaking points, sixteen in all, to present to Major in an effort to win the British premier over to a fundamental reappraisal of Northern Ireland policy. These dwelt on the financial, human, political, and social costs of a conflict that seemed to be unending and that ultimately was in the interests of both governments to resolve and stabilize. Although each politician knew more about the secret peace process than he was prepared to admit, there was only one veiled reference to the clandestine diplomacy that had been going on in the background for over five years. “There are factors working for progress,” said one account of the exchange. “There is a strong tide of public opinion in favour of peace in both communities in Northern Ireland and in Ireland as a whole. The futility and horror of violence must surely become more evident even to its proponents with every passing year and we know that at least some of these elements are reflecting on this.”[30] Haughey’s arguments succeeded, and Major agreed to initiate a comprehensive internal review that the two leaders would consider in the early part of 1992.

The meeting between them never took place. An old scandal concerning the tapping of politicians’ and journalists’ telephones resurfaced, and in early February 1992 Haughey was forced to resign. Albert Reynolds was chosen by the Fianna Fail parliamentary party to succeed him, and a day later the group of British and Irish civil servants charged with considering the new initiative met at the Cabinet Office in London. Butler, Chilcot, and Thomas told their Irish counterparts there could be no further movement on the initiative until after that year’s British general election. Various problems were raised by the British, not least their unwillingness to talk directly to the Provisionals and the difficulties of selling the language contained in “Draft 2” to the unionists.

Those on the British side were less than forthright with the Irish officials about their own lengthy dialogue with Gerry Adams, and made no mention of their complete familiarity with the concepts and language that permeated “Draft 2.” The British said they wanted an assurance of an end to violence before committing any more resources to the project. But there was a way ahead, they said. It lay in a British declaration of neutrality and an assurance by the Irish concerning the principle of consent in Northern Ireland. The meeting ended with agreement on those points and on a strategy to deal with any leaks to the media of John Hume’s role as an intermediary. But it was a doubtful, uncertain meeting; no one quite knew what the political landscape would look like in a few weeks’ time.

As it turned out, Albert Reynolds embraced the peace process with enthusiasm and John Major survived the British general election. Haughey had fully briefed the new taoiseach on the state of play, and Reynolds had immediately agreed to keep Martin Mansergh on his staff. Alec Reid’s visit to Government Buildings in Dublin increased, as did the number of meetings between Hume and Adams. In December 1993, after months of exhausting, often frustrating negotiations with the Major administration, the Downing Street Declaration was finally unveiled by the two leaders to a curiously mixed reception in Northern Ireland.[31] While unionists were relaxed about its contents, rank-and-file Provisional supporters were bewildered. Some feared that their leaders had been drawn into a deadly trap from which there could be no escape; none suspected that the declaration had emerged from the secret initiative launched by their own leader.

The reason for their confusion was contained in the declaration’s paragraph 4, which incorporated Alec Reid’s consent principle, although none of the IRA’s supporters who read it could have been aware of that. “The British Government,” it said, “agree that it is for the people of the island of Ireland alone, by agreement between the two parts respectively, to exercise their right of self-determination on the basis of consent, freely and concurrently given, North and South, to bring about a united Ireland, if that is their wish.”[32]

The declaration was an astonishing document and one of the seminal pieces of Anglo-Irish diplomacy. The fact that it elevated the principle of consent and dictated that this had to be a cornerstone of any Northern Ireland settlement was secondary. The two governments had long agreed on that. What made the Downing Street document exceptional was that it was modeled on ideas and concepts evolved, initiated, and developed in a secret dialogue whose instigator was the head of the political organization pledged to overthrow the principle by gun and bomb.

In June 2006, after a lengthy illness, Haughey finally succumbed to the rigors of prostate cancer and died at his home in Kinsealy where some twenty years earlier he and Father Reid had met to begin the peace process. Haughey’s supportive role in the peace process at its most vulnerable and crucial period—at its birth—had rarely been acknowledged in Irish public life, perhaps because it conflicted so starkly with the dark image that his political enemies and many in the Irish media preferred to project. But at his graveside Haughey’s protégé and successor as Fianna Fail leader and taoiseach, Bertie Ahern—once famously described by Haughey as “the most clever, the most cunning, the most devious of them all”—paid an overdue tribute to the role Haughey had played in ending the Troubles in the North: “His courageous decision to open a secret channel of communication with the Provisional leadership paved the way to the banishing of the bomb and bullet, North and South, in our time.”[33]

10. “No Idle Boast”

Dublin’s Mansion House has witnessed many dramatic and historic moments in its nearly three hundred years of life. Built as a town house in 1710 in the middle of Georgian Dublin by a wealthy County Derry property developer, Joshua Dawson, it has been the official residence of the city’s lord mayors since 1715. Its famous Round Room, added to mark the visit of King George IV in 1821, has been the venue for some of Ireland’s most momentous political gatherings. In January 1919 the First Dail, dominated by the revolutionary Sinn Fein party, met there to ratify the 1916 proclamation and adopt Ireland’s Declaration of Independence. Only twenty-six of the sixty-nine successful Sinn Fein candidates were at liberty to attend the session, and although the British authorities permitted the rebel parliament to meet, the proceedings were closely monitored from a building across the road, by the inspector general of the Royal Irish Constabulary, Sir James Byrne, and his Dublin counterpart, Colonel Wedge-worth Johnston. The Second Dail also convened in the Mansion House more than two years later but was dissolved after only a year as civil war gathered Ireland in its terrible grip. And it was in the Mansion House that Eamon de Valera had presaged the civil war when he told delegates to the Sinn Fein Ard Fheis of October 1921 that “Ireland’s representatives will never call upon the people to swear allegiance to the English King.” When a few weeks later Collins and the Treaty delegates brought home from Downing Street an agreement requiring just that, Irish republicanism was convulsed by violent division.

More than sixty years later, on November 2, 1986, as once again the forces of law and order watched events from a safe distance, the Mansion House was witness to another great Irish political rupture when Provisional Sinn Fein delegates, after nearly five hours of tense and at times emotional debate, elected to drop the party’s long-standing opposition to taking seats in the Irish parliament. To some in the hall this formal act of recognizing the hated partitionist settlement was a case of déjà vu. The same issue had split the movement in 1969 and 1970, with the IRA and Sinn Fein taking separate but equally divisive stands on the issue. The resulting quarrel had torn the movement apart, and the consequences were violent and long-lasting, leading not just to the formation of the Provisional IRA but also to the creation of a feud culture within republicanism in which ideological differences were often settled by the shedding of blood.

There were, however, key differences this time that would make this split less divisive and provide evidence that Gerry Adams had learned important lessons from the mistaken way Cathal Goulding and his allies had handled the issue in their day. In 1970 Goulding had advocated entering all three partitionist parliaments at the same time—the Dail in Dublin, or Leinster House, as republicans preferred to call it; Stormont in Belfast; and the House of Commons at Westminster.

In 1986 Adams made a much less ambitious proposal to the republican movement. He wanted Sinn Fein to enter only the Irish part of the 1921 settlement, Dail Eireann; the British parts, Stormont and Westminster, would still be off limits. There was no ideological reason for this, since once the principle of abstention was ditched for one assembly it applied to all three. But by advocating entry first into the Dail, a far-off place about which Northern republicans knew little and cared less, Adams minimized likely opposition from Northern IRA units.

Goulding’s move also came at a time when his internal critics were complaining that he was running down the IRA’s military capacity and had failed to ensure that Catholic districts of Belfast were adequately protected against loyalist attack. No one could level that charge against the Adams-led IRA. The debate on dropping abstentionism came just as the IRA was receiving huge shipments of weaponry from Colonel Qaddafi. Far from running the IRA down, Adams and his colleagues were promising an escalation in both the quality and the quantity of IRA operations, and on such a scale that they could possibly tilt the military and political balance in the IRA’s favor. In 1969 the IRA had split after a general Army Convention had voted in favor of the policy switch, but at a similar Convention, held in County Donegal in September 1986, delegates were able to reconcile their differences without a rupture. Reassured by the leadership’s promise of an intensified war, most of the delegates who voted against the motion stayed in the IRA’s ranks. In 1970 they had walked out.

In 1986 some of those attending the Sinn Fein Ard Fheis did walk out in protest, but they never presented the threat posed by the dissidents of 1969. Led by Ruairi O Bradaigh and other veteran activists, they went on to form Republican Sinn Fein (RSF). Despite the purity of their beliefs, the O Bradaigh wing failed to persuade enough Provisionals to join them. The split was small and contained, confined largely to older republicans, many of whom were based south of the Border miles away from the war zone, where the Provisional IRA still carried the greater appeal for grassroots militants. Although RSF supporters did set up a military wing, such was the fear of reprisals from the Provisionals that it was years before the Continuity IRA revealed its existence. That in itself was a comment on the new group’s frailty.

To most delegates and nearly every media representative crammed into the Mansion House that day, there seemed little doubt that the underlying reason for that was the Adams leadership’s apparently unswerving support for armed struggle. It had been clear to all in 1969 that Goulding was intent on demilitarizing the republican movement. There was, apart from members of the O Bradaigh–O Conaill wing, no such suspicion by 1986. The debate about abstentionism seemed to be about what the leadership said it was about, an attempt to make Sinn Fein more relevant on the southern side of the Border by abandoning a policy stance that was unpopular with the bulk of voters there.

What only a few in the Mansion House could know was that the Adams leadership had already commenced a secret diplomacy with the leadership of Fianna Fail and that dropping abstentionism was a key part of the strategy. Changing Sinn Fein and IRA policy was in fact vital to the prospects of creating an alliance with Charles Haughey; after all, how could the Fianna Fail chief even contemplate a relationship with Adams as long as the Provisionals rejected the legitimacy of the state he led? The shift would undoubtedly help Sinn Fein win votes, but that would take many years to happen and require an IRA cessation. Primarily, the ending of abstentionism prepared the way for constructive dialogue with Haughey.

When Father Reid sent Haughey his lengthy message in 1987, he cited the outcome of the 1986 Ard Fheis as evidence of the strength of the Sinn Fein leadership and the freedom it had to initiate and organize its own policies. Adams, in other words, was the man with the skills and the trust of the rank and file needed to do the job. “Alec would say,” recalled one participant in the peace process, “that not even de Valera or Collins was able to accomplish such a huge shift in policy and get away with what was really an insignificant split and no upheaval worth the name.”[1] What he meant, of course, was that while de Valera and Collins had both steered the IRA of their day into constitutional politics, it had been at the cost of serious and bloody splits. Adams was doing the same but largely kept the organization intact. That meant that he might even be able to deliver a fully intact IRA into the peace process.

To the outside world, dropping abstentionism seemed to be the logical conclusion of Adams’s efforts to politicize the Provisionals. He had started to move the organization in this direction during the Cage 11 days in the mid-1970s, when he had argued in favor not only of social and economic radicalism but of republican involvement in agitational politics in the South. Throughout the development of this strategy, Adams and his allies had justified it on military grounds. The argument was disarmingly simple; the more republicans identified with the lives and needs of the poorest sections of Southern society, the more support, tacit or otherwise, the IRA would get for the war of national liberation in the North. Adams would use the same argument to urge that abstentionism should be dropped.

In practice, however, it worked the other way around. When Sinn Fein tested the electoral waters in the Republic, the party discovered that the violence of the IRA’s campaign and Sinn Fein’s associations with the Northern conflict were liabilities. The early toleration of the IRA’s campaign in Northern Ireland had virtually disappeared by the late 1970s and had been replaced by a fear that the conflict would spill over the Border, as it had in May 1974, for instance, when loyalist bombs killed thirty-three people in one day in Dublin and Monaghan. Many Southerners blamed the Northern republicans for this and strongly resisted voting for Sinn Fein. Although Sinn Fein attributed this to state censorship, particularly section 31 of the Broadcasting Act, which banned the party from radio and television, the truth was that the IRA and its violent activities were the reason for the party’s electoral failure.

Adams’s push for involvement in Southern politics began in 1977, not long after his release from Long Kesh. It was there in Jimmy Drumm’s “long war” Bodenstown address in June that year and was made explicit in January the following year, by which time Adams had become IRA chief of staff.

As in the North, it was the hunger strikes of 1981 that strengthened Adams sufficiently to enable him to accelerate Sinn Fein’s electoral activity south of the Border. H Block candidates had done well in the Republic’s general election of June that year when nine of them won nearly 43,000 votes, some 2.5 percent of the electorate, and two won seats, both taken by Long Kesh prisoners. If nonprisoner candidates, pledged to take their seats, had run instead, they would have held the balance of power in the new Dail. Flushed with these victories and the separate successes in the North, delegates to the 1981 Ard Fheis had given the party leadership a free hand in deciding whether to contest future elections, North and South. The first opportunity to do that south of the Border came in February the following year when the FitzGerald government elected in the turbulent summer of 1981 suddenly collapsed.

As it turned out, the election was a disaster for the republicans. Sinn Fein ran in seven constituencies, two fewer than in 1981, and selected a decidedly military-leaning slate. Two of the best-known candidates were IRA prisoners. Joe O’Connell, the leader of the Balcombe Street IRA unit, which had bombed and shot its way around southern England and central London in the mid-1970s, ran in Clare, while Seamus McElwaine, an IRA leader from the Monaghan-Fermanagh Border, then on remand in Crumlin Road jail, Belfast, ran in Cavan-Monaghan. His agent was Caoimhghin O Caolain, a former bank clerk from Monaghan town, who had been talent-spotted by Adams during the hunger strike and subsequently groomed for the seat. He eventually became TD for the constituency in 1997. The overall result revealed a collapse in the H Block vote, from 43,000 votes, or 2.5 percent, to just under 17,000, or 1 percent. The message was clear: Southern voters might come out at an emotional moment and support dying hunger strikers, but otherwise they would spurn the IRA and its political wing. By way of contrast, republicans who had run on a platform of taking their seats did well. Bernadette Devlin-McAliskey tripled the H Block vote in Dublin North-Central, while Sinn Fein’s old rivals in the Goulding’s Officials, now calling themselves Sinn Fein–The Workers Party, won three seats.

Sinn Fein drew in its horns in 1982 after that poor first result. That Easter the Army Council reaffirmed the traditional disdain of constitutional methods: “Only through armed struggle will we be listened to…, ” it said.[2] When another Dail election was called in November 1982, Sinn Fein decided not to contest it, conserving its resources for the election to Jim Prior’s “rolling” Assembly in Belfast, which was held the month before. Nevertheless, the move toward dropping abstentionism in the South continued apace as Adams and his allies chipped away at the opposition at one Ard Fheis after another. First Sinn Fein agreed to contest European elections—a proposal Adams had resisted when O Bradaigh had suggested it—and to take its seats in the unlikely event of winning. Then it agreed to register as a political party in Dublin, a move that de facto recognized the hated Southern state. At the next conference a motion declaring abstentionism to be a tactic, not a principle, was passed, although short of a large enough majority to change the party’s constitution. Slowly, incrementally, the party was moving to end its ban on taking seats in the Irish parliament and recognizing the state that the IRA’s predecessors had vowed to destroy.

THE IRA held one of the keys to winning the issue, or rather Colonel Qaddafi did. By October 1986, a month before the Ard Fheis and just after the IRA Convention, the IRA had smuggled some 130 tons of Libyan weaponry into Ireland in four separate shipments and had successfully hidden these in secret dumps in some of the most isolated parts of Ireland. As the Army Convention and the Sinn Fein Ard Fheis neared, the number of key IRA activists who knew that substantial amounts of arms had been successfully landed slowly grew. “People were told there was big gear come in,” remembered one IRA member, even if only a select few knew the details.[3] The origin of the arms, the fact that they were a gift from Qaddafi, was still a secret to all but the Army Council. With tons of weaponry stored away and many middle-ranking IRA activists aware that a big offensive was in the pipeline, the notion that Adams was about to sell out just seemed absurd. Colonel Qaddafi had helped make the outcome of the abstention debate a foregone conclusion.

Before going to the Sinn Fein conference with the proposal, Adams first had to win over the IRA. Once that was done, Sinn Fein would be easier to handle. In early 1986 the Army Council set up a special subcommittee to examine the possibility of holding an IRA Convention. When the decision to go ahead was made, it moved very fast. “It was done very hurriedly, in less than two weeks, when normally it would take a month to organize,” remembered one delegate.[4] Before the Convention met, the ever-reliable Brian Keenan, by then into his seventh year in various English jails, made his support for Adams known publicly. In a letter to An Phoblacht–Republican News, co-signed by three other IRA prisoners held in Britain, Keenan wrote, “It is time for a change to enable elected representatives to carry out revolutionary work in the corridors of power. We do not believe any republican principle is involved in this issue.”[5] A secret letter went to the Army Council saying the same thing.

At the Convention, the Adams camp, including Adams himself, argued that a yes vote would assist the armed struggle by broadening the base and producing more safe houses for the IRA. Martin McGuinness predicted that Sinn Fein would win up to five or six seats, and if that happened no government in Dublin would dare move against the IRA. The IRA could launch attacks in the North safe in the knowledge that its rear was secure against assault. Danny Morrison claimed that the dead hunger strikers, especially Bobby Sands, would have voted yes, and Seamus Twomey declared his support for the change. Opposition was strongest from around the Border in South Armagh, Louth, and South Down. But the message from the outgoing IRA Army Council was that the war was going on regardless of how the vote on abstention went, and this swayed the meeting. “The way Adams did it was to say to the IRA men, ‘This is politics. You get on with your thing—the war—and it won’t be affected,’” remembered another delegate.[6] The Convention was held in a room that sloped downward from the back rather like a cinema; somewhat disconcertingly for the delegates, the leadership sat in the rear seats, watching and noting the speeches and who voted which way.

The vote went three to one in favor, comfortably exceeding the required two-thirds majority. There were two bonuses. The Convention once again upheld General Army Order no. 8, which had prohibited IRA actions against the security forces in the Republic for some forty years. The effect of this would be to reinforce Father Reid’s conciliatory message to Haughey. Adams also got a restructured Executive, the first to be elected since 1970 and one much more sympathetic to his politics, at least initially. A small minority eventually left the IRA after the vote, but there was no walkout from the Convention, as there had been in 1969.

ADAMS’S PROBLEM was not with the IRA but with Sinn Fein. The IRA had been under the sway of the Belfast leadership ever since the establishment of Northern Command, and since it was larger in the North the units there would always have a greater say at a Convention. But Sinn Fein was as numerous on the ground in the South as in the North, and feelings in the South were stronger against change. Many of the Southern members of Sinn Fein came from families and backgrounds shaped by the bitter divisions of the civil war in 1922 and they literally hated the system of government in Dublin. The Southerners had stymied Adams at the 1985 Ard Fheis when a move to declare abstentionism a tactic rather than a principle failed to get a two-thirds majority. Having foiled Adams once, they could do it again. The difficulty that Adams and his allies faced was the possibility that they would end up with the same result that Goulding had got in 1969 when the IRA had voted for the change, but Sinn Fein was against. If sufficient numbers split from Sinn Fein, they could act as a magnet for military dissidents, and the division could worsen. With Father Reid’s approach to Haughey building up speed, Adams’s need to avoid a big public split in Sinn Fein assumed greater significance.

