This publication is copyright Marie-Therese Fay, Mike Morrissey and Marie Smyth 1999 and is included on the CAIN site by permission of Pluto Press and the authors. You may not edit, adapt, or redistribute changed versions of this for other than your personal use without express written permission. Redistribution for commercial purposes is not permitted.

List of Tables

List of Figures

Acknowledgements

We wish to acknowledge the help of the following people:

Dr John Yamell, Department of Public Health, Queen’s University and the Health Promotion Agency; Dr Debbie Donnelly, NISRA; Wendy Hamilton, General Registry Office; Alison Hamilton, Central Services Agency; Ronnie McMillen, John Park, Social Services Inspectorate; Yvonne Murray, Linenhall Library; Andy White; Conor Barnes; Jonathan Blease; Alan Breen; Survivors of Trauma, Ardoyne; Greencastle Women’s Group; Damien Gorman, An Crann/The Tree; Tony McQuillan, Northern Ireland Housing Executive; Staff in the Central Library, Belfast, Queen’s University Library; David McKittrick; Professor John Darby; Dr Andrew Finlay, Trinity College Dublin; Dr Maggie Martin, Eastern Connecticut State University; Brandon Hamber, Centre for the Study of Violence and Reconciliation, Cape Town; Patrick Ball of the American Association for the Advancement of Science in Washington, DC for providing some of the international data; The Community Information Technology Unit, Belfast; The Centre for Childcare Research at the Queen’s University of Belfast; Dr R. Scullion of the Central Statistics Unit; Arlene Healey from the Young People’s Centre; John Park from the Social Services Inspectorate of the Department of Health and Social Services; Dr Brian Tipping, Research Services Ireland; young people from various communities in North and West Belfast, Derry and elsewhere contributed enormously to our understanding of their position; over sixty adult interviewees participated in our enquiries, and many gave us much by way of understanding of parental roles, and their experience. Dr Roger McGinty at INC ORE provided feedback on a draft of the book. Our colleagues, Gwen Ford, Sarah Oakes and Mark Mulligan at The Cost of the Troubles Study, and fellow directors of The Cost of the Troubles Study, particularly David Clements, Brendan Bradley, John Millar, Hazel McCready, Sam Malcolmson, Linda Roddy, Mary Donaghy and Shelley Prue have provided assistance, guidance and feedback. Tracy Wong of the Urban Institute assisted with the statistical analysis. Others who offered feedback were Paul Morrissey, Mary Enright, Bel McGuinness, Marguerite Egin, and Dr Yallowly from the Department of Public Health at the Queen’s University of Belfast, who also provided useful suggestions about other research. We thank those who funded the research on which this book is based: the Central Community Relations Unit of the Central Secretariat; Making Belfast Work, North and West teams; the Special Support Programme for Peace and Reconciliation through the Northern Ireland Voluntary Trust; the Joseph Rowntree Charitable Trust; a private donation; Barnardo’s Northern Ireland, Save the Children Fund; the Cultural Diversity Group of the Community Relations Council, the Belfast European Partnership