Just before the Ard Fheis, Adams moved to try and neutralize O Bradaigh, and he used the Libyan weapons to lure his opponent, hinting that a number of big arms shipments had been brought in and that the war effort would be intensified. “Ruairi was told there was good news in that regard,” said one source familiar with the meeting.[7] But Adams’s efforts to win over O Bradaigh failed. On the day of the debate he again asked to see O Bradaigh and his supporters. They met backstage in the Mansion House at lunchtime, both accompanied by supporters, for what turned out to be a bad-tempered and fruitless encounter. Adams had brought Micky McKevitt along as evidence that even though the powerful quartermaster’s department had voted against the change at the Convention, it was nevertheless prepared to stick by the Adams leadership afterward, once more hinting at the arms shipments and the coming IRA offensive. Daithi O Conaill had been barred by the Adams camp from the meeting, and it ended with a threat to the dissidents: if they set up a rival Army, the Provisionals would take O Conaill out.[8] One of O Bradaigh’s supporters, Des Long from Limerick, slammed a table with his walking cane in anger, and they stormed out.

THE ARD FHEIS DEBATE started just before 11:00 A.M. on the Sunday morning, but many believed the key contribution had been made the evening before, when Gerry Adams gave his presidential address. His speech set out the classic arguments for change. The IRA had met in Convention and approved the move without staging a walkout, he said. It followed that anyone who opposed a yes vote was actually opposing the IRA. Critics had gone to the establishment media with accusations that the leadership was going down the same road as Goulding’s “Stickies.” “To compare us with the ‘Stickies’ is an obscenity,” he protested. “For anyone who has eyes to see, it is clear that the Sticky leadership had abandoned armed struggle as a form of resistance to British rule as part of their historic departure into British and Free State constitutionality. For our part, this leadership has been actively involved in the longest phase ever of resistance to the British presence. Our record speaks for itself. We have led from the front and from within the occupied area.”[9] And, he added, the armed struggle would continue until victory. “We all have a part to play in it and those of us who remain committed to it will ensure, regardless of the dangers it holds for us, that this struggle is going to continue until Irish independence is won. That is no idle boast.”[10]

Others echoed the military arguments. Another Adams ally, County Donegal man Pat Doherty, opened the Sunday debate on the abstention motion and argued that the major difference between Sinn Fein and other republicans who had entered parliament in the past was this leadership’s commitment to the IRA’s armed struggle. John Joe McGirl, a former chief of staff, repeated the message. He had gone to other veterans, J. B. O’Hagan, Joe Cahill, and Seamus Twomey among them, he told the delegates, and they all supported the change. “We have an army fighting 16 years which will continue to fight until British rule is defeated,” he declared.[11] Joe Cahill said that the Goulding leadership “had sold out the military spirit,” but he was confident that by the time the election after next came, the deadline for progress set by Adams, “the freedom fighters of the IRA will have forced the Brits to the conference table.”[12] Speaker after speaker who backed the move invoked the IRA and the leadership’s commitment to the armed struggle.

It was left to Martin McGuinness to deliver the hardest and most uncompromising military message of the debate, establishing a precedent for the peace process in which McGuinness’s militancy would be regularly flourished to reassure the rank and file that there would be no sellout. A large enough number of the delegates would have known exactly who Martin McGuinness was—that he was the current Northern commander in charge of the day-to-day war against the British. That day the Derry man was the voice of the IRA.

He began his speech with a commitment on behalf of the leadership never to enter Stormont or Westminster and then turned to allegations that he and other Republican leaders had plans to abandon the armed struggle. “I reject any such suggestion and I reject the notion that entering Leinster House would mean an end to Sinn Fein’s unapologetic support for the right of Irish people to oppose in arms the British forces of occupation,” he said. “That, my friends, is a principle which a minority in this hall might doubt but which I believe all our opponents clearly understand. Our position is clear and will never change. The war against British rule must continue until freedom is achieved.”[13] The opposition to the change, he continued in a less than subtle reference to O Bradaigh and O Conaill, was not about abstentionism but about the 1975 cease-fire. “The reality is that the former leadership of this Movement has never been able to come to terms with this leadership’s criticism of the disgraceful attitude adopted by them during the disastrous 18 month ceasefire in the mid-1970’s.”[14] In other words, the only people who were talking about the war’s being ended had themselves nearly brought the IRA to defeat through a foolish cease-fire. How could anyone think that this leadership, which had rescued the IRA then, could or would make the same mistake?

At around four-thirty that afternoon the debate ended, and the Mansion House, which had been packed with media and spectators, waited expectantly for the result of the vote. When it came, the sigh of relief from the Adams camp was almost audible. Fully 429 delegates had voted for the change, 161 against, and 38 abstained. A two-thirds majority was 418, and the Sinn Fein leadership had made it by the votes of just 11 delegates.

In the excitement of the moment no one noticed that the number of delegates attending the Ard Fheis had inexplicably doubled from its usual figure. At the 1985 conference, the year before, the motion seeking to define abstentionism as a tactic and not a principle had been lost by 181 votes to 161; a total of 341 delegates had cast their votes. Yet just a year later the number of delegates at the Ard Fheis soared to 628, almost double; the following year, in 1987, it reverted to its normal 350 or so delegates. That was also the number of SF delegates, more or less, who voted overwhelmingly to back the Good Friday Agreement when Sinn Fein held a special Ard Fheis in May 1998, twelve years later, to discuss the political deal. Each year after 1986 and before 1998 had seen more or less the same number of delegates at each Ard Fheis. The puzzle is why the number of delegates suddenly jumped to over 600 in that one crucial year.

The explanation, according to a number of republican sources, is that the 1986 Ard Fheis vote was really organized and manipulated by the IRA, with all the care and preparation normally reserved for a military operation. This exercise, which began as early as 1984, had been twofold in character. One well-placed Belfast Sinn Fein source active at the time described what happened:

They went about it in two ways. Over a two-year period beforehand released IRA prisoners loyal to Adams were ordered to join Sinn Fein cumainn [branches] and take them over by replacing hostile or unsympathetic officers. In one instance in Andersonstown that I remember, two of these people got themselves chosen as delegates to the 1985 Ard Fheis and just point-blank refused to propose a motion that was seen as critical of Adams.

The other way was that they just invented Sinn Fein cumainn. All you needed was five names, and you got two delegates to the Ard Fheis. They were set up all over the country and in Belfast. I personally saw faces at the ’86 Ard Fheis I had never seen before or since. There must have been a hundred or more of these cumainn but after ’86 they just petered out. It was done over a two-year period with a big push in 1986, slow at first, but then it became obvious.[15]

No one noticed the disparity in numbers, or if anyone did, they chose to remain silent. The reality was that afterward the reason for the sudden surge in membership hardly mattered. The Ard Fheis result had objectively changed republican politics in two ways: Sinn Fein, and by extension the IRA, had been edged significantly closer to constitutionalism, while the Redemptorists’ mission to Charles Haughey had been armed with a crucial piece of evidence about the bona fides and skills of the Adams leadership.

Within six months of the Ard Fheis and in great secrecy, Reid delivered the cease-fire offer to Haughey, and his trips to Kinsealy on behalf of the Sinn Fein leader increased. The peace process was accelerating, and the 1986 Ard Fheis and its historic political turnaround had played a crucial part in achieving that.

ALTHOUGH SINN FEIN was by now well on the way to an even more comprehensive political transformation, its leaders continued to assure the rank and file that the IRA’s armed struggle would remain sacrosant, no matter how unsettling the changes. Between 1987 and 1989 Adams, McGuinness, the IRA GHQ, the Army Council, An Phoblacht–Republican News, and other influential figures and bodies gave one promise after another that the IRA would carry the war to the British until they agreed to leave Ireland and that whatever the talks with the SDLP were about, they did not encompass an IRA cease-fire.

At the same time this was happening, the Adams leadership gradually introduced some of the key elements of the still-secret “stepping stones” agenda into the republican vocabulary. But not all of them, by any means. Some concepts, like the redefinition of British withdrawal, were just too heretical to be allowed into the public domain. Nor were the rank and file, or for that matter the Army Council, told about the secret strategy that underlay the new concepts. To call what happened between this point and the declaration of the first IRA cease-fire in 1994 a “debate,” as more than one observer has, would be something of an exaggeration. Ideas were certainly put into circulation, but the ideas belonged mostly to a Sinn Fein leadership that preferred to talk at rather than to or with their base. Dissension was frowned upon. One internal theoretical magazine, Iris Bheag (Little Magazine), which had been thrown open to the rank and file, was quietly suppressed around 1990, after IRA prisoners had used it to criticize aspects of the Adams strategy. Iris Bheag was replaced by two glossy productions, The Captive Voice, for IRA prisoners, and The Starry Plough, which featured mostly articles written or approved by SF headquarters. If there was a debate, it was mainly on the leadership’s terms.

In April 1987, just before Tim Pat Coogan carried Alec Reid’s letter to Haughey, Sinn Fein issued an eight-page discussion paper entitled “A Scenario for Peace.” Couched in the traditional idiom of republicanism, the document repeated the conventional demand for British withdrawal, called for the disbanding of the RUC and UDR and the release of IRA prisoners, and demanded reparations for British misrule. It seemed to be a routine repetition of republican dogma and was ignored by the media and most political rivals. But “A Scenario for Peace” did two things. It scrapped the simplistic and rigid “Brits Out” slogan of the 1970s and replaced it with the much more subtle and flexible phrase “national self-determination” (NSD), a concept that was one of the cornerstones of the Reid-Adams strategy. NSD was a formulation within whose generous frame two utterly conflicting definitions could coexist quite happily, one the traditional republican one, which envisaged physical British expulsion, and the other Reid’s version, which said that any move by the British to leave Northern Ireland had to be with the consent of unionists, not just nationalists. One other “stepping stone” idea was contained in the paper, the proposal for a constitutional conference at which elected representatives from all the Irish traditions, nationalist and unionist, would hammer out a settlement. Again worded in republican language, this idea nevertheless could happily sit alongside the Reid-Adams formulation. Adams’s efforts to move Sinn Fein away from its ban on taking seats in the Dail had been a slow, cautious, and incremental journey; it would be the same with the bid to make the Reid-Adams diplomacy republican policy. Slowly, the building blocks for a settlement based on the constitutional status quo, by and large, were being laid—and Sinn Fein would be part of that settlement.

Within two years another plank in the Reid-Adams strategy, pannationalism, fell into place when the notion of an alliance with constitutional nationalism became de facto Sinn Fein policy. The groundwork had been done in the months before. Against a background of regular assurances throughout the spring of 1988 from Martin McGuinness that talks with the SDLP had nothing to do with an IRA cease-fire, a troop of Adams’s supporters, Mitchel McLaughlin, Tom Hartley, and Danny Morrison among them, made public calls urging the adoption of a pan-nationalist strategy. It was all in preparation for a special internal conference of Northern Sinn Feiners at the end of June 1988 that would consider a special report prepared by the SF general secretary, Tom Hartley, arguing for an alliance with the SDLP.

This was unfamiliar and unfriendly territory as far as many Provisional activists were concerned. Most still shared the scathing view of their constitutional rivals that had been expressed by IRA prisoners at the end of the 1981 hunger strike, and six years of electoral rivalry between the parties had, if anything, deepened the hostility at grassroots level. The June 1988 conference also heard a critique of the IRA’s armed struggle from a Belfast councillor, Martin O Muillear, a post–hunger strike, non-IRA recruit to republicanism who told the delegates that there were contradictions between “the armed struggle and our political work,” not least in the area of job creation. As a councillor he wanted jobs for his constituents, but the IRA was pledged to deter inward investment with its bombing campaign. “[L]et’s have enough savvy to tell the difference between what is a position of political suicide,” he told the 150 delegates, “and what is an intelligent and pragmatic political position.”[16] O Muillear’s message was clear: the IRA was becoming an electoral liability, while Hartley’s message, the need for an alliance with the SDLP, offered an alternative.

Leadership figures were pushing the new ideas, but at the wider grassroots level reservations about dealing with the SDLP at such close quarters were deeply held and seemingly staunch. In the mostly pseudonymous columns of Iris Bheag, Tom Hartley’s discussion paper “Towards a Broader Base,” was given a dusty response by republican activists and IRA prisoners alike. Hartley’s principal contention that “Britain can only be moved on the National Question if the British state is confronted by the combined forces of Irish Nationalism”[17] implied a compromise that went too far for many. In August 1988 one group of IRA prisoners identified the central danger of the approach, as they saw it. “We don’t believe that Britain can be moved ‘only by the combined forces of Irish Nationalism,’ mainly because we feel that no constitutional party will agree to ‘formal’ alliances with us unless we reject the use of armed struggle (and we don’t believe such is possible)…”[18] The same prisoners also complained about the inconsistency in Hartley’s approach, recalling the long-held Provisional view that the SDLP’s role in Northern Ireland was to support British strategy. “We cannot simply tell our membership today ‘SDLP bad—Sinn Fein good’ and then tomorrow tell them ‘maybe SDLP not so bad, Sinn Fein good.’”[19]

IRA prisoners in H5 of the Maze prison were equally dismissive: “The central position of an alliance with the SDLP is impracticable,” they complained, “given the class position and interests of their leadership, and their stance on Hillsborough was a conscious reinforcement of imperialist rule. Neither the SDLP (nor the Dublin government for that matter) were pushed into this treaty, they jumped! Do we seriously believe we can force Hume or Haughey to adopt a more ‘Republican’ position?”[20] Tony Doherty, SF education officer in Derry, put it more bluntly: “The Republican Movement is clearly identified as the arch-enemy of the SDLP. We [must] show them that we make a very formidable enemy.”[21] Iris Bheag was closed down not long afterward.

Although Adams and his closest allies continued throughout 1988 to urge pan-nationalism on the republican base, the level of internal hostility toward it and the fact that in the autumn of that year Gerry Adams and John Hume took the Sinn Fein-SDLP talks into private and secret mode forced a more cautious and circuitous approach on the leadership. At the Ard Fheis of January 1989 talk of pan-nationalism had disappeared, replaced by a slogan calling for the setting up of a left-wing “anti-imperialist” mass movement that would bring together “nationalists, patriots, socialists, republicans and democrats” to press for national and economic self-determination.[22] The Adams leadership—Tom Hartley, Martin McGuinness, Jim Gibney, Mitchel McLaughlin, and Adams himself—lined up to recommend the new slogan to the delegates, and it was easily approved. Although one agitational group, the Irish National Caucus, did emerge, little resulted in terms of longterm practical initiatives. As a strategy it disappeared from the leadership’s lexicon not long thereafter, replaced once again by pan-nationalism, albeit rechristened “the Nationalist consensus.” Not for the first or last time, Adams used the appeal of radical politics to bring his supporters to the desired goal.

The 1989 “anti-imperialist front” proposal got the rank and file accustomed to the notion of Sinn Fein’s working with other parties in broad fronts, an essential prerequisite for an alliance with the SDLP. The motion, Adams told the Ard Fheis, was a recognition of the republican movement’s weakness: “[W]e can’t win this struggle alone…,” he said.[23] This would reemerge years later as the central argument for calling a cease-fire. Crucially the strategy also envisaged Sinn Fein’s making alliances with parties that were opposed to the IRA’s violence. “Participation in this movement,” a leadership document stated, “should not be dependent on one’s attitude to armed struggle. Those involved in the movement would have the right to their own individual positions on armed struggle.”[24] In practice, as a moment’s reflection would have revealed, prospective constitutional allies would refuse to consider a relationship with Sinn Fein unless ending the armed struggle was on the agenda. The H Block prisoners were, in their terms, right. In an important sense this was history repeating itself. It was precisely this fear that constructing alliances with non-Republicans would dilute the armed struggle that had persuaded the founders of the Provisionals to reject Cathal Goulding’s version of broad-front politics, the “National Liberation Front,” exactly twenty years before.

The move to drop abstentionism contained a hugely significant message for Haughey and the British, but it was by no means the only signal sent in these crucial years. In March 1987, a month before the publication of “A Scenario for Peace,” the IRA’s Northern commander, Martin McGuinness, bowed to Catholic Church pressure in Derry and announced that the IRA would review the practice of firing volleys over the coffins of IRA Volunteers in church grounds. McGuinness’s retreat came in the wake of a public protest from the bishop of Derry, Edward Daly, after two masked men had fired handguns over the coffin of Gerard Logue, a twenty-six-year-old from the Gobnascale area of the city who had been killed accidentally by his own weapon while on active service. Logue’s funeral cortege had been surrounded by police and soldiers in a bid to prevent a paramilitary display, but at the door to St. Columba’s Long Tower church the IRA had seized the chance to give their fallen comrade the traditional farewell salute. Daly accused the IRA of reneging on an understanding not to violate church grounds in this way, and he said that in the future the bodies of paramilitary members would not be allowed on church premises in the city.[25]

Bishop Daly, unlike his namesake in Belfast, had supported the secret diplomacy of Father Alec Reid and was an enormously popular figure in Derry. Television coverage of him waving a bloodstained handkerchief as an improvised flag of truce at heavily armed British troops as he tried to secure safe passage for one of the fatally wounded civilians shot on Bloody Sunday became one of the visual icons of the Troubles. McGuinness’s decision not to seek a fight with Daly over the funeral issue was understandable in the light of all that, but it also served to highlight a largely unnoticed IRA policy change that had significantly demilitarized its funeral rites. The practice of firing volleys over IRA coffins, whether in church grounds or not, had at one time been part of the ritual of defiance, but by this stage had ended in Belfast. The change coincided with the expulsion of Ivor Bell and his hard-line, militarist allies and with the beginnings of Father Reid’s secret peace enterprise. Instead the IRA now paid “tribute” to martyred colleagues by firing volleys over their wreath-bedecked photograph, usually after the funeral had taken place. The effect of McGuinness’s concession was to extend the change to Derry, the second city of Provisional republicanism. The change had been partly forced upon the IRA by a tougher British security force presence at IRA funerals, but the fact that the leadership chose not to make an issue out of this, when it could easily have sought violent, destabilizing street conflict by staging firing parties, was a subtle signal to both governments and a concession to the Catholic Church allies whose friendship Sinn Fein sought.

Republicans still tried to bury their fallen colleagues with other military trappings. Tricolors and IRA berets and gloves were draped over caskets, and corteges would be accompanied by marching color parties in displays that often led to ugly confrontations with RUC riot squads. These caused violence and tension, but in circumstances less calculated to cause death and more likely to generate sympathy from a wider nationalist audience.

Similar signals were sent by an Army Council decision to demilitarize the color party that led the march into Bodenstown graveyard each June for the annual commemoration of Wolfe Tone. The Goulding leadership had used this public ceremony to advertise its own ideological shifts, and ever since the split in 1969 the Provisionals had insisted that their color party at the annual ceremony be an IRA one, complete with all the customary paramilitary trappings, masks, and uniforms. It symbolized their republican purity and the adherence to armed struggle. In June 1987, however, the color party was suddenly civilianized, much to the disappointment and puzzlement of the rank and file. In the Iris Bheag of September 1987, a correspondent called “The Weasel” complained bitterly. “Of all the people I spoke to about the change in policy in relation to colour parties at Bodenstown,” he wrote, “not one single person agreed that it was a good policy and nearly all agreed that it was sapping the morale which went with previous parades before the policy change.”[26] In the same issue, “Sea Hag” commented, “I wouldn’t have seen [the colour party] go by only that someone next to me said it was passing.”[27] “The Weasel” unwittingly detected the subliminal message contained in the move: “Defiance was always one of our strong points, without it we will be left to the history books.”[28]

There were other adjustments in Provisional theology every bit as significant. In February 1988, a year later, a Catholic civilian, twenty-four-year-old Aidan McAnespie, was shot dead as he walked through a British army checkpoint at Aughnacloy, on the Tyrone-Monaghan Border, en route to a local Gaelic football game. Despite British protestations to the contrary, McAnespie’s family alleged the killing had been deliberate and cited a history of harassment and verbal abuse directed at the dead man by soldiers manning the checkpoint. The Irish prime minister, Charles Haughey, chose to cause a diplomatic row with the British over the incident. He ordered a senior Garda officer to investigate the death, and quietly the IRA and Sinn Fein instructed its councillors and supporters to cooperate with him. The importance of the move was enormous; traditional republican dogma forbade cooperation with the Irish police force in such matters on the grounds that the police enforced the laws of what was seen as an illegal state. Again a writer in Iris Bheag hinted at the internal controversy this had caused: “Some thought it was wrong because it legitimised what we in Sinn Fein do not see as legitimate.”[29] The underlying message to Haughey and to the Southern security authorities was unmistakable. The Army Council, under Adams’s direction, was in a very real way conceding to Haughey its traditional claim to be the only legitimate government in Ireland.

A similarly significant signal was sent to the British the following month when, for the first time since the split with Goulding, the organizers of the annual Easter Provisional republican parade to the IRA plot at Milltown cemetery in West Belfast filed for permission from the RUC to hold their march. The traditional refusal to ask official leave carried legal penalties, but rather than extend recognition to the Northern Ireland state that filing for an RUC permit implied, the organizers were normally quite happy to bear the consequences. Later in the year Sinn Fein began to use the courts in Northern Ireland to challenge unionists who were refusing to give the party’s councillors seats on committees. It was the first time ever that republicans had resorted to British justice in this way to seek redress.

The effect of all this was to flag an important message that only a few could detect. The bulk of the republican grassroots were distracted by the noise of the IRA’s ongoing war, which was by 1988 in the midst of the postponed “Tet offensive,” and those who did notice were satisfied with the explanation from Adams’s colleagues in the Sinn Fein leadership that it was all a ploy to win sympathy in the South and among the greener fringes of constitutional nationalism in the North. None of them could have guessed the real significance and meaning of the changes.

11. Death in Tyrone

Loughgall, County Armagh, Friday, May 8, 1987. It was just after 7:15 on a bright spring evening when the shooting began. Witnesses later said that the gunfire seemed to go on forever, but in such situations the mind often plays tricks and events can seem to happen in slow motion. Nonetheless the shooting was relentless. When finally it ceased, an eerie silence descended over the shattered remnants of the village’s police station. The scene that greeted the Special Air Services (SAS) soldiers and policemen as they carefully emerged from their hiding places was shocking even by the standards of a Northern Ireland whose seventeen years of bloody conflict had been regularly punctuated by multiple loss of life. Scattered around the bullet-riddled shell of a Toyota Hiace van lay the bodies of the cream of the Provisional IRA in County Tyrone. The eight-man active service unit (ASU), the cutting edge of the East Tyrone Brigade, had been wiped out in the carefully planned ambush, cut to ribbons in a withering fusillade of automatic fire. Estimates of the number of rounds fired by the British that evening reach as high as twelve hundred.

The East Tyrone ASU’s plan had been to destroy Loughgall police station with a huge bomb. The device had been placed in the bucket of a mechanical digger hijacked from a farm near Dungannon and then driven the twenty-five minutes or so to the picturesque village situated deep in Armagh’s rolling apple orchard country. Loughgall police station was a part-time post, manned by just three or four RUC officers, which opened only in the mornings and afternoons. The IRA men expected to find it unoccupied when they arrived, and anticipated no resistance. Three of the ASU traveled with the digger, while the remaining five drove ahead in the Toyota van, scouting for patrols and checkpoints. The digger easily broke through the wire mesh rocket-proof fencing that surrounded the police station, and it was rammed into the building so that when the explosion came it would cause maximum damage. The fuse was lit, but just as the bomb detonated the SAS opened fire.

The East Tyrone IRA had walked into a carefully laid trap. More than three dozen British soldiers drawn from the elite SAS regiment were lying in wait for them. Armed with heavy machine guns and automatic rifles, they poured bullets into the startled IRA men from at least four points. Backing them up were scores of officers from the RUC’s elite paramilitary wing, the Headquarters Mobile Support Unit, while shadowing the IRA operation as it unfolded that day were many more police and military surveillance personnel. The scale of the British operation spoke eloquently to the quality of intelligence that had come their way. The conclusion was difficult to avoid: someone, somewhere in the IRA had betrayed the East Tyrone Brigade.

The Loughgall ambush resulted in the heaviest single death toll suffered by the Provisional IRA in all the years of the Troubles. The only comparable loss in IRA history had happened way back in the middle of the Tan War in February 1921 when, in an operation characterized by the same military ruthlessness, a fifteen-strong IRA company was pinned down in a cottage in Clonmult, County Cork, by a mixed party of British troops and auxiliary police. After a fierce firefight the IRA men surrendered, but when they did they were set upon and twelve were killed. Although the scale of the slaughter at Loughgall was not as great, it probably dealt a more devastating blow. In fact the ambush had two damaging consequences for the Provisionals. Not only was the morale of IRA activists and Provisional supporters throughout Ireland rocked by the killings, but the organization lost a number of irreplaceable members, skilled and determined operators who had been slated by GHQ to play a key role in the planned post Eksund “Tet offensive.”

Even more damaging were the subsequent suspicions that have surrounded the identity and motives of those who apparently betrayed the East Tyrone Brigade. The questions have lurked in the minds of the IRA in the county ever since, hanging over Tyrone republicans like a huge black cloud that threatens to explode into a storm yet never quite does. The years since 1987 may have passed, but the speculation has never ceased about whether the traitor came from Tyrone or from elsewhere in the IRA and, if so, how high up, or, alternatively, whether the British stumbled upon the operation by expert surveillance or whether flaws in the restructured IRA sealed the unit’s fate. Even darker and more sinister are the questions that link the ambush to the then growing peace process and the impact on its development.

Whatever doubt exists about the identity or role of the Loughgall informer, there is no doubt that the ambush robbed the IRA of some of its best fighting men. Perhaps the most feared and certainly most wanted by the British was Jim Lynagh from Tully in County Monaghan. One of fourteen children, thirty-one-year-old Lynagh had joined the IRA as a teenager and was soon making his name as a fearless activist. In 1973 he had a narrow escape when a bomb he was carrying exploded prematurely. He was badly injured but survived and then was sentenced to ten years’ imprisonment, becoming one of the few Monaghan men ever to serve time in Long Kesh. In 1978 he was released, immediately rejoined the IRA, and a year later was elected a Sinn Fein councillor on Monaghan Urban District Council. Three years later he was caught carrying bullets by the Irish police and was sent to Portlaoise prison. Freed in April 1986, he once again resumed active service with the East Tyrone Brigade.

Lynagh was a ruthless IRA gunman, whose politics were simple and straightforward, his belief in military methods as unshakable as his readiness to conform to the party line was undependable, as a former colleague remembered:

He was outside the charmed circle. He was regarded [by Sinn Fein leaders] as not quite respectable enough, a bit too wild for them. He was no saint. He had been involved in the odd punch up, and his brother had been charged with shooting a bouncer. He was also a hard-line republican. I can remember when the word came [from Sinn Fein headquarters] not to mention the North or the war on the doorsteps while canvassing [for election] and to concentrate instead on social and economic issues, Lynagh objected and then just ignored it.[1]

Despite the leadership’s doubts about Lynagh, GHQ had chosen him to play a key part in the “Tet offensive”; he was one of a small group of trusted operators taken to Libya in 1986 for training in the weaponry being smuggled from Qaddafi’s arsenals.

Lynagh’s ideological soulmate was thirty-two-year-old Padraig McKearney, a member of a renowned republican family from The Moy, on the border between Tyrone and Armagh, whom Lynagh had met in jail and befriended. McKearney’s maternal grandfather, Tom Murray, had fought with the Roscommon IRA in 1920, and the family tradition lived on with the Provisionals. The oldest McKearney brothers all joined the IRA as teenagers in the early 1970s, and a sister, Margaret, has lived in Dublin, out of the reach of the British authorities, since the mid-1970s, not long after Scotland Yard branded her as “the most dangerous and active woman terrorist” operating in Britain.[2] Their youngest brother, Sean, was killed along with a lifelong friend, Eugene Martin, in May 1974 when a bomb intended for a local garage exploded prematurely as they were transporting it. It was probably his first IRA operation, and he was just eighteen at the time. The oldest brother, Tommy McKearney, rose through the ranks to become a member of the IRA’s GHQ until he was arrested in 1976 and processed through the Castlereagh conveyor belt to a cell in the H Blocks of Long Kesh. When the IRA prisoners decided to fast for the return of political status in the autumn of 1980, Tommy McKearney volunteered and was one of the seven inmates chosen to go on hunger strike. When the protest ended in confused circumstances on December 18, he had gone without food for fifty-three days. The Troubles were to touch the McKearney family again more than a decade later when the youngest son, Kevin, was shot dead by Ulster Volunteer Force (UVF) gunmen in the family butcher’s business in The Moy and an uncle, Jack, fatally wounded. Later that year another uncle, Charlie Fox, and his wife, Tess, were shot dead in their home by the same UVF. Few republican families had been as devastated by the Troubles as the McKearneys.

Padraig McKearney joined the IRA in 1971, when he was seventeen, and served two spells in jail, the first in 1974, when he was convicted of blowing up a factory, and again in 1980, when he was caught with a loaded Sten gun. He was sentenced to fourteen years’ imprisonment for that offense but never served the full term. In September 1983 he and thirty-seven other IRA prisoners broke out of Long Kesh prison in a spectacular escape. Like his close friend and fellow escaper Seamus McElwaine, another republican nonconformist, Padraig McKearney reported back for duty with the IRA as soon as he was safely across the Border.

Along with twenty-nine-year-old Gerard O’Callaghan, Lynagh and McKearney were the senior members of the East Tyrone ASU that day. Their commander was Patrick Kelly, the thirty-two-year-old OC of Tyrone IRA, who drove the Toyota van into Loughgall. Regarded by some as the architect of the IRA’s strategy in Tyrone, Kelly was seen differently by the activists. “He was there for diplomatic reasons, to represent the interests of Northern Command and McKenna [the chief of staff],” commented one. But two generations of IRA men were shot dead in Loughgall. Declan Arthurs (twenty-one), Seamus Donnelly (nineteen), Tony Gormley (twenty-five), and Eugene Kelly (twenty-five) were all chums who lived in and around the staunch republican village of Galbally, home of the dead IRA hunger striker Martin Hurson, whose death in 1981 persuaded them to join the IRA. According to those who knew them, their role model and the IRA man they most looked up to was Jim Lynagh.

THE LOSS of the East Tyrone Brigade devastated the Provisional IRA. None could have guessed it back in 1987, but the killings at Loughgall marked the start of a concerted undercover British and loyalist offensive against organized republicanism in the county, its supporters, and uninvolved Catholics which would leave over twenty IRA members and twice that number of nationalists dead within five years. As a contribution to the internal and external pressures on the IRA to go down the peace process road, this slaughter in Tyrone should never be understated.

Tyrone had always played a crucial role in the annals of Irish republicanism, and a setback to the IRA in Tyrone would be bound to damage the cause throughout Ireland. It had been that way for over three hundred years. Ever since the great Tudor campaigns against the remnants of Ireland’s Gaelic chieftains, Tyrone had stood out for the strength of its resistance to English rule. The defeat and death in 1648 of Owen Roe O’Neill, great chieftain of the Tyrone clans who had returned from European exile to lead his people into battle, marked the end of opposition throughout Ireland to the English occupation. It was not until a hundred and fifty years later that the flame of Irish rebellion was lit again when a Southern Protestant, Theobald Wolfe Tone, made common cause with the Northern Catholic Defenders to mold the United Irishmen. The defeat of O’Neill also consolidated the plantation of Ulster by lowland Scots and Northern English Protestants whose seizure of Catholic land gave England a buffer against French or Spanish invasion through its Irish back door but left a deep and enduring scar on the Tyrone psyche. “Rural memories are long,” commented one Tyrone republican. “People know which land they lost in the plantation and who took it. They can identify their fields, and in all probability the same Protestant families who took it from their ancestors are still living there.”[3]

Military defeat and repression created two tendencies in Northern nationalism, based partly on differences in social class, partly on conflicting political interest. One, from which the Provisionals sprang, preached physical resistance to British and unionist rule and the attainment of full independence, while the other, a Hibernian, strongly Catholic tendency, argued for slow constitutional advancement and the placing of sectional before national interest. The Hibernian tendency thrived in most of pre-partition Ulster and was dominant in most counties. Tyrone was the exception. There the two forces were historically more evenly balanced and are even to this day, the advantage tilting in one direction or the other, depending on which is thought better placed to give unionism the hardest time.

Tyrone republicans were to play a significant part in the 1916 Rising and the subsequent Anglo-Irish war, not least because of ties with Irish-American sympathizers on the east coast of the United States. Following the collapse of an alliance between Parnell’s parliamentarians and the revolutionary Fenian Brotherhood in the late nineteenth century, Irish republicanism found a new home in New York, where the Fenians regrouped to form the secret, elitist Irish Republican Brotherhood (IRB), whose Military Council in Ireland and the United States plotted and planned the Easter rebellion.

The IRB set up an American wing called Clann na Gael, and Tyrone men would always figure prominently in its affairs. The best known of these was the “father” of the Easter Rising, Tom Clarke, one of the 1916 Proclamation signatories, whose mother was from Tyrone and who had been reared in Dungannon, where he joined the IRB. Clarke emigrated to New York, but the Clann sent him back to Ireland, where he was arrested and then imprisoned in England for fifteen years. After his release he quietly recruited for the IRB in Dublin in preparation for the Easter Rising. Using a tactic that was to be imitated by future republican groups, the IRB infiltrated and ultimately controlled the much larger and open Irish Volunteers, who provided the manpower for the Rising. Clarke paid for his militancy with his life. Along with Patrick Pearse, who was regarded by the British as the leader of the failed insurrection, Clarke was in one of the first groups of rebel leaders executed by a British firing squad in the days after the Rising.

Clarke’s activities on behalf of the Clann began a long tradition of Tyrone involvement in the group’s clandestine support for militant republicanism in Ireland. In the buildup to the Rising, the Clann was headed by J. J. McGarritty, from Carrickmore in Mid-Tyrone, who sent Roger Casement to Germany in a fruitless search for weapons to arm the 1916 rebels. He later helped raise thousands of dollars for the IRA’s 1919–21 campaign and hosted de Valera during his lengthy stay in America. The Tyrone influence in Clann na Gael lived on, and during the Provisionals’ campaign most of its leading New York members were from the county, many of them prosperous figures in the city’s construction industry. While the media focused on Noraid as the largest U.S. fund-raiser for the Provisionals, the fact was that the secretive and close-knit Clann was a much more significant source of support.

In 1916 Tyrone was chosen by the IRB as the rendezvous point from where the Northern IRA was to mobilize and play its part in the Rising. Fearing that sectarian violence would engulf Belfast and Derry, the IRB had ordered that there be no military action in the North and instead told Northern units to gather in Coalisland, from where they were to link up with Liam Mellowes in Galway. The idea was that they would help hold the west, in a line from Cavan to Limerick. When the orders were countermanded by the Volunteers’ commander in chief, Eoin MacNeill, most of the Tyrone men returned to their homes but some went on to fight.

Tyrone’s tradition of militant republicanism went hand in hand with an often stubborn streak of dissent against distant leaderships, whether in Dublin or in Belfast. Liam Kelly was perhaps the best example of this breed. Born and reared in Pomeroy, Kelly almost single-handedly forced the IRA leadership into launching its 1956 Border Campaign. Interned by the Stormont government in the 1940s, Kelly, who had joined the IRA at sixteen, was OC of the IRA in Tyrone, and he was impatient for battle. When he was expelled from the IRA in 1951 because of unauthorized military action, he took most of the Tyrone membership with him and formed Saor Uladh (Free Ulster), which scandalized unionists when it occupied and sealed Kelly’s home village of Pomeroy. In 1953 he was elected to the Stormont parliament as an abstentionist MP on a platform of utter simplicity: “I do not believe in constitutional methods, I believe in the use of force; the more the better, the sooner the better—that may be treason or sedition, call it whatever the hell you like.”[4] Nearly a quarter of a century later his nephew Patrick would lead the Loughgall unit to their deaths.

Curiously, Kelly’s political line mirrored that of the post-1986 Sinn Fein, although his methods were much more violent than those with whom figures like Gerry Adams would ultimately be comfortable. Kelly happily accepted de Valera’s 1937 constitution, linked up with Sean MacBride’s strongly republican Clann na Poblachta, which was a partner in the 1948 coalition government and was happy to take a seat in the Irish Senate, courtesy of MacBride in 1954. Kelly’s quarrel was not with the Irish state but with the Northern state, and his pursuit of that made him an irritating thorn in the flesh of the IRA leadership. In November 1955 Saor Uladh bombed and shot up Roslea RUC station in County Fermanagh. Although Kelly was forced to flee and eventually sought refuge in New York, the unionist government was startled by the attack and the IRA leadership was forced to contemplate advancing its own military plans or risk defections to extremism. A year later Operation Harvest was launched.

The same streak of independence was evident in the years following the 1969 IRA split. Most of Tyrone stayed loyal to the Goulding leadership, if only because the Army Convention of that year had backed it, but the line between the Officials and the Provisionals was always blurred. The ideological differences that had split the movement in the South mattered little in Tyrone. What did count was which IRA was able to hit the British hardest. The killing of the Official IRA OC, John Paddy Mullen, and another member of the Official IRA, Hugh Heron, in October 1972 illustrated the point. British soldiers shot them dead at a roadblock, and there were angry allegations that they had been killed in cold blood. The Officials’ failure to strike back, despite Goulding’s pledge at their funerals that they would, strengthened support for the Provisionals, and Mullen and Heron were quietly transformed into posthumous Provos. Their names now appear among the Tyrone dead in the republican roll of honor published each Easter in An Phoblacht–Republican News. As late as 1977 the local pro-Goulding Republican Club would make no distinction between the Provisional and Official IRA when it contributed to the Prisoners Dependants Fund. Goulding eventually expelled them.

The leading Provisional IRA figure in Tyrone after the split was Kevin Mallon, whose trial during the Border Campaign had been a cause célèbre. Accused of the 1957 murder of an RUC sergeant killed in a booby trap bomb near Coalisland, Mallon was acquitted when evidence was produced showing he had been ill treated by his police interrogators. Mallon had a strong following in Tyrone, but he kept those under his control more or less independent of the Provisional Army Council after the 1969 split until, as a former colleague put it, “he was caught up in the tidal wave”[5] and gave allegiance to the Provos.

Lynagh, McKearney, and other members of the East Tyrone ASU showed the same independent and rebellious streaks. They opposed much of the Adams agenda, especially the dropping of abstentionism, fearing that recognizing and accepting the Irish parliament would lead inevitably to endorsing the Stormont and Westminster bodies too. Had they lived, they almost certainly would have also opposed the peace process and perhaps might have brought significant numbers of the IRA in Tyrone and Monaghan with them. What distinguished their opposition to the leadership was that it was expressed in both political and military terms. Lynagh and McKearney had worked together to oppose the dropping of the absten-tionist rule and had used exactly the same arguments as O Bradaigh and O Conaill, principally that it would lead to constitutionalism and inevitably to the dilution of armed struggle. At the Tyrone-Monaghan brigade convention, which chose delegates for the Army Convention, they spoke out against the motion and clashed with the chief of staff, Kevin McKenna, a fellow Tyrone man but at that stage a strong Adams supporter. They were counting on the support of another veteran republican (J.B.) Joe O’Hagan, a former quartermaster general and IRA Executive member from Lurgan, County Armagh, who had found refuge in Monaghan. They believed they had his support, but when the brigade convention voted, O’Hagan, like the majority of delegates, went with Adams. Despite that setback, Lynagh and Paddy Kelly attended the 1986 Army Convention and voted against Adams, although they were in a minority of the brigade delegation.

Their military critique was also calculated to cause friction with the leadership. It opposed two tenets of the Army Council’s strategy as it was developed in the mid-to late 1980s, both closely associated with the Adams concept of republican struggle: namely the notion that the IRA’s war was a piece of armed propaganda and that Britain would be forced to move when enough soldiers were killed. “[Lynagh and McKearney] didn’t believe sending Brits home in boxes would work, because the British army wasn’t a conscript army like the Americans in Vietnam,” recalled a former associate.

They were working on the basis that a radical departure had to be made. The idea was either total war or no war at all, to force the British out of their bases and to make the place ungovernable. They said that either the IRA should take it to that level or finish with the war; killing the odd UDR man did nothing. They believed the “Green Book” [the IRA manual of tactics and regulations] was shit, that it was based upon the false idea that the IRA would be able to operate from its home base and at the same time be able to resist interrogation at Castlereagh. Their response was that the enemy will not allow you to survive in his bosom. Would Castro have survived if he had been in Havana rather than in the mountains? That was the question they asked.[6]

The pair had gone to Chief of Staff Kevin McKenna with a radical proposal to build a flying column, consisting of perhaps twenty or thirty trusted activists, which would be based deep in the South, with its own dedicated training facilities. The column would never break camp, in a conscious imitation of the flying columns that had run the British ragged in Cork during the 1919–21 conflict. This was meant to ensure that it would be more secure; an informer in the midst of the column would reveal himself if he separated from the group, as an informer would have to do in order to communicate with his or her handlers. The column would strike three to five times a year, aiming to cause maximum damage and disruption to the British administration. Satellite groups of two or three men would all the while attack on a harassment basis and collect intelligence for the bigger strikes.

McKenna turned down the idea, condemning it as too impractical, too ambitious, and not sustainable, and rejected their request for a separate training camp on security grounds. Friction between McKenna and the Lynagh-McKearney team intensified. There were attempts to separate the pair, apparently aimed at weakening their influence. “McKenna tried to put Lynagh in one zone and McKearney in the other,” recalled an associate. “It was possibly also an attempt to divide and conquer, to cause rivalry over assets and the like.”[7] Another IRA source suggested that McKearney may have been seen as the greater irritant: “Padraig and McKenna had never really got on. After the escape in ’83 the men were offered the choice of going to America or back to active service. Padraig got the feeling McKenna would have liked to see him take the boat. For his part Padraig was very strongly opposed to the 1986 decision [to drop abstentionism]. He felt afterward that people who were in favor of running down the war were being pushed forward.”[8]

McKenna and the Army Council did not like these militant views, but they could not afford to ignore them. Lynagh and McKearney had powerful friends and allies. Among those who backed their agenda or at least were sympathetic to it, according to IRA sources in Tyrone, were some of the most active IRA gunmen and bombers of the day. The claim is not, it has to be stressed, supported by any independent evidence, but it has been made by sources who claim a deep and close knowledge of the relationship between Lynagh-McKearney and others of a like mind. The other dissidents and malcontents, according to this version of events, included not only the bulk of the Loughgall unit but also Seamus McElwaine and Kieran Fleming, who had escaped from Long Kesh with McKearney in 1983, Antoin MacGiolla Bhride from South Derry, Dessie Grew from Armagh, and the Ardboe men Michael “Pete” Ryan and Liam Ryan, the former OC of the New York IRA who was brigade intelligence officer when the Loughgall ambush happened. The area covered by the dissenters embraced not just East Tyrone but Armagh, South Derry, and Monaghan, the most important and active operational area outside of South Armagh. Like the Loughgall ASU, McElwaine, Fleming, MacGiolla Bhride, Pete Ryan, and Grew were all to be killed in British undercover ambushes, while Liam Ryan died in an expert UVF gun attack on his bar.

Although McKenna had turned down the East Tyrone flying column proposal, Northern Command did order the unit to bomb and mortar police and UDR bases in 1985, although in this case the strategy was a limited one, aimed at frustrating the repair and renewal of security bases. The East Tyrone ASU spearheaded much of this activity, and a sort of flying column was set up—but it came together only for specific operations, and that made it very different from the full-time and secure column Lynagh and McKearney had in mind.

In December 1985 the East Tyrone ASU blew up Ballygawley RUC station and killed two policemen who were stationed inside. Then it mortared bases in Castlederg and Carrickmore. In the autumn of 1986 the police station in “The Birches” near Portadown, County Armagh, was devastated by a bomb carried to its target in the bucket of a mechanical digger. The Loughgall attack was supposed to be a repeat performance. East Tyrone was also to the fore in the offensive against building contractors. Two weeks before the Loughgall ambush, five ASU members took fifty-two-year-old Harry Henry out of his Magherafelt, County Derry, home in his stocking feet and shot him in the head. Henry’s brother owned a building firm that specialized in doing work for the security forces.

Although the IRA leadership had moved to meet some of the East Tyrone demands, Lynagh and McKearney felt it still fell far short of the sort of tactic they believed could cause a qualitative shift of fortune in the IRA’s direction. Relations between McKenna and McKearney in particular were worsening, and the number of operations carried out by East Tyrone kept to a minimum, as one dissident confidant recalled: “[McKenna] kept on knocking back ideas for operations and separating Padraig from the rest of the [East Tyrone] unit. He was excluded from the attack on The Birches for example, and it was like McKenna was trying to tell the others they didn’t need Padraig.”[9]

In the months between the 1986 Convention and the Loughgall ambush, McKearney began putting out feelers in Dublin and elsewhere in a bid to acquire weapons that would be used to arm the rebels if they decided to break free of the Provisionals. He was also in touch with the faction that had coalesced around the Provisional IRA renegade turned INLA leader Dominic McGlinchey, whose violent campaign in the South Derry countryside in the late 1970s was matched in its ferocity only by that in East Tyrone. McGlinchey, who had even less regard for the Adams leadership than the East Tyrone men, was in jail at this point, but his formidable wife, Mary, an experienced and ruthless operator in her own right, had assumed leadership of his group. The contacts between McKearney and the INLA were designed to test whether they could find common ground, but when Mary McGlinchey was killed in January 1987, shot dead by unknown gunmen as she was bathing her two infant sons, the liaison ended.

Nonetheless, in the spring of 1987, just weeks before the Loughgall ambush, the dissidents met to discuss their plans. The venue was Seamus McElwaine’s family home on the Monaghan-Fermanagh Border. “They sat down in North Monaghan and asked their colleagues whether it would be feasible to break away and form a separate organization which would put their flying column idea into practice,” recalled a former associate. “It was an informal meeting, seven or eight of them were present.”[10] Naturally, the meeting was kept secret from mainstream IRA loyalists. Another source familiar with these events added, “The gist [of the meeting] was that they [decided that they] could get arms from their own sources and move away from the Provos.”[11]

Jim Lynagh and Padraig McKearney did not live long enough to put whatever plans they had made into action. Since their deaths in the Lough-gall ambush the search for an explanation of what went wrong that day has obsessed many Tyrone republicans. Was there an informer and, if so, who was he? The obsession caused near-paranoia to grip the IRA in the brigade area. Patrick Kelly’s successor as OC, Brian Arthurs, went as far as splitting East Tyrone in two to see if this could isolate the guilty party, and for a long time brigade operations were suspended altogether while internal investigations were carried out.

Certain features of the Loughgall operation suggested the possibility of a more innocent explanation. Glaring mistakes were made in the planning and execution of the bombing which inadvertently could have put the British on the trail, mistakes that spoke of a reckless overconfidence and care lessness. There were, for instance, no probes made around Loughgall before the attack. This was routine practice in South Armagh, where, before am bushes or other IRA operations, sheepdogs were sent into adjoining fields to flush out under cover soldiers. Nor was there any effort to give the attackers the protection of covering fire just in case something went wrong. Such sloppiness at this late stage possibly indicated that other lapses had occurred earlier in the preparatory work and it is conceivable that this is how the British learned of the plan. An internal IRA investigation held afterward concluded that something like this might have happened, and that there had been a leak at the Monaghan end of the operation, although the precise source was never identified.[12] The IRA leadership remains as much in the dark about what really happened as anyone else.

The number of theories suggesting that an informer had betrayed the unit have multiplied with the years. An Ardboe woman, Colette O’Neill, came under suspicion in 1989, for instance, and was briefly kidnapped by the IRA but rescued by the RUC, which seemed to have excellent information about her plight. She later made a deal with the IRA in which she agreed not to give evidence against her abductors if the IRA left her alone. In an interview with the author not long afterward, she denied being the Loughgall informer, although she admitted that a phone call had been made from her home to members of the ASU on the day of the operation, clearing the way for the bomb to be picked up. Another theory blames a ninth member of the ASU, who was supposed to be on the Loughgall mission but missed it. Although the man protested his innocence, not surprisingly he left for England not long afterward and has not been seen since.

Some British sources have claimed that electronic surveillance of the homes of two other Tyrone IRA members, Gerard and Martin Harte, near Omagh, put the British on the trail of the Loughgall operation.[13] The same eavesdropping operation directed against their home led a year later to the death of the Hartes, lured into an ambush near their home by the SAS. The Harte brothers’ role in the Loughgall operation was to hijack the mechanical digger, and the sources say that electronic bugs picked up details of this part of the plan. Against this, local sources claim that while this may have happened, it did not necessarily mean the British would have known the target.

“Liam Ryan did the intelligence work for Loughgall, and he insisted that he had compartmentalized everything, that no single participant would have known all the details,” recalled a confidant of the Tyrone Brigade intelligence officer. “He felt [the leak] had to be at a centralized level; he would have concluded that it had to be higher up than Tyrone.”[14] Other Tyrone IRA sources corroborate this and say that some of the minor members of the Loughgall team were not told about the operation until fourteen or fifteen hours beforehand, not enough time for the British to mount an SAS ambush on the scale of that which destroyed the ASU.[15]

Another factor may have led to the Loughgall ambush. Since the mid-1980s the IRA leadership had exercised tighter and tighter control over the IRA’s day-to-day activity as it became clearer that bad IRA operations, those in which civilians were put at risk or killed, could seriously erode electoral support for Sinn Fein. During the 1983 general election a huge bomb aimed at Andersonstown RUC station had wrecked homes in the streets where Gerry Adams had been canvassing for votes not long before. The Ivor Bell–Belfast Brigade revolt against Adams during 1983 and 1984 had led to the deaths of Edgar Graham, Jimmy Campbell, and Mary Travers, daughter of the Catholic magistrate Tom Travers. Adams had warned of the need for “controlled and disciplined” IRA actions at the 1983 Ard Fheis.[16] After Sinn Fein’s poor performance in the June 1984 Euro election, he returned to the theme: “[T]here are a number of people who, while they voted for us in June 1983, may not have been able to tolerate some aspects of IRA operations,” he cautioned. “I think it is fair to say there are varying degrees of tolerance within the Nationalist electorate for aspects of the armed struggle.”[17] The ballot box was beginning to curb the Armalite.

Just before the death of Seamus McElwaine, killed by the SAS in County Fermanagh in April 1986, Northern Command got permission to vet most IRA operations in Northern Ireland in a bid to forestall further electoral damage. “There had been some bad operations, politically bad operations, and this was done to correct that,” recalled one activist. “McGuinness [the Northern commander] got authority from the Army Council to vet operations. Before that, area commanders would run through their plans in very general terms, for example, ‘I have a policeman or a British patrol,’ with the chief of staff or director of operations. Now people had to go into the detail of the operations.”[18]

IRA activists had an almost pathological fear of sharing operational intelligence with people they did not know or who did not come from their area. No one ever suggested that Martin McGuinness or any other senior figures at his level were passing on information to the British, but the very fact that the circle of operational knowledge was widened beyond those chosen to go on the mission heightened fears of leaks, surveillance and even treachery.

The vertical IRA structures introduced by Adams after his Cage 11 days combined with the greater political control exercised over operational matters made it easier, not harder, for British intelligence to penetrate the IRA’s nerve centers. The old battalion and company architecture was leaky, for sure, but whatever damage an informer did was usually confined to one small area. Even then it was a relatively easy task to isolate and identify the traitor. A well-placed agent in the remodeled IRA could by contrast cause enormous harm throughout the length and breadth of the organization and be pretty sure of getting away with it. Those on the ground who fought the IRA’s war were acutely aware of these dangers.

Loughgall was a case in point. The bombing was a Northern Command–directed operation, but Lynagh objected strongly to having to share vital details of the mission plan with others. Before the go-ahead was given for the operation he traveled to Monaghan to see Kevin McKenna to complain, and the two had a blazing row. “McKenna and Lynagh never saw eye to eye,” remembered one associate.[19]

But this alone does not explain the deep gulf of mistrust that separates many Tyrone republicans from their leadership beyond the county. To explain that, it is necessary to go back to the secret meeting of East Tyrone activists who gathered in North Monaghan in the spring of 1987 to discuss splitting from the Provisionals. “They made a big mistake by including someone who really shouldn’t have been there,” concluded an associate.[20] This was a reference to a Belfast IRA man whose family had a record of long and loyal service to the leadership. No proof or evidence has ever been produced to show that he or anyone else told the leadership about the dissidents’ plot, but in the hothouse world of the IRA that hardly mattered. “This is the reason for all the suspicion around Loughgall,” said the same source.[21]

FROM THE TIME of the Loughgall ambush onward, it was more or less open season on the Tyrone IRA as far as the SAS and British intelligence were concerned. In August 1988 the SAS killed the Mid-Tyrone IRA members Gerard Harte, Martin Harte, and Brian Mullin. In October 1990 Dessie Grew and a ferocious young gunman from Galbally, Martin McCaughey, were ambushed and killed by undercover soldiers not far from Loughgall. In June the following year the remnants of the East Tyrone ASU died in an SAS trap in Coagh, between Ardboe and Cookstown. Lawrence McNally and Pete Ryan, both Monaghan-based operators, had been lured to the village by a sighting of a UDR soldier who, because of his alleged connections to loyalists, had long been an IRA target, but their car was destroyed in an SAS gun and rocket attack. Their driver, Tony Doris, was an IRA member from Coalisland. “After McNally and Pete Ryan that was it; we had nobody left,” commented one Tyrone republican.[22]

The SAS was not finished, however. In February 1992, undercover soldiers cut a swath into the next IRA generation when they killed the twenty-one-year-old student Kevin Barry O’Donnell, twenty-three-year-old Sean O’Farrell, nineteen-year-old Peter Clancy, and twenty-year-old Daniel Vincent in a set-piece ambush in the parking lot of the Catholic church in Clonoe, between Coalisland and Ardboe. The four died after a particularly pointless machine-gun attack on Coalisland police station mounted from atop a truck as it sped through the town. Once again, according to IRA sources in a position to know, Northern Command had approved the operation.[23] In all four ambushes the British security forces’ intelligence had been excellent.

The impact of the British offensive is evident from the war statistics, which reveal a picture of a steady decline in IRA activity in Tyrone. In 1986, before the Loughgall ambush, according to the author’s analysis of the weekly IRA “War News” column published in An Phoblacht–Republican News, East Tyrone accounted for 21 percent of all IRA operations in Northern Ireland. But in 1987 that dipped to 9 percent. In 1988 there was some recovery in IRA activity to 16 percent; but thereafter the decline resumed: in 1989 to 13.5 percent, in 1990 to 16.4 percent, in 1991 to 8 percent, in 1992 to 11 percent, and in 1993, the year before the first cease-fire, to 9 percent.[24] The IRA death toll tells the same tale. Up to April 2000, the IRA in Tyrone had lost 53 members, the highest death toll for any brigade area. But over half, 28, were killed in the five years between May 1987 and February 1992, compared with 25 in the seventeen years between 1970 and 1987. In other words, IRA deaths in Tyrone increased fivefold after the Loughall ambush.

The SAS was not alone in putting East Tyrone in its deadly sights. The years 1987 and 1988 also saw the beginnings of a concerted UVF assault against republicans and nationalists in the brigade area that in its way was more enervating than the undercover military ambushes. In an important sense the East Tyrone Brigade brought the calamity upon itself. The killings of Harry Henry in April 1987 and then a year later of Ned Gibson, a Protestant sanitation worker who was also a part-time soldier in the UDR, shot dead as he collected garbage in Ardboe, were regarded by Tyrone’s unionists as overtly brutal sectarian assaults, and they responded in kind. It was, however, the nature of the IRA’s response to the loyalist offensive that made the carnage in Tyrone such a significant milestone on the road to the 1994 cease-fire.

In the first twenty years of its existence the Provisional IRA had a simple if brutal attitude toward loyalist killings. This was to retaliate with excessive but deliberately directed violence against the unionist community. Two notorious incidents stand out as examples of what in practice this policy meant. The first occurred in early January 1976 when Provisional IRA gunmen halted a minibus carrying textile workers home from work near Kingsmills in South Armagh. They singled out the lone Catholic on board, told him to get out of the way, and then lined ten Protestants up against the side of the bus and riddled them with automatic fire. The incident was so horrific that the IRA was reluctant to admit responsibility, and instead a cover name, the South Armagh Republican Reaction Force, was used. The Kingsmills slaughter was the IRA’s response to the UVF killing in the days before of five South Armagh Catholics, members of two families.

The second example came in January 1981 when an eleven-man IRA unit, reportedly led by Jim Lynagh, broke into Tynan Abbey, an enormous mansion set in eight hundred acres of lush farmland near the Armagh-Monaghan Border, and killed its two occupants, eighty-six-year-old Sir Norman Stronge and his merchant banker son, forty-eight-year-old James. They then planted incendiary devices that set the abbey alight, destroying it in the fire. The family had been in Tynan for eight generations, and the Stronges were scions of the unionist establishment. Sir Norman had been Speaker of the Stormont parliament for a quarter of a century and his son had been a unionist MP at Stormont; his great-grandfather had been Speaker of the old Irish House of Commons. The reason for the IRA attack was clear. Five days before, UDA gunmen had made an unsuccessful attempt on the life of the Tyrone republican leader Bernadette McAliskey, which had left her and her husband, Michael, barely alive. The UDA attack was the latest in a series of assassinations of prominent nationalist and republican figures. The killing of the Stronges was designed to bring them to a halt.

The point about the Kingsmills and the Stronge killings is that most Provisional supporters believed that they had worked and that a speedy and violent IRA reaction of that caliber was the best, perhaps the only, way to stop such loyalist violence in its tracks. Such retaliations became accepted as part of the Provisionals’ view of the world. “It’s a lesson you learn quickly on the football field,” commented a Tyrone republican and GAA veteran. “If you’re fouled, you have to hit back.”[25] The reprisal policy was part of what the Provisional IRA was about; it reached deep down into the group’s Defenderist roots.

But by the time the Mid-Ulster UVF began its killing in the autumn of 1988, IRA policy had begun to change. In November 1988 the UVF targeted the home of the Moortown Sinn Fein councillor Francie McNally, not far from Ardboe on the shores of Lough Neagh. Shots were fired through the kitchen window, and McNally’s younger brother Phelim was fatally wounded as he played an accordion. Three months later, in February 1989, the UVF struck once more against Sinn Fein. This time the target was fifty-eight-year-old Magherafelt SF councillor John Joe Davey, who was cut down in the driveway of his isolated rural home. Davey had been named under privilege in the House of Commons by the DUP MP for Mid-Ulster, the Reverend William McCrea, as an accomplice in the killing of a local Protestant security contractor in 1986. A veteran of the Border Campaign, Davey had twice been interned by the Northern authorities. The UVF was exacting revenge.

Army Council policy in relation to reprisals had changed significantly by this stage. Retaliations of the Kingsmills and Stronge variety were banned, as one IRA source familiar with the policy change explained: “John Davey was the first Sinn Fein councillor to be shot, but there was to be no retaliation because of [an Army Council directive] against political assassinations. Although there could be exceptions—Ken Maginnis [ex-UDR captain and the former Fermanagh–South Tyrone unionist MP] was one—and there could be special requests, basically that type of reprisal was politically unacceptable to Adams and company.”[26]

Instead the IRA was allowed to strike back only at named, identified targets, and this meant that only those who could be shown to have been directly involved in the loyalist killings or who were known to be pulling their strings in the background were legitimate targets. Again Northern Command would vet each operation, and local brigades would have to justify the choice of targets. That was necessarily a drawn-out process that delayed the IRA response. “The time frame was crucial, otherwise the message was lost,” recalled a Tyrone republican. “Loyalists retaliated fast, while in the IRA they had to battle for permission to strike back. You virtually needed a jury trial.”[27]

Retaliation for the McNally and Davey killings came in the first week of March 1989 when East Tyrone IRA members sought out Leslie Dallas, a UVF member and a leading member of one of the four UVF families in the East Tyrone–South Derry area. He was gunned down in a garage in Coagh, not far from Ardboe, but the shooting was a disaster for the IRA. Two elderly Protestant men, Ernest Rankin and Austin Nelson, neither of whom had any association with the UVF, were caught in the gunfire and shot dead.

In Belfast, Gerry Adams distanced Sinn Fein from the killings. Referring to the deaths of Rankin and Nelson, he echoed the Army Council line. “Our position is clear,” he said. “Sinn Fein does not condone the deaths of people who are non-combatants. There can be no legitimate reason for any uncertainty about Sinn Fein’s attitude to such killings.”[28] On the ground in East Tyrone, the Provisionals’ grassroots, by contrast, had demanded a much more drastic response. “The cry was a councillor for a councillor, for Willie McCrea or a bomb in the council chamber,” explained a Tyrone republican. “That was the gut feeling of the rank and file.”[29] They didn’t get their way.

After the Coagh shootings, the UVF campaign in the East Tyrone, North Armagh, and South Derry areas intensified. Their targets included not just uninvolved Catholics but IRA and Sinn Fein members and their families, and its effect was to deeply undermine republican morale. “In Tyrone the SAS went for IRA members while the UVF went for the families,” was how one republican source described the tactics in these years.[30] The UVF killed twice more in 1989, five times in 1990, and fifteen times in 1991, eight of the victims gunned down in the first three months of the year. Again the facts and figures speak for themselves. Between 1988 and August 1994 eighty-six people died violent deaths in the East Tyrone operational zone, and the UVF was responsible for forty of them, nearly half the slaughter. Davey’s death began an open season on Sinn Fein councillors and activists. Fourteen Sinn Fein members were killed in the four years that followed the killing of the Magherafelt councillor, over half of all the Sinn Fein personnel killed since 1970. The IRA was powerless or unwilling to stop it.

Relatives of Republican activists were also picked off, sometimes in exceptionally brutal circumstances, sometimes in circumstances suggesting that the loyalist killers had excellent intelligence. Roseanne Mallon, for instance, shot dead in Lisgallon in May 1994, was related to the most senior IRA figure in the county. In one of the worst examples of this kind of killing, a mother of five, Kathleen O’Hagan, was gunned to death in front of her children at her home near Cookstown in August 1994. Her husband had served eight years for possessing an IRA gun, and that made her a target, as it made by implication the spouses of other IRA personnel. Mrs O’Hagan was seven months pregnant at the time of her brutal death. There were other clues that the killers were well briefed. Patrick Shields, shot dead with his son Diarmuid in January 1993 in his grocer’s shop near Dungannon, had been in the IRA in the 1970s but had quit long before. He was, however, still a contact of Kevin Mallon’s, by this stage no longer a Provisional but a dissident sympathizer. Not many people would have known that. A month later his son’s girlfriend, Julie Statham, overwhelmed by grief, committed suicide.

The IRA’s new, restrained retaliation policy had two effects. The IRA concentrated all its energies in the hunt for the head of the Mid-Ulster UVF, Billy Wright, who was based in Portadown and who was believed to have had a hand in most if not all the killings. The IRA made at least five attempts on Wright’s life, including a booby trap bomb placed underneath his car, but the UVF leader lived a charmed life and survived them all. He was later shot dead by the INLA under extraordinary circumstances, inside the Maze prison, but the IRA’s inability to deal with him or his associates discredited the organization. “They were always talking about a night of the long knives against the UVF, but it never did happen,” remembered one Tyrone IRA man.[31] Another recalled approaching a prominent Sinn Fein politician in the county after learning of a UVF threat against his family: “His response was ‘I wonder if we could get in touch with the UVF and sort it out.’”[32] A week later the UVF killed his brother.

The IRA’s failure to stem the loyalist killings struck at the core of its raison d’être. If it couldn’t protect its own, especially in Tyrone, wondered many republicans, how could the IRA expect to drive the British out of Ireland? “As the killings grew, the demand to do something grew as well,” recalled one activist. “People were scared because it seemed the loyalists had a free hand. People were afraid to be identified with Sinn Fein, not just the IRA. You could be shot for having the same name as someone in Sinn Fein like poor Tommy Molloy. Meanwhile the IRA was doing nothing to protect people.”[33] Molloy was killed apparently because he shared a surname with Francie Molloy, the leading Sinn Fein figure in the area, who later became an Assembly member for Mid-Ulster. By the end of 1990, according to republican sources in the area, Sinn Fein was having great difficulty persuading people to run in council elections.[34] The grassroots demoralization that flowed from all this nourished the psychology of ceasefire, making the peace process acceptable and even welcome, as one astute observer of Tyrone republicanism noted. “People are terrified of going back to war because of their memories of that UVF campaign,” he said.[35]

The IRA leadership’s new attitude toward loyalist killings sometimes meant that lies were told about the true allegiance of some victims. At least three Sinn Fein councillors killed by the UVF in the area were also key IRA activists, but this was never acknowledged even years later. When Liam Ryan was shot dead by the UVF at his Battery Bar near Ardboe along with a civilian, Michael Devlin, his IRA membership, never mind his role as brigade intelligence officer, was denied by Sinn Fein, although three weeks later an IRA firing party did fire a volley of automatic shots over his grave. The IRA’s worst loss at the hands of the Mid-Ulster UVF came in March 1991, when three young IRA members were shot dead as they drove into the car park of Boyle’s Bar in the strongly republican village of Cappagh, County Tyrone. Twenty-three-year-old John Quinn, seventeen-year-old Dwayne O’Donnell, and twenty-year-old Malcolm Nugent were cut down in a hail of bullets, and a Catholic noncombatant, fifty-two-year-old Thomas Armstrong, killed by a stray round as he used the toilets in the bar. The IRA did not acknowledge the three for a year, and they were depicted in Sinn Fein propaganda as innocent Catholic victims, the result of an order relayed to the Tyrone Brigade on behalf of Northern Command by Jim Gibney, a senior Belfast-based adviser of Gerry Adams.[36] The deception fooled no republicans in Tyrone and very few in the rest of Northern Ireland, but was aimed at currying sympathy south of the Border, where people were much less likely to be aware of the truth.

“Adams’s attitude was always to encourage the view of nationalists as being the underdog, the ones at the receiving end of this sort of violence,” recalled a senior IRA source. “He would say we want to be seen as the oppressed, and that was why the Cappagh men were disowned.”[37] Denying the Cappagh dead enraged many Tyrone republicans, but the IRA on the ground was assured that there were good reasons. “It was to do with making life easier in the South, particularly in relation to the safe houses and the like which we could get, especially from Fianna Fail people,” explained one former senior figure. “After disasters like Enniskillen, houses were shut to the IRA; after Cappagh when what happened [was] seen as security force collusion with loyalists to kill Catholics, the houses open again.”[38] There was a political dimension to this as well. If Fianna Fail supporters could open their homes to the IRA, then perhaps the Fianna Fail leadership could do the same at a political level with Sinn Fein. This provided the real reason for the shift in Army Council policy on loyalist killings. The change dovetailed into Gerry Adams’s quest for pan-nationalist unity that in 1988 and 1989 was at the center of his secret diplomatic overtures to the Fianna Fail taoiseach, Charles Haughey.

THE FUNERALS of the Loughgall ASU were some of the biggest seen in Ireland since the hunger strikes. Jim Lynagh’s removal and burial became an occasion for the Provisional grassroots to demonstrate a disdain for the Republic and its institutions that had been sharpened by suspicions that the Irish police, the Gardai, may have played a part in laying the ambush, possibly by passing on intelligence to the British. Lynagh’s body was escorted back across the Border on the Monday after the killings by hundreds of IRA supporters. At Emyvale in County Monaghan the cortege stopped in the main street and was joined by an IRA color party. From a side entry stepped three masked IRA men armed with automatic rifles. They fired three single shots over the coffin and then a wild volley into the air that roused the crowd. As they turned to melt back into the sea of faces, they found their way blocked by a carload of the elite Garda antiterrorist unit, the Task Force armed with Uzi submachine guns. Seeing this, the crowd surged forward and manhandled the police car into an empty drain, upturning it with one of its plainclothes occupants still inside. Other policemen fired wildly in the air, and for a few moments it looked as if a serious confrontation might follow. Later that day a Sinn Fein picket gathered outside Fianna Fail’s headquarters in Lower Mount Street in central Dublin. Angered by Fianna Fail denunciations of the IRA in the wake of the ambush, the picketers strung a banner across the footpath which read, “Fianna Fail, ‘The Republican Party,’ Collaborators with SAS Murderers.”

Lynagh was buried two days later, on May 13, and Gerry Adams gave the oration at his graveside and lashed the Fianna Fail government and its leader, Charles Haughey. “A few short months ago,” he said,

the people of this State elected a Fianna Fail government of sorts. Their leader made many brave noises about a British withdrawal being a prerequisite for peace in this island. He described the Six-County State as a nonviable social and economic unit. He chose Bodenstown to denounce British policy, [Garret] FitzGerald’s collusion in that policy and the actions of the British Crown Forces. That was when he was looking for votes. I have some questions for all Fianna Fail supporters and for all nationalists. Did you elect a government to support Thatcher’s terrorism? The British government understands Charles J. Haughey… as it understood FitzGerald and Spring. It has always understood the shoneen clan—it bought them off with partition. It does not understand the Jim Lynaghs, the Padraig McKearneys or the Seamus McElwaines. It thinks it can defeat them. It never will.[39]

It would have been later the same day or not long afterward that Tim Pat Coogan was ushered into Charles Haughey’s presence to hand over the lengthy letter from Father Alec Reid that outlined Gerry Adams’s proposals for an alliance between Fianna Fail and Sinn Fein and the extraordinary offer of an IRA cease-fire. The letter, fifteen pages and 7,000 words long, had been written two days before, when the gunfire over the coffin of Jim Lynagh was still echoing around the streets of Emyvale and the angry shouts of Sinn Fein picketers were ringing outside Fianna Fail’s offices.

12. “The War of the Twilight”

As the year 1988 dawned, there were two sharply contrasting moods in the IRA. At grassroots level, disappointment at the loss of the Eksund the previous November was more than compensated for by the subsquent disclosure that other shipments had gotten safely through from Libya, well over 120 tons according to reports in the British and Irish media, and at the realization that the IRA was now better armed than at any other time in its history. Ever since the first shots were fired at British troops in 1970, the Volunteers had dreamed that one day they would have plenty of modern weaponry with which to take on British forces, and now that day had arrived. Whatever doubts they had about the political direction the Adams leadership was taking, whatever questions had been raised by the movement’s recognition of the Irish parliament a few months before, now vanished like snow off a ditch in spring.

At the level of Army Council, the mood was entirely different. Those who had spent the previous three years putting the Libyan smuggling venture together and who had been planning the “Tet offensive” knew just what a disaster the betrayal of the Eksund was. They knew it spelled military stalemate for the IRA, robbing the offensive of the one element that would have guaranteed a major impact—surprise. But that was not the only disheartening aspect of the Eksund’s loss for IRA leaders. On board had been military mortars that could have devastated British barracks and RUC bases throughout the North, enabling the IRA to launch damaging attacks from safe distances. Also gone were the powerful 106-millimeter canons that were to be used to sink the Royal Navy’s patrol boat that cruised Carlingford Lough. There had also been plans to sink boats in Belfast harbor with the weapons, blocking access to the city’s docks and disrupting trade. But these ambitious schemes too would now have to be scrapped.

Also lost with the Eksund was the ambitious political agenda that the Army Council expected would be advanced by the “Tet offensive.” The immediate objective of the plan was to force the British into reintroducing internment against the IRA, the weapon last used in 1971 but discredited then as both ineffective and iniquitous. Internment would, the IRA leadership calculated, bring a number of political benefits. The use of the weapon would likely anger and alienate nationalist opinion in both parts of Ireland as it had done in 1971, thereby fueling support for the IRA’s campaign, and would also demonstrate the bankruptcy of British policy in Ireland, discrediting Britain in the eyes of world opinion and perhaps provoking public opinion in Britain to search for a more radical and long-lasting solution to the Irish problem.

The Army Council may have miscalculated the IRA’s ability to use the Libyan weaponry to the best advantage. Not everyone in the IRA was convinced that the organization had the wherewithal to deliver such an ambitious enterprise, as one middle-ranking commander recalled. “The strategy was to mount a massive campaign, but I had been going around the units and I was not convinced it would work,” he said. “We weren’t capable of that. There hadn’t been enough organization, and our security and training weren’t good enough. We didn’t have enough intelligence work done either. If we had tried to mount it, I think it would have been a disaster. I believed we needed a lot more time, but people had got carried away with all the heavy gear.”[1] In a sense that did not matter, for the purpose of the “Tet offensive” was, like that of its Vietnamese original, to show the world how deep and violent the opposition in Ireland was. After all, the Vietnamese had been given a bloody nose during their offensive, yet the violence had helped transform American public opinion.

The plan came so close to success that the Army Council at one point succumbed to bragging. After the Villa had landed 105 tons of arms in October 1986, the leadership’s spirits had soared, and at the following Easter’s republican commemorations it could not resist dropping a hint of what was in the pipeline. In tune with the military mood, the Council’s statement was distinctly dismissive of electoralism:

We agree with the building of a revolutionary, republican party in the course of the armed struggle and we refuse to be over-euphoric with electoral victories or disillusioned or deterred by electoral setbacks. The British will only be talked out of Ireland through the rattle of machine-guns and the roar of explosives…. It has been a long time since the Irish Republican Army has felt so confident, so sure of victory, so happy about the future even though for many of us the coming fight may mean imprisonment and death.[2]

The move to abandon abstentionism in the South became part of the plan. At the 1986 IRA Convention the leadership sold the notion to the rank and file that recognizing and taking seats in the Dail in Dublin would help neutralize the Republic’s government and isolate the British. If Sinn Fein won seats to the Dublin parliament, it was argued, then it would be much more difficult for the Republic’s government to join with the British in an assault against the IRA. Furthermore, if Sinn Fein TDs held the balance of power in the Dail, as could happen, then the party would hold the whip hand and the IRA would be free to bomb and shoot at will. “We knew there were going to be major casualties and draconian measures introduced like internment,” recalled one IRA source familiar with the “Tet” strategy. “That was the story at the ’86 Convention: ‘We’ll have a couple of TDs by then, and if [the Republic’s government] introduce[s] internment we’ll be able to block it.’”[3]

There would be other succulent fruits to pluck if all went according to plan. If the British responded by reintroducing internment, the deal agreed with Nasser Ashour of the Libyan Intelligence Service envisaged Qaddafi sending more consignments of weapons to Ireland and pumping more money into the IRA. And there were other, enticing possibilities. “By inflicting such big casualties, you’d get the support of radical governments elsewhere in the world,” recalled the same source.[4]

The capture of the Eksund changed everything. The more astute among the IRA leaders fully realized the consequences. “We were gung ho before at the prospects, but now it was the war of the twilight,” recalled the source. “It was over, and it led directly to a stalemate situation which then fed into the peace process.”[5] The IRA’s grassroots were jubilant over the organization’s newfound strength, but even this was illusory. The IRA had lots of weapons, but it was by no means certain that they were the best that could be had. The value of the AK-47s, for example, had been exaggerated. “The Volunteers thought they could fire round corners,” remembered the same source. The heavy Soviet machine guns, the “Dushkies,” were pretty much useless. “They took three men to carry and only fired eighty rounds to a belt; they fired too slow,” recalled a rural IRA activist.[6] The SAM-7s were virtually obsolete. They dated back to the 1960s, and the batteries in the firing mechanisms were dead and useless. Without Libyan assistance, and that was cut off when the Eksund was lost, they could not be replaced, and so the SAM-7s stayed in their dumps while the IRA’s spin doctors did their best to exaggerate their threat. The IRA had scores of Webley revolvers, but their ammunition was of too low a caliber for its needs, while many of the millions of rounds of ammunition shipped from North Africa had been rendered unusable by old age. Qaddafi had been generous to the IRA, but his gifts had come mostly from the back shelves of his arms stores.

The IRA had plenty of Semtex, but that, really, was all. In the coming years the inventive resources of the IRA’s engineering department would be stretched to the limit as improvised weapons were developed one after the other. Homemade rocket launchers, grenades, coffee jar bombs, booby trap car bombs, and mortars poured out of the IRA’s factories hidden deep in the Republic, each device designed to utilize the explosive power of Semtex. At the time all this was seen as evidence of the IRA’s ingenuity, skill, and strength, even though in fact it was a symptom of the weakness caused by the loss of the Eksund.

The IRA leadership was forced to make do with what was available, but the big military breakthrough was now as elusive as ever. The plan put together by the director of operations, Tom “Slab” Murphy, and approved by Chief of Staff Kevin McKenna before the capture of the Eksund, envisaged a three-pronged offensive that would start in Northern Ireland and then spread to continental Europe, where British army contingents in West Germany would be targeted, and to England, where the targets would be military, political, and economic. Following the Eksund’s loss, IRA structures in the Republic were disrupted for weeks by the huge twenty-six-county-wide search for the earlier shipments, but nonetheless the go-ahead was given to begin the campaign in Europe.

A probing attack had been made in early 1987, when a 300-pound car bomb exploded outside the officers’ mess of the British army’s West German headquarters at Rheindahlen, injuring thirty-one people. It looked like a reasonably successful operation, but, in fact, the IRA had had a miraculous escape. The only reason people were not killed was that the ASU had been unable to position the car bomb close to the mess, because the car park was full of vehicles. Unknown to the IRA, most of the vehicles blocking their way were owned by West German military officers who had been invited to spend a social evening with their British counterparts. Had the car bomb exploded as it was supposed to, many of these officers could have been killed and the beginning of the IRA’s European campaign would have become a diplomatic and military disaster.

The campaign proper began on as bad a note for the IRA when on Sunday, March 6, 1988, two seasoned IRA activists, thirty-one-year-old Mairead Farrell and thirty-year-old Danny McCann, and a young, twenty-three-year-old bomb maker, Sean Savage, were gunned down by the SAS as they made their way out of Gibraltar, Britain’s quaint cosmopolitan colony perched on the southern tip of Spain. The trio, who were all from West Belfast, were unarmed when shot, and this helped transform their subsequent inquest into a virtual trial of Margaret Thatcher’s tough approach to the Northern Ireland problem. A celebrated Thames Television documentary, Death on the Rock, produced eyewitness evidence that supported the allegation that the three had been shot down in cold blood while trying to surrender, a contradiction of the version of events put about by the Thatcher government, which implied that the SAS believed they might have been about to detonate a bomb or reach for weapons when killed. An inquest jury cleared the SAS, but the controversy lingered on.

The IRA’s plan was to place a large Semtex car bomb in the center of the colony and detonate it just as a British military band assembled for a weekly changing-of-the-guard ceremony outside the governor’s official residence. Usually some fifty soldiers took part in the parade, and the target was doubly tempting, offering the IRA substantial casualties in one of the few remaining British colonies. If it worked, the attack would be a military triumph full of political symbolism. The three were on a final scouting mission when they were killed, and had left an empty car in a parking lot near the band’s assembly point, presumably to reserve a spot for the real car bomb two days later. On the trio’s way back to the border with Spain, the SAS struck.

At the inquest, held amid considerable publicity in Gibraltar itself, it became clear that substantial British intelligence resources had been deployed to forestall the attack, and it seemed there was little that the authorities did not know about the IRA’s plans. Apart from the SAS soldiers, teams of MI5 “watchers” had been assigned to follow the ASU in Gibraltar. The Garda Special Branch had watched the team leave Dublin airport for Malaga, but while the Gardai did not inform the British, the IRA later learned, Interpol in Vienna was told. Interpol in turn contacted Spain, and via the Spanish police the British were told the team was on its way. IRA intelligence later established that the Spanish had sought assurances from the British that if they tailed the three and kept MI5 and the Gibraltar police informed of their movements, they would not be killed. The British reneged on the deal.[7]

The Gibraltar operation was a disaster for the IRA, although it probably did win the subsequent public relations battle with Thatcher’s government. The true extent of the calamity for the IRA was, however, never acknowledged by the organization and was kept hidden from the rank and file. The story of Gibraltar is the story of an operation that should never have been allowed to happen.

In the weeks leading up to the operation, warning signals were clearly visible to IRA GHQ and senior Army Council personnel—red flags in military parlance—which indicated that the British had access to the IRA’s plans in Europe. They were ignored.

ALONG WITH the Gibraltar bombing, the IRA was planning another European spectacular to open its Continental campaign. The IRA had learned that the British foreign secretary, Sir Geoffrey Howe, traveled regularly to Brussels for meetings of NATO and European Community ministers, and, crucially, there was a regular pattern to his movements in the city. He was driven to his destination by the same route each time. A plan was devised to assassinate him by means of a remote-controlled bomb placed in a car parked at the side of the street. When Howe’s car passed, the bomb would be detonated and one of Mrs. Thatcher’s most senior ministers blown to pieces.[8] The operation was approved by the IRA leadership, and the bomb put in place. The IRA team charged with carrying out the mission arrived, but Howe’s car, for the first time in months, failed to turn up. The IRA concluded that British intelligence had discovered the plan and changed Howe’s itinerary. The bomb was dismantled and left in a lockup garage in Brussels where, in January 1988, Belgian police found it.

There was, however, one puzzling aspect to the affair. The British clearly knew all about the IRA plot, but they did not tip off the Belgian authorities, who could easily have arrested the ASU. That would have been something of a coup for all concerned. The reason the British and Belgians let matters lie was that the Gibraltar and Brussels operations overlapped. The IRA chiefs who had planned the Gibraltar attack and two members of the Gibraltar ASU were also involved in putting together the attempt on Sir Geoffrey Howe’s life. Had the authorities moved against the IRA in Brussels, thereby indicating that there had been a leak or treachery, GHQ would have been obliged to abandon the Gibraltar operation and Mrs. Thatcher would have been denied the chance to give the IRA a bloody nose in the streets of her colony.

Even so, the evidence suggests that the British knew weeks in advance about the IRA’s plans for Gibraltar and that the IRA should have suspected this. The original intention had been to bomb the Rock in December 1987, but the operation had to be abandoned when suddenly and for no obvious reason the changing-of-the-guard ceremony outside the governor’s residence was cancelled. But early in the New Year the IRA returned to Gibraltar and found that the parade had been resumed. The intelligence on the operation was resubmitted, and the bombing was once again authorized by the IRA leadership. No effort had been made to discover why the parade had been halted at a crucial point during the execution of the first plan or if this might have indicated that the British knew something about the IRA’s intentions. Nor did anyone in the IRA hierarchy link that with the failed attempt on Sir Geoffrey Howe’s life.

The Army Council ordered an internal inquiry into what had happened at Gibraltar and blamed loose talk by members of the ASU, three of whom were dead and could not answer back. The investigation claimed there had been a leak in West Belfast and one in County Louth, just across the Border. There was other independent evidence to indicate carelessness on the part of some of the team which suggested there was validity in the complaint. Just after the killings a West Belfast republican told the author that Danny McCann had been seen drinking in a Lower Falls social club a week before, and nearly everyone who knew him could tell he was going on an operation. “He had dyed his hair,” the source explained.[9] McCann had been allowed back into the IRA not long before and forgiven for his rebellious liaison with Ivor Bell. The IRA inquest did not, however, consider the possibility that there had been a leak or a traitor at a higher level. Given the occurrence of the Gibraltar debacle within months of the loss of the Eksund and after the Loughgall massacre, that possibility should have been considered.

THE CELLULAR REORGANIZATION of the IRA prompted by Adams and his colleagues in Long Kesh in the mid-1970s had been motivated by the alleged looseness of the old company and battalion structure. But the changes had failed to stem security breaches or agent infiltration and recruitment on the part of the British. In fact there has always been a strong school of thought in the Provisionals that held that the centralized control systems introduced in the mid-1970s may have facilitated rather than hindered such penetration. Not least it meant that one well-placed agent in the new IRA could do more damage than a dozen informers distributed around the old company and battalion system. In the months after the betrayal of the Eksund, more and more evidence surfaced suggesting that the IRA had been infiltrated in this way, at a high level and in different sections. So serious was the problem that it effectively hobbled the revised “Tet offensive.”

The autumn and winter of 1989 provided the evidence. In October a mass breakout of remand prisoners from Crumlin Road jail in Belfast was frustrated by the British. At least thirty IRA men had planned to escape by using smuggled Semtex to blast their way out of the central block. The Belfast Brigade had organized backup. A bomb would blow a hole in a perimeter wall while up to a dozen cars had been positioned around the prison to whisk the escapers to freedom. But on the appointed day the British were waiting in large numbers and clearly knew everything there was to know. A smuggled “comm” from remand prisoners to a colleague on the outside a few months later made the prisoners’ suspicions clear. The secret communication, apparently intercepted by the authorities, was quoted in Jack Holland’s Hope against History and read, “Our conclusion is the BB [Belfast Brigade] may well be compromised.”[10] It was one of several bad operations highlighted by the prisoners.

There were other telltale signs of serious penetration by the British. The original plan for the post-Eksund offensive had called for the establishment of flying columns along the Border and in Belfast and Derry. The columns, which would be brought together for specific operations and then dispersed until needed again, were to be the cutting edge of the first assaults, but in the wake of the Loughgall and the Eksund setbacks, GHQ and Slab Murphy drastically scaled down the plan. Eventually a single, experimental flying column was put together under the command of Michael “Pete” Ryan from East Tyrone, composed of around twenty of the IRA’s most experienced and trusted operators from throughout the North who were specially drafted in.

The column’s first and, as it turned out, only operation was devised by Northern Command, two of whose staff vetted the details. The plan was shared with the column members just before it was put into action. The aim was to destroy a large British army checkpoint at Derryard, near Roslea on the Fermanagh-Monaghan Border, and kill perhaps a dozen soldiers. While some members of the column sealed off approach roads to prevent civilians from stumbling onto the scene, a truck carrying eleven IRA men was to drive into the checkpoint and attack the base with two heavy machine guns, half a dozen automatic rifles, grenades, and a flamethrower, one of a consignment that the Libyans had added, unasked, to the IRA’s shopping list. Under the cover of all this fire the column would then place close to the soldiers’ living quarters a 400-pound van bomb, which would cause carnage when detonated. The operation went according to plan, and two soldiers, members of the King’s Own Scottish Borderers, were killed in the initial phase. Suddenly the column itself came under attack. Heavy gunfire was directed at its members from fields about fifty yards away, while a British army Wessex helicopter appeared from nowhere over a nearby hill. The column fled, leaving behind the primed van bomb.

To the outside world the Roslea attack looked like a success. Unionists protested angrily and called for a tough security response, while the media pondered the significance of this sudden intensification of IRA violence. But the truth was that the flying column’s first outing had been betrayed, and its members were lucky to have survived. “It was a disaster,” said a source familiar with the operation.”[11] The van bomb failed to explode, and an internal investigation concluded that the detonating mechanism had been sabotaged, a sign that a traitor had been at his or her work. A limited number of people had known about the operation, and although no one was pointing fingers, at least openly, the failure at Derryard illustrated for many IRA activists the dangers of sharing operational details with Northern Command or indeed with anyone outside their tight, trusted circle.

That was the end of the flying column idea. “The fear of a high-level informer made people afraid to go on column operations,” recalled another IRA source.[12] Not only had an informer sabotaged the original “Tet offensive,” but the revised campaign had been undermined apparently in a similar fashion. The loss of the Eksund had dealt a devastating blow to the IRA’s ambitions, and now the campaign patched together in the wake of that disaster was falling to pieces. Slowly but surely the IRA’s military options were being closed off.

A MONTH LATER one of Gerry Adams’s closest advisers and confidants, Danny Morrison, was arrested in a house in Andersonstown, West Belfast. Inside police found the disheveled, exhausted, but extremely relieved figure of Sandy Lynch, a Special Branch agent who had infiltrated the IRA in North Belfast. Financed by the RUC, Lynch had purchased a car, started taxiing, and offered his services to the IRA, who had gratefully accepted. It was one of the oldest tricks in the intelligence book, yet it always seemed to work. Eventually Lynch came under suspicion, and he was lured to the Andersonstown house, where he was interrogated and quickly confessed. Those questioning Lynch were members of the IRA’s security department, a specialist counterintelligence unit established as part of the Cage 11 reorganization plan.

The arrest of Danny Morrison proved to be a serious blow to the Adams leadership. A skilled and affable propagandist, Morrison had revamped Republican News back in the mid-1970s and overseen the “fusion” of the Belfast paper and An Phoblacht in Dublin. Regarded as a figure who could always be relied upon to loyally echo the Adams line, Morrison rose through the ranks. A member of GHQ staff, he was the IRA’s and Sinn Fein’s director of publicity and by 1989 had sat on the Army Council for the best part of a decade.

Morrison found himself in the Andersonstown house because of the activities of another informer, Joe Fenton, or rather because of the Belfast brigade’s decision in February 1989 to kill Fenton before he could be thoroughly interrogated and the extent of his treachery—and his dealings with key brigade staff—discovered. The decision had angered senior figures in the IRA, who realized that a golden opportunity to uncover a network of spies had been lost. There were angry accusations of a cover-up and favoritism toward the relatives of key West Belfast figures. To quell the unease, the Army Council ordered that in the future one of its number would have to check whether a suspected informer had been properly and fully questioned before any action could be taken. The British security forces had nicknamed Morrison “The Lord Chief Justice” after he took on this role, and they clearly relished the prospect of catching him red-handed. On January 7, 1990, their opportunity came. Almost as soon as Morrison turned up at the house, soldiers and RUC officers swooped. He was arrested and subsequently convicted and sentenced to eight years.

The significance of the swoop was that it suggested that the IRA department whose job it was to track down and kill informers had itself been penetrated by the British. The RUC Special Branch and British military intelligence had long identified the IRA’s security department as a prime target for infiltration and for very good reasons. The security department had an intimate and unrivaled knowledge of the organization’s affairs. Its job was to investigate every botched, failed, or aborted operation to see whether or not the IRA had been compromised. It was work that meant that the security department would be aware of the identities and detailed role of IRA members throughout the organization. The department also had the job of vetting recruits. It was the Achilles’ heel of the reorganized IRA. An agent placed inside the security department could provide priceless information. There already had been strong suggestions that this had happened. In July 1988, Markets IRA man Brendan “Ruby” Davison was shot dead by the UVF at his home. An irritated RUC later let it be known to the media that the loyalists had killed a key double agent. The RUC would have been grossly negligent had it not made considerable efforts to replace Davison, and Morrison’s arrest suggested that it may well have succeeded. The Sandy Lynch affair added to a general sense of paranoia in the IRA, inhibiting its ability to intensify the revised “Tet offensive” to the levels hoped for by the Army Council.

Between them these events pointed to the existence of agents or leaks at leadership, Northern Command, European department, Tyrone Brigade, Belfast Brigade, and security department levels. It was difficult to see how the IRA could have been more thoroughly compromised.

NOT EVERYTHING the IRA touched turned to dust. The European department resumed its campaign on the European mainland and enjoyed immediate success, killing three RAF personnel in one weekend in May 1988, but inevitably, it seemed, there were botched operations leading to civilian deaths, seven in all over the next two years, including that of a six-month-old baby girl and the German wife of a British serviceman. Ten members of the European department were arrested, again suggesting more leaks or treachery. The IRA’s England department enjoyed better luck. Between August 1988 and July 1990 its members killed fourteen British servicemen, eleven in a single bomb attack on the Royal Marine School of Music in Deal, Kent. The IRA also assassinated Ian Gow, Margaret Thatcher’s close friend and a political ally of Northern Ireland’s unionists. In two bombings it came close to causing multiple British army deaths. Later MI5 would make serious inroads into the IRA’s English department, so many that the IRA was forced to virtually abandon the use of “sleeper” units, groups sent in to merge with the local population and then activated, but the first two years of the campaign in Britain had shown no obvious sign of agent penetration.

In Northern Ireland, Adams’s strategy of encouraging pan-Irish nationalist unity was by this stage formal and public republican policy, although the details of his discussions with Haughey and the contents of the “stepping stones” document remained a closely guarded secret even from the Army Council. Instead Adams argued that pan-nationalism could politically isolate the British and the unionists, and increasingly he urged that the IRA’s military strategy be tailored accordingly.

So it was that the Army Council ordered its units to concentrate their energies on attacking British military personnel, to seek out and kill British soldiers in preference to locally recruited and mostly Protestant RUC and UDR members, a change in strategy strongly supported by Adams and his allies.

“Adams’s argument was that we needed to build a nationalist consensus and had to tailor the armed struggle accordingly,” recalled an IRA source familiar with the debate. “We needed fewer car bombs, more attacks on the Brits, more use of mortars against bases, and so on, but the IRA couldn’t deliver.”[13] It was this thinking that had shaped the European and English departments’ campaigns, and the same would be attempted in Northern Ireland.

Once again there were public clues to the private shift which became evident during the spring and summer of 1988. In June Adams told the Observer’s veteran Ireland correspondent, Mary Holland, that it was “vastly preferable” for the IRA to target British soldiers rather than the RUC and UDR. “Callous as it may seem,” he told her, “when British soldiers die it removes the worst of the agony from Ireland…. It also diffuses the sectarian aspects of the conflict because Loyalists don’t see the killing of British soldiers by the IRA as an attack on their community.”[14] A month later Adams overtly linked the military tactic with Sinn Fein’s quest for pan-nationalist unity. In an interview with the author he said,

It is a fact which the Dublin establishment might not like to admit, but there is… a broad acceptance in the 26 Counties which straddles elements of all the political parties that people don’t have any problems about operations against British Crown forces and particularly against what you would call British Army units which are not domestically recruited—no problem whatsoever. If you start off from the basic position that this is morally the right way to do it and add all these other considerations then not only is it the right thing to do but it’s also the clever thing to do—to pursue the armed struggle in such a way that it helps either to broaden the base or doesn’t obstruct the broader aims of the Movement.”[15]

Within the IRA the tactic was justified on the basis that the British would move to leave Ireland if enough soldiers were sent home in coffins. The appalled British reaction to the deaths of two British corporals in March 1988, in the wake of the Gibraltar funeral, was cited as an example. Mistaken for loyalist gunmen about to ambush the funeral of an IRA member, the two men were dragged from their car, beaten by a mob, stripped, and then shot dead by the IRA. Much of the incident was filmed by television crews and the scenes horrified British viewers. Mrs. Thatcher herself turned up at the airport when their bodies were returned to England, testimony to the traumatic impact of the deaths.

The Army Council’s endorsement of the Adams analysis was a highly significant move whose underlying message was not detected at the time. Leaving aside the IRA’s capability or lack of it, the notion that Thatcher’s government or any British government could be influenced by military deaths was a contentious one. The British army was a volunteer, professional outfit, and the precedent from other colonial-style conflicts, from Cyprus to Vietnam, showed that it was only when conscript armies suffer casualties that domestic pressure to withdraw could become a potent factor, as it was in the United States during the Vietnam War and in Britain during the Cyprus emergency. The decision to scale down attacks on the RUC and UDR, in practice unenforced and unenforceable, disguised a deeper political shift. For years the IRA had targeted local security forces, the UDR in particular, knowing that such attacks, in the words of one Tyrone republican, “stop the unionists doing a deal with the SDLP.”[16] In other words, the IRA’s attacks on Protestant security force members kept Northern Ireland unstable. Even though the new military tactic was in this regard largely observed only in rhetorical terms, it indicated a subtle change in attitude toward a Stormont-based political settlement on the part of Republican leaders.

The security statistics tell the tale. The number of British soldiers killed by the IRA between 1988 and 1990 rose to fifty-six, as many as had been killed in the six previous years. The IRA’s efforts in Europe, Britain, and in Northern Ireland, where in 1988 seventeen soldiers were killed between mid-June and mid-August, meant that the killing rate against the British military doubled. But the IRA could not sustain the campaign. In 1988 military fatalities dropped by over a half, to twenty-two soldiers. The next year there was a slight rise to twenty-four, but then in 1990 the IRA accounted for only ten soldiers. In 1991 British fatalities fell to five, to more or less the level they had been at before the revised offensive was launched. The offensive had proved to be unsustainable and was now over.

The failure of the campaign demonstrated that the British were more or less on top of the IRA. Thatcher had responded to the offensive with a public package of measures, including tougher prison terms, a media ban on Sinn Fein, and a restriction on the right of silence normally enjoyed by suspects in police custody, but a secret package, a vast increase in the Irish budget of MI5 and other intelligence agencies, caused more damage. “Money was being pushed through letter boxes,” recalled one activist.[17] Much of the extra finance was directed to recruiting spies in the IRA, but GHQ intelligence noted a sharp rise in efforts to recruit informers in Sinn Fein, so-called “agents of influence,” whose job would be to steer the party in certain political directions. There were even attempts to recruit Adams’s senior advisers, on one celebrated occasion while the target was on a trip to Barcelona.

The British were ahead of the IRA in technical terms as well. New flak jackets worn by patrolling soldiers could not be penetrated by any bullet in the IRA’s armory. Only one weapon could fire a powerful enough round, and that was the U.S.-manufactured Barrett Light 50, an anti-helicopter sniping rifle that fired a huge half-inch armor-piercing slug. Eventually the South Armagh Brigade managed to acquire three of the weapons from its own American contacts, and sent one up to Belfast, where it was promptly lost to the security forces by the Belfast Brigade. But by then it hardly mattered, as the IRA was well on the way to the 1994 cease-fire. The IRA had also lost the battle over radio- and remote-controlled bombs. The British had learned how to block the IRA’s radio signals, even though the IRA had learned to modulate its signals. Bombs triggered by landlines could be detected thanks to infrared sensors fitted to British helicopters that were able to detect heat given off by the detonating cables, while culvert bombs, which had once made rural areas a terror for military patrols, had also been mastered by the British. Stuffed into the drains that tunneled underneath rural roads all over Northern Ireland, these land mines had killed scores of soldiers in the 1970s and made South Armagh too dangerous for the security forces to travel in by road. But by the late 1980s every single culvert in Northern Ireland had been blocked with wire mesh, making it impossible for IRA engineers to booby-trap them.

One way or another IRA strategy was in a state of flux. The demands of pan-nationalism had radically changed IRA policy toward loyalist violence, with enervating results on the republican base in Tyrone. Now the objective of building an alliance between Sinn Fein and Fianna Fail had created a demand for military goals that the IRA had been unable to deliver. The IRA was fast running out of military options. The more Machiavellian members of the organization came to a bleak conclusion. “It was a way of running down the war without saying so,” deduced one former IRA commander.[18]

IF THE NEEDS of pan-nationalism were pushing the IRA in the direction of politics, then Sinn Fein’s electoral reverses were pushing it in the same direction from another. With its constitutional opponent, the SDLP, revitalized by the 1985 Anglo-Irish agreement, Sinn Fein’s electoral march upward and onward had been halted. By 1988–89 the party’s share of the vote had settled at around the 11 percent mark, almost what it had been in 1982 when the electoral strategy was launched, while the SDLP enjoyed twice that level of support. At the level of council elections the SDLP made gains at Sinn Fein’s expense, and then in 1992 disaster struck when the SDLP showed it could repeat that sort of success even against the best that Sinn Fein could field. The jewel in the Sinn Fein electoral crown was snatched away when Gerry Adams lost his West Belfast seat to the SDLP’s Joe Hendron, an affable but politically lightweight local family doctor.

In the Republic it was an even sorrier story. There were no signs at all of the power-broking TDs promised by Martin McGuinness at the 1986 IRA Convention, while Gerry Adams’s assertion that the second election after the Ard Fheis would be the real test of the benefts of dropping abstentionism showed that Sinn Fein’s electoral appeal to Southern voters had collapsed. In the 1987 poll the party won 1.7 percent of the poll and no seats; two years later the Sinn Fein vote fell to a miserable 1.2 percent. In the June 1991 council elections Sinn Fein managed to win only 6 out of the 883 seats at stake and a derisory 0.7 percent share of the vote.

Virtually all the commentators agreed on the cause—a series of botched and bungled IRA operations that began in the autumn of 1987 and continued through to 1992. Each disastrous IRA operation highlighted the glaring contradictions between the IRA’s Armalite-based approach and Sinn Fein’s espousal of electoral, or ballot box, politics, as it was called. With every botched and bungled operation, with every uninvolved civilian killed, public anger grew and electoral support shrank. Gerry Adams’s press aide, Richard McAuley, put it most succinctly in 1992. Once a supporter of the anti-Adams camp led by Billy McKee and David Morley in Long Kesh in the mid-1970s, McAuley had long since joined the Adams camp and later became the Sinn Fein leader’s trusted press attaché. In September 1992 he wrote, “We’re not going to realise our full potential as long as the war is going in the North and as long as Sinn Fein is presented the way it is with regard to armed struggle and violence.”[19]

ON SUNDAY, November 8, 1987, just days after the discovery of the Eksund, local Protestants were mingling with members of the RUC and soldiers of the Ulster Defence Regiment at the cenotaph in Enniskillen, the county town of Fermanagh situated in one of the most picturesque spots in Ireland, where the two Erne lakes, the upper and lower loughs, meet. It was Remembrance Day, the annual occasion when the dead of two world wars and the conflict in Ireland were commemorated in a solemn ceremony that culminated in the laying of poppy wreaths at the base of the ornate war memorial. For Protestants it was a sacred day. Suddenly, and without warning, a bomb, hidden in a building overlooking the scene, exploded, sending tons of rubble and brick onto the heads of the crowd. When the dust settled, eleven people lay dead or dying, six of them woman, many of them elderly, and all of them Protestants. It was one of the worst IRA atrocities of the Troubles.

Condemnation was widespread, but it was particularly intense in the Republic. The reason for that was the impact made by Gordon Wilson, the father of the youngest victim, his daughter Marie, a local nurse. Father and daughter had been caught up in the blast, buried together under the rubble. He had survived, but Marie had been crushed to death His emotional, eloquent, but forgiving account of the ordeal, given in a Border brogue, caught the Southern imagination. In many ways Wilson was the South’s idealized Protestant—liberal, open-minded, and compassionate. He even read the Dublin-based daily, the Irish Times. The fact that the IRA had killed his daughter and had violated a commemoration of Protestant war dead deeply offended many Southerners. Gordon Wilson became an iconic figure in the Republic and was later appointed to the Irish Senate.

The political damage to the IRA was almost beyond calculation, and to compound the disaster the organization had been caught lying in its explanation for the slaughter, falsely claiming that the British had triggered the bomb with a radio sweep for bomb signals. Compelling evidence provided by the British army showed that the device had been set off by a mechanical timer that could have been set only by the hand of an IRA bomber.

It was the start of a catalog of disasters for the IRA. In March 1988 a twenty-one-year-old Protestant woman, Gillian Johnston, died in a machine-gun attack on her car near the Fermanagh village of Beleek that seriously wounded her boyfriend. Again the IRA lied in its account, claiming that the attack had been meant for her brother, a soldier in the UDR. In June the Fermanagh IRA set off a bomb in a school bus near Lisnaskea. The device was meant to kill the part-time UDR driver, but instead it injured children. In July a sixty-year-old Catholic mother and a twenty-four-year-old man, both West Belfast constituents of Adams, were killed in a booby trap bomb placed in the public swimming baths on the Falls Road. A soldier was also killed. Later the same month a land mine meant for a senior Catholic judge as he crossed the Border near Newry instead killed a family driving a similar car. Robert Hanna, a heating contractor, his wife, Maureen, and their seven-year-old son, David, were blown to pieces.

In the nine months since November 1987 the IRA had killed eighteen civilians, not an unusually high level for an organization that had, on the average, claimed the lives of twenty civilians during each year of the Troubles. As in every war, it seemed that civilian casualties were an unavoidable consequence of the Troubles. Two factors, however, made civilian deaths a new and urgent issue for the IRA in 1987 and 1988. The first was the fact that the British had learned from earlier mistakes and no longer killed as many civilians as they used to. In the 1970s and even the 1980s, IRA and British atrocities had more or less canceled each other out. But by this stage only the IRA was killing noncombatants, and there was a public relations and political price to pay for that. The political price was exacted at the polling stations, where bad IRA operations, as they were called internally, lost Sinn Fein support in the voting booths. In fact the damage, the seepage of support for the IRA in effect, could be calculated down to the last vote, and inevitably it had an impact on the organization’s politics.

Only a year before the Enniskillen deaths, figures like Adams and McGuinness had promised their grassroots that dropping abstentionism would not weaken or dilute the armed struggle. In one celebrated interview Adams had gone as far as saying that if Sinn Fein ever repudiated the armed struggle, the party would have to look for a new leader.[20] Two years later, as he sought the go-ahead for talks with the SDLP and Haughey’s representatives, he had assured the Army Council that military policy would remain supreme. But now, as the armed struggle spawned electoral difficulties, Adams and, to a lesser extent, McGuinness began to criticize the IRA’s conduct of the war, criticism that implied the need to curtail and restrict operations.

Adams highlighted the negative consequences of Enniskillen days after the bombing, and he was to return to this theme repeatedly in the next two years. “What is clear is that our efforts to broaden our base have most certainly been upset in all the areas we have selected for expansion,” he said. “This is particularly true for the South and internationally. Our plans for expansion will have been dealt a body blow.”[21] After the Lisnaskea school bus bomb, Adams said he was “unable to condone” the IRA action.[22] Following the Falls Road swimming baths attack he called on the IRA to “get its house in order,”[23] while the deaths of the Hanna family brought stronger words. He was “very shocked,” he told the BBC, and “considerably annoyed.”[24] The language of his condemnations grew sharper with each atrocity.

Then, at the Sinn Fein conference in January 1989, Adams took the bull by the horns. “I want to speak directly to the active service volunteers of Oglaigh na hEireann,” he told delegates. “You have a massive responsibility. At times the fate of this struggle is in your hands. You have to be careful and careful again…. The morale of your comrades in jail, your own morale… in the field can be raised or dashed by your actions. You can advance or retard this struggle.”[25] Adams’s criticism intensified as one botched IRA operation followed another until finally, in June 1990, it reached a sort of watershed. At the launch in West Belfast of a book he had written about his life as an IRA prisoner in Cage 11, his patience ran out. That morning the IRA had placed a booby trap bomb under the car of a retired policeman, sixty-five-year-old James Sefton, in North Belfast. The bomb, fixed by a magnet to the underside of the vehicle and known in the IRA as “up and under,” killed Sefton and his sixty-six-year-old wife, Eileen. Asked what he thought of the operation, a visibly angry Adams replied that he no longer believed that Sinn Fein had to explain, justify, or defend every IRA operation. “You can best describe Sinn Fein’s position as one of critical support for the IRA,” he said.[26] The significance of the remark was that only seven years earlier Sinn Fein’s election candidates had been obliged to give unambivalent support to the IRA, lest anyone thought that fighting elections might inhibit the IRA’s armed struggle. Now Adams didn’t care what the IRA thought. Times had changed.

IT WAS, of course, next to impossible for the IRA to conduct its violence without risking civilian life. Given the conditions under which it fought, especially in urban areas, and the obvious limits on its ability to rehearse operations, civilian casualties were inevitable. Unable to wage the “clean” war that the Sinn Fein leadership demanded and incapable of delivering the quantity and quality of “military operations” necessary to inflict significant casualties on the British army, the IRA saw its war gradually, slowly run down. The preconditions for a public move into the real peace process agenda had arrived.

Adams turned to an institution that he had utilized before when he had wanted to influence the IRA’s direction. In the past he had used it to push the Army Council in a radical military and political direction, but now he would use the Revolutionary Council in an effort to steer the IRA away from violence and toward the still-secret peace strategy. In its first manifestation the Revolutionary Council had been an exclusively IRA body, but when Adams again turned to it, he ensured that Sinn Fein was invited to send members to its meetings. The Army Council was, of course, represented at its gatherings, as were the heads of GHQ departments, but this time the Revolutionary Council would also include key members of the Sinn Fein Ard Comhairle in its deliberations, figures who, like Adams, were unhappy at the conduct of the IRA’s violence. Composed of some twenty-five to thirty people, the cream of the IRA and Sinn Fein, and chaired by Adams himself, the Revolutionary Council would meet two or three times a year, and although its decisions and opinions were not binding on the Army Council, it was in practice difficult if not impossible for the IRA’s leaders to ignore them. Although billed as a way of improving communication and understanding between the IRA and Sinn Fein, the Revolutionary Council would in this form be used by the Adams camp as a tool to rein in and curb the IRA.

One IRA source familiar with its deliberations explained how this worked.

Its function was supposed to be to explore the best way forward. Sinn Fein people raised bad operations, and the IRA people raised criticisms of Sinn Fein statements, but really it was aimed at pointing out to the IRA the error of its ways. It began restricting operations—attacks on school buses were banned, an attempt made to stop killing off-duty UDR men was made but failed, booby traps were restricted and then phased out because they were too indiscriminate, the bombing of factories stopped, and so on. Sinn Fein people said we were putting our own people out of work. They came out against the hoaxes that we were using to cause economic damage. At one stage they were regular Friday events in Belfast, but first they got them banned in the mornings because they were hitting their people going to work and then they were banned in the evening because they affected their people going home. Eventually they were phased out altogether. The Adams faction won the day.[27]

Although the Revolutionary Council was moving the IRA in the direction Adams wished, it was often a hard struggle, and on at least one occasion he was forced to play his ace, in the knowledge that his value to the movement was inestimable. On that occasion, after two Australian tourists mistaken for off-duty British soldiers had been killed in Holland in May 1990, he threatened to quit the IRA and then sought Army Council permission to criticize and condemn IRA operations. When he was refused, he then came up with a second idea. This was to formally and publicly separate Sinn Fein from the IRA, although in practice he and other prominent Sinn Fein figures in the IRA leadership, like McGuinness, would continue to hold their seats on the Army Council. “It lost,” recalled one source, “when a consensus emerged that if Adams went down that road, he and others could no longer hold leadership positions in the Army. McGuinness didn’t back it either. It would have been a sham division but would have allowed Sinn Fein to criticize the IRA and distance themselves from it. But it was rejected out of hand so strongly that it was never heard of again.”[28]

Adams was frustrated but nevertheless later managed to put some formal distance between Sinn Fein and the IRA. Sinn Fein let it be known in January 1991 that the party would no longer act as “proxy spokespersons” for the IRA. A Sinn Fein source told the Irish Times, “The IRA can speak for itself.”[29] Adams’s biographers noted that from this time on it became increasingly difficult to meet the IRA; such meetings were no longer held in Sinn Fein offices, and Sinn Fein stopped faxing IRA statements to the media.[30] The Sinn Fein’s publicity machine had been separated from the IRA’s. The author had a bizarre personal experience of what this meant. On one occasion an appointment to meet an authorized IRA spokesperson turned into a scene from a B spy movie, complete with instructions on which newspaper to read at which café table. The instructions were too complicated, and the meeting never happened. Prior to this a simple phone call would have ensured a rendezvous.

In 1991 Martin McGuinness joined in the criticism echoing Adams’s impatience with IRA mistakes and distancing himself from the organization publicly, even though he was the organization’s military chieftain in Northern Ireland.

Nowhere in the Sinn Fein constitution does it state that if you become a member or supporter of Sinn Fein do you have to support armed struggle. It doesn’t say anywhere whatsoever that you have to support the IRA and I think personally from within the republican movement that that is a position which we should be moving towards in the future. That we have to say to people that you can support Sinn Fein and you can support the republican analysis of what’s wrong in the six counties and you don’t necessarily have to support every single IRA operation which takes place.[31]

Skillfully employing the Revolutionary Council–Army Council mechanism, the Adams camp managed to control and gradually restrict IRA operations, but still the PR disasters—“fuck-ups” as the Provo base called them—continued. The worst of these came in January 1992, when eight Protestant workmen were killed in a 500-pound IRA land mine explosion at Teebane Crossroads on the main road between Cookstown and Omagh in County Tyrone; the blast destroyed their van and tossed their bodies into nearby fields. They had been working for a building company contracted to carry out repairs at the British army base at Lisanelly, Omagh. Opposition to the Provisionals sharpened in the Republic afterward and took tangible form when Dublin City Council banned Sinn Fein from using the Mansion House for its annual Ard Fheis. The council cited Sinn Fein’s support for the IRA as the reason.

Despite all the warnings from figures like Gerry Adams, it seemed that the IRA was determined to carry out operations that embarrassed Sinn Fein, demoralized republican supporters and activists, and, arguably, served to undermine the armed struggle it self. To the outside world it sometimes looked as if all this was deliberate, that the IRA was, at best, uncontrollable and, at worst, at odds with Sinn Fein leaders like Adams over the movement’s political direction. There was a major difficulty with this view and that was that the IRA leadership did not exist in an insulated, self-contained compartment. Adams was a long-standing member of the Army Council, and the Council, as the supreme IRA body, decided Army policy. Metaphorically, the Army Council’s offices were just across the corridor from the offices of the people who ran the war on a day-to-day basis in GHQ, the English department, the European department, and Northern Command and who took their orders ultimately from the Army Council. The heads of these departments were sometimes also members of the Army Council and were often in close, intimate contact with each other. In the latter years of the IRA’s campaign, for example, three key figures—the organization’s director of operations, its Northern commander, and his deputy, the Northern Command adjutant—all had seats on the Army Council. If all these people had been privy to the peace process strategy—as should have been the case—and appreciated the need for a refined military strategy, Gerry Adams would not have had to lecture them so often about the need to be careful in the way that they fought the war. There could be only two explanations: either the Army Council was deeply divided about the strategy, or its members were largely ignorant of it.

IN THE MID- and late 1980s the Army Council had authorized Northern Command to vet brigade operations in Northern Ireland, and as the years went by that control tightened, much to the irritation of the IRA rank and file, who much preferred to operate on a loose leash. The reason for tightening control was simple. It meant that the leadership could more easily ensure that the IRA’s military strategy dovetailed with its political approach, a consideration that assumed enormous importance when Sinn Fein began contesting elections. Despite the requirement for a precisely directed military strategy, the record shows, according to well-informed IRA sources,[32] that Northern Command approved and sometimes initiated the very operations that so often imperiled Sinn Fein’s political strategy, operations that frequently claimed civilian lives. They were not the haphazard, ill-thought-out enterprises that the Adams-McGuinness criticism made them out to be. In the case of McGuinness, he sometimes denounced operations that either he, as Northern commander, or his colleagues on Northern Command staff had approved or endorsed. In the case of Adams, he was condemning operations that had been approved in many instances by his closest political ally. Even though much of the criticism by Adams and McGuinness concerned the manner in which the operations had been carried out as much as the concept behind them, it was an extraordinary, not to say bizarre, inconsistency.

According to these IRA sources, Northern Command had, for example, given the go-ahead for the Enniskillen bombing. Attempts by the IRA at the time to blame the disaster on local activists do not square with the fact that there were similar operations elsewhere that day, evidence of the sort of coordination that only Northern Command could organize. A bomb was found at Tullyhommon War Memorial on the Fermanagh-Donegal Border, set to go off at 11:00 A.M., when the Poppy Day service was set to start, while in Belfast a carload of explosives was defused at the headquarters of the Royal British Legion, which organized the annual war tributes.[33] Northern Command had also vetted the Lisnaskea school bus bomb, the bomb at the Falls Road swimming baths, and the Border land mine that had killed the Hanna family, while the deaths of James and Eileen Sefton, which had so exercised Gerry Adams, came as a result of a plan vetted and approved by Northern Command.

IT WAS AGAINST this background, and with the full approval of the entire leadership, that the IRA organized one of the most cold-blooded operations in its history, one that would be imitated years later on a much larger scale in lower Manhattan and on the streets of Jerusalem by Islamic extremists less concerned about the effect of their actions on public opinion. The use of the human bomb did not begin when Al Qaeda hijackers pointed passenger jets at the towers of the World Trade Center or when Hamas bombers blew themselves and scores of Israeli partygoers to pieces. It began in Derry a decade before.

The first use of a human bomb by the IRA was one of those examples of Northern Command initiating an operation rather than approving one suggested by activists lower down the chain of command. In the autumn of 1990, the Command staff won the approval of the Army Council to mount the first of what was to be a series of human bomb attacks, although there was a crucial difference between these and the events that devastated the United States on September 11, 2001. Unlike the Islamic extremists of Al Qaeda, IRA men would not die in their version of the human bomb. The IRA’s device instead would be delivered by an uninvolved outsider, usually an innocent civilian, forced to ferry the explosive while his family was held hostage. As an operation calculated to undermine the IRA’s armed struggle, alienate even its most loyal supporters, and damage Sinn Fein politically, it had no equal; no other single act of violence perpetrated by the IRA during this phase of its campaign could match it.

Armed and masked IRA men took the family of Patsy Gillespie hostage in the early hours of October 24, 1990, and forced him to drive a car loaded with 1,000 pounds of explosives to the British army checkpoint at Coshquin on the Derry-Donegal Border. Gillespie was chosen to deliver the device because he worked in the canteen of a local British army base, and in the minds of the IRA organizers that made him a legitimate target. He lived in the strongly nationalist Shantallow area of Derry and had refused to relocate even though four years earlier he had been forced to drive another IRA bomb into the city, although on that occasion his life had been spared. As soon as he arrived at the checkpoint, the bomb was detonated by remote control, tearing Gillespie and five British soldiers to pieces. A similar attack later the same day killed a British soldier near Newry, while another human bomb near Omagh was intercepted and defused. Attacks like these continued for a month, culminating in a foiled attempt to destroy the checkpoint at Roslea, County Fermanagh, with a 3,500-pound trailer bomb. In that incident members of a hostage family were tied to a tree while a relative drove the bomb to its target.

Predictably, the human bomb tactic was a public relations disaster. What made the use of the human bomb so difficult to comprehend was that it had been proposed and supported by an IRA leadership that supposedly was in the midst of implementing a strategy aimed at winning political and electoral support on both sides of the Border. That similar attacks were repeated for weeks after Coshquin made the episode even more extraordinary. Adams’s lieutenants told journalists they were in despair at what the IRA had done, conveying the impression not only that the Sinn Fein leader had had nothing to do with it but that he was actively opposed to the tactic. One key adviser told the author at the time, “It’s gone down very badly with the base from the Ma’ [mother] on down. It might be different if it had been a one-off, but because it was done again it has had such a bad effect.”[34] But the criticism and public outrage were entirely predictable and raised obvious questions. The IRA leadership, including Adams, was capable of seeing the negative consequences that resulted from, for example, placing a bomb on a school bus, yet they had unhesitatingly supported a tactic that involved forcing a father of three to drive a huge bomb to an army base and then, before he had the chance to escape, blowing him to smithereens.

Only one thing can be said with certainty. The human bomb tactic fortified the peace camp within the Provisionals and weakened the militarists. The BBC journalist Peter Taylor, a veteran reporter of the Irish Troubles, put it well. “By actions such as this and the revulsion they provoked within the community,” he wrote, “the IRA inadvertently strengthened the hand of those within the Republican Movement who argued that an alternative to the armed struggle had to be found.”[35]

13. The Derry Experiment

The deaths at Coshquin in October 1990 caused outrage in Derry, and there is little doubt the attack marked a low point in the fortunes of the IRA’s Derry Brigade. The local nationalist weekly paper, the Derry Journal, devoted three pages to the story. The banner headline on the front page called the attack “Bloody Wednesday,” and an editorial said the anger in the city was at a level not seen since Bloody Sunday 1972, when British troops had killed thirteen local men, gunning them down without warning or mercy.[1] While nationalist and unionist politicians vied with each other to condemn the atrocity, and the city’s Catholic bishop, Dr. Edward Daly, called it “a callous, cynical, crude and horrible deed,”[2] behind the scenes the deaths animated a mini–peace process that was to see the IRA in the city gradually and secretly de-escalate its violence in tandem with the British security forces. Although limited in scope and cautious in its application, this mutual de-escalation made Derry a laboratory experiment where the viability of a wider peace process was tested and the IRA enabled to signal to the British authorities a willingness to bring its long war to a controlled and phased end. In the process, both sides began to build the trust that would be needed if the IRA was to declare a much wider cease-fire.

The IRA’s campaign in the rest of Northern Ireland would last for another four years, but in Derry the end came much quicker. Nobody could know it at the time, but the five British soldiers who perished in the Coshquin explosion were the last military personnel to die at the hands of the Provisionals’ Derry Brigade. Although seven more people were killed as a result of republican violence in the city before the IRA called its August 1994 cease-fire, only two, both RUC officers, were members of the security forces. Once ranked as the second or third most active brigade in Northern Ireland, the IRA in Derry soon became one of the quietest in the whole organization.


The mechanism that achieved this was modeled on Cold War superpower diplomacy of the sort that for so long had prevented an outbreak of nuclear war between the United States and the Soviet Union. Ultimately it led to both sides’ reducing and scrapping much of their arsenals of nuclear warheads. Known by the acronym GRIT (Graduated and Reciprocated Initiatives in Tension Reduction), the strategy was devised when President Kennedy and Premier Khrushchev were in office and it was employed by them a total of sixteen times in the early 1960s.[3] Instead of demanding concessions from each other, GRIT involved one party to a conflict making a unilateral and publicly verifiable concession to the other. The prior agreement of the other party was not necessary for this to happen, and the strategy could be employed in the absence of formal talks or contact. The effect nonetheless would be to invite the other party to do the same. Failure to reciprocate meant that the erring party shouldered the blame for lack of progress, while agreement accelerated the process. GRIT was a safe way of pressurizing each party to de-escalate. Ideally the strategy would be progressive, as one academic expert explained: “With each exchange of concessions, trust grows and tension is reduced.”[4]

In the context of Derry’s violent past, the IRA’s decision even to discuss its participation in such an enterprise was extraordinarily significant. Unlike the Provisionals in Belfast, the IRA in Derry was not driven by sectarian considerations or by the need to defend its community. The Derry IRA was instead created and sustained by conflict between the city’s nationalist population and the security forces, first the RUC and then the British army. By agreeing to take part in a process that endeavored to reduce the opportunities for conflict or that softened hostility between its constituency and the British forces, the IRA was in effect offering to cut off the supply of its own lifeblood. It would be difficult to find a more convincing way of signaling a willingness to end the war.

TO GRASP the significance of the events at and after Coshquin, it is first necessary to understand the sanguinary history of Derry itself. Although it is inaccurate to do so, most Northern nationalists place and date the beginning of the Troubles to events in the city on October 5, 1968, when a small civil rights march was batoned and hosed off the streets by a force of RUC men. Their orders from the unionist home affairs minister, Bill Craig, were to stop the crowd marching from Duke Street, on the eastern banks of the Foyle, across Craigavon bridge into the historic, walled center of the city.

The reaction of the police was out of all proportion to the threat posed by the crowd of four hundred or so, but their behavior betrayed the enormous importance unionists attached to retaining their grip on this most nationalist of Northern Ireland’s cities. Built and developed as a commercial venture by the City of London companies in the northwest of Ireland in the early 1600s, the walled city was constructed as a Protestant bastion, a garrison of the Scots-English plantation of Ulster. Its fortified walls made the city a place of Protestant refuge during successive Irish native rebellions. A lengthy siege in 1689 mounted by the Catholic King James II won Londonderry, as Protestants always call the city, a special place in loyalist mythology. The fact that it resisted the siege earned it the title “the Maiden City,” and hard-line Protestants determined that Catholics would never breach its sanctity.

Catholics had lived outside the walls of Derry from the late eighteenth century in a rat-infested collection of hovels called the Bogside. The population nevertheless grew; by the mid-nineteenth century Catholics were in a numerical majority and by 1920 made up a majority of voters, although gerrymandering meant that unionists held on to control of the corporation. The introduction of proportional representation in the same year changed all that, and for the first time in Derry’s history a coalition of nationalists and republicans took control and a Catholic was made mayor. But after partition Derry became part of the Northern Ireland state, and unionists set about reasserting their control. Proportional representation was abolished, electoral boundaries were redrawn so that Catholic votes were devalued, and a property qualification was introduced for elections to local government. According to one estimate, a third of the adult population of the Bogside in 1964 was not allowed to vote in council polls.[5]

Characterized by discrimination and deprivation and with huge male unemployment rates, Derry symbolized nationalist complaints about life in unionist-dominated Northern Ireland. Things did not begin to improve for the city’s nationalist community until after the Second World War, when the benevolent effects of the British welfare state, especially the availability of college-level-education, filtered through the Catholic population. A more assertive and impatient generation emerged and demanded change.

Anger at the poor quality of public housing and the unionist-dominated corporation’s refusal to build new developments, part of a calculated policy to confine Catholic voters to overcrowded electoral wards, provoked the first mild street protests, led mostly by left-wing activists. The same factors were working on the Catholic psyche elsewhere in Northern Ireland, and in August 1968 the first civil rights march, from Coalisland to Dungannon in County Tyrone, was held but was barred by the RUC from reaching the center of Dungannon. A few weeks later the Derry housing agitators invited the organizers, the Northern Ireland Civil Rights Association, to lead a similar protest in October.

Not everyone in Derry was in favor of such street demonstrations. The Catholic Church exercised the single most powerful influence on the minds of Derry nationalists. The city’s bishop at the time, Dr. Neil Farren, one of the most conservative Catholic clerics in Ireland, consistently aligned himself with the forces of law and order, while moderate Catholics expressed alarm both at the radical politics of those behind the civil rights agitation and at its potential for unrest. John Hume, for instance, was ready to lead a motor cavalcade of Derry people to Stormont to protest the unionist government’s decision to locate Northern Ireland’s second university in the Protestant town of Coleraine but declined to put his name to a document notifying the RUC of the route of the first civil rights march in Derry, something that would have meant accepting legal responsibility for the event.[6] Tensions between these conservative elements and Derry’s radicals—and later militant republicans—was to be a defining feature of nationalist politics in the following years, but security force excesses repeatedly drove the moderates into the hard-line camp or neutralized them.

That process began on October 5, 1968. Dozens of marchers were hospitalized by the initial RUC charge, while scores more were forced to run a gauntlet of batons or were drenched by a water cannon operated by policemen whose commander, District Inspector Ross McGimpsie, a local version of Bull Connor, enthusiastically joined the fray. A prominent West Belfast MP, Gerry Fitt, the founder of the Republican Labour Party, was the first to get his skull cracked, while the presence there of a number of British Labour MPs ensured that the events would be impossible to ignore. As it was, a cameraman from the Irish broadcasting service, RTE, captured the police violence on film, and the scenes shocked public opinion in the Republic and in Britain. In the Bogside the events in Duke Street triggered three days of stone throwing, and poorly constructed barricades were erected to keep out the RUC. The October 5 melee was a public relations and political disaster for unionism, but it set Derry’s nationalists on a course of increasingly violent confrontation with the state and its uniformed guardians.

The situation only worsened thereafter. In January 1969 some four score radical students from the Belfast-based People’s Democracy group staged a march from Belfast to Derry in deliberate imitation of the American black civil rights march from Selma to Montgomery. The seventy-three-mile route took the students through staunch loyalist areas of Counties Antrim and Derry, and they were constantly blocked and harassed by mobs, some led by associates of the fundamentalist unionist Ian Paisley amid allegations of police collusion. At Burntollet bridge, on the eastern outskirts of Derry, the march was ambushed by an organized gang of stone throwers, many of them off-duty members of the Protestant militia, the B Specials. The RUC fled, leaving the students to the mercy of the loyalist mob. The remnants of the march were again stoned by loyalists as they made their way to Guildhall Square, on the edge of the Bogside, where the arrival of bloodstained survivors sparked five days of fierce rioting. In the early hours of the first night, drunken RUC men went on the rampage in part of the Bogside, smashing windows, shouting sectarian insults, and beating up any Catholic resident unfortunate enough to cross their path. After that, the barricades, this time more effective structures, sprouted throughout the area. Someone painted a slogan on a gable end in the Bogside that read, “You are now entering Free Derry.” A myth had been born.

Derry’s moderate nationalists had condemned the Burntollet march, as it became known, and eventually talked the barricades down, but events were slipping out of their control. In April there was another outburst of violence, and this time the mostly teenage rioters used gasoline bombs as well as stones to fend off the RUC. Again the police went on a late-night rampage and badly beat a Bogside man, Samuel Devenney. He died three months later, and the conviction grew that RUC batons had hastened his death.

There were more riots on July 12, when Orangemen marched through the city on their way to their annual gatherings, but the real crisis came in August, when the Apprentice Boys of Derry staged their traditional parade through the walled city. Founded in memory of youthful Protestant heroes who had defied attempts by their leaders to surrender the city to the Catholic King James in 1688, the Apprentice Boys parade was always an occasion for sectarian coat-trailing, but in 1969, after a year of accelerating nationalist confrontation with the unionist state, the potential for serious trouble was obvious. Attempts by moderate Derry nationalists to get the march stopped failed in the face of a nervous unionist government’s need to appease its extremists. The result was predictable. The ensuing violence pitched not just Derry but the whole of Northern Ireland into the most serious and violent crisis since partition.

In Derry, skirmishes between the Apprentice Boys and Bogside Catholics on August 12 soon developed into a full-scale battle, when the RUC took the side of the loyalists as they made efforts to invade the Catholic area. This time the Bogsiders repulsed the police charges with volleys of stones thrown from behind barricades and gasoline bombs tossed from the top of a tall block of flats, while the police replied by soaking the area in clouds of acrid CS gas, disabling rioter and innocent resident alike. Roused by a communal fear of what defeat might bring, the resistance offered by the Bogsiders was determined and fierce.

After two days the Battle of the Bogside had been won by Derry’s nationalists, and an exhausted RUC was obliged to withdraw. The victorious Bogsiders celebrated as, on the afternoon of August 14, a company of the Prince of Wales Own Regiment took up positions in William Street at the mouth of the Bogside. Elsewhere in Northern Ireland the Derry riots had transformed the political situation. The Irish taoiseach, Jack Lynch, had gone on Irish television the night before to warn that the Republic could not stand by and see innocent people injured and perhaps worse. Lynch’s speech raised unionist fears and nationalist hopes that the Irish army would invade, but arguably the real effect was to oblige the Wilson government in Britain to send in the troops. Meanwhile pleas from the besieged Bogsiders that nationalists in Belfast and other towns stage protests aimed at drawing off police resources had been answered with tragic and fatal results. A Catholic man was shot dead by B Specials in County Armagh, and five people were shot dead in Belfast as rioting engulfed interface areas of North and West Belfast.

The arrival of the British army in Derry was greeted by most Catholics as a huge political victory over the unionist government in Belfast, and although some voices warned against giving the troops a welcome, these were in the minority. What everyone did agree on, however, was that in less than twelve months the nationalists of Derry had been transformed from quiescence to militancy, thanks mostly to the actions of the RUC.

IF DERRY PROVIDED proof that nothing radicalizes people quicker than the thump of a police or army club, then it would be hard to find a better example of this than the experience of Martin McGuinness. A nineteen-year-old apprentice butcher at the time of these turbulent events, McGuinness was later to occupy virtually all the top positions in the Provisional republican movement: Derry IRA commander, Northern commander, chief of staff, chairman of the Army Council, Sinn Fein vice-president, Mid-Ulster MP, and Sinn Fein minister of education. Back in August 1969, though, he was just one of the hundreds of so-called Young Hooligans who would throw stones and gasoline bombs at the police. McGuinness’s journey to militancy was typical of so many of his contemporaries, and his story is the story of the rise and growth of the IRA in Derry.

Born just at the dawn of the welfare state, McGuinness had a background mirroring that of most families in the Bogside. His mother and father were both devout Catholics and daily communicants. Politics, he remembered in a 1989 interview, were “never discussed” in his family, although, like most Bogsiders, his parents voted for the Derry-based nationalist leader Eddie McAteer with as much devotion as they practiced their religion. Militant republicanism was just not an issue. Back in the late 1960s the number of republicans in Derry could be counted on the fingers of one hand. While Belfast had a large network of republican families, Derry republicanism was dominated by just two veteran figures, Sean Keenan and Neil Gillespie, who could trace their involvement back to the 1940s. In the McGuinness household the IRA was a distant and strange thing. “It was never a subject for discussion,” he later recalled.

We had been through what you term the Border campaign, from ’56 to ’62. I do remember vaguely discussions with my friends about what is the IRA and people were saying the IRA is this and the IRA is that. I had no real interest in it and it meant nothing to me. Our lives revolved around attending Derry City football matches, playing football ourselves, playing Gaelic football and hurling and, when we got old enough, having a good time at the weekend.[7]

Nor were McGuinness and his contemporaries always hostile to the RUC:

The cops were people who came to street corners to chase you away if somebody sent for them because you were playing football on the street… they were never seen as a political thing at all. There was actually a cop, I forget his name, who was fairly involved with the local football club here. The cops were strolling around the Bogside and nobody took a second look at them. The older people had a resentment towards them but it was never really discussed or talked about to us young people. I can never recall my father or my mother or any older people saying “these guys are bad news” or “they’ve done this” or “they’ve done that.”[8]

At the early stage of the civil rights agitation McGuinness shared the moderate views of Catholic leaders like John Hume. Reform, not revolution, was uppermost in his mind:

[I was] very pacifist, absolutely and I agreed with them at the time. I thought it was dead sound because at that time I wouldn’t have been saying let’s fight back, let’s use violence against these people because they’re using violence against us. I never felt that the situation had deteriorated so badly that that could be justified. There was always the hope that somewhere along the lines of government would catch themselves on and grant the Catholic people the demands they were asking for.[9]

The beating and subsequent death of Samuel Devenney began a change in McGuinness’s attitudes. “The innocuous policeman who was involved in our football club had suddenly overnight become a monster because Catholics were demanding civil rights,” he recalled. “So at that stage we regarded our community as being under attack by the RUC, that these people had turned into monsters.”[10] After this he began throwing bricks and stones with the best of them.

McGuinness was neutral about the British army when the Prince of Wales Own Regiment replaced the RUC on August 14, and was ready to be influenced by the way subsequent events unfolded. “[T]here were people saying that the troops eventually would be used against the people [but] I never took sides in that debate at all. I never said this side’s right, we should welcome them. After they arrived I just went home… and took up no position at all on whether the coming of the troops was a good thing or a bad thing.”[11]

THE BRITISH TROOPS and the nationalists of Derry enjoyed a honeymoon period after the Battle of the Bogside, but like all honeymoons it was fated to be a short-lived affair. At the end of September, the Derry Citizens Defence Association, set up in anticipation of the August riots by the presplit Republican Club, dismantled the last remaining barricades around the Bogside, but almost immediately there were sectarian clashes that developed into running battles between Protestant youths from the Fountain area and Bogside Catholics. In the riots that followed, a middle-aged Protestant man, William King, was beaten, suffered a heart attack, and died.[12] Local unionists, already deeply unsettled by the events of August, demanded that the British troops react vigorously to this incident and to the continued defiance represented by Free Derry.

The British army’s response to the death of King was governed in no small way by the background of the senior military commanders now charged with policing the city. Many had recently served in colonial trouble spots, and this shaped their attitude to the Bogsiders, as the most detailed study of the city at this time noted: “Brigadier Peter Leng, for example, the commander of the British troops in Derry, had been a battalion CO in Aden from 1964 to 1966, when the British withdrew. It is hardly surprising that such officers should draw on their colonial experiences, in Aden, Malaya, Kenya, Cyprus or Oman, when dealing with the situation in Derry.”[13] Within days of King’s death British troops erected a “peace-ring”, which comprised checkpoints and military barriers, that cut off the Bogside and the Creggan above it from the city center, treating it like an Arab souk that had to be insulated from the rest of civilized humanity. Under pressure from unionists, the British army was beginning to isolate and identify Derry’s nationalists as the problem. That they did so without taking into consideration the objections of the city’s moderate Catholics served only to further isolate this important sector of nationalist opinion and foster the psychological climate for further confrontation.

It was not long after this that Martin McGuinness had his first brush with the law, when he was arrested and charged with a breach of the peace during a confrontation with the RUC and British troops near the Bogside. Accused of shouting abuse at soldiers, he was bound over to keep the peace for two years.[14] Relations between the soldiers and Catholics in Derry gradually worsened in the following months. After clashes between off-duty troops and Catholic youths, often sparked by rivalry over girls at dances, the army banned city center cinemas from showing movies at nighttime. As an example of taking a sledgehammer to break open a nut, it could hardly be bettered.

The number of clashes increased as 1970 wore on. In February a rally by the loyalist leader Ian Paisley sparked confrontations between Catholics and British troops, who pursued stone-throwers into the Bogside, just as the RUC used to do. For the first time the military resorted to snatch squads equipped with long, heavy batons to arrest alleged rioters, often beating them severely. At Eastertime there were more riots and incursions by troops into the Bogside after a protest at Strand Road RUC station, where a Union Jack had been hoisted to coincide with the Official Rep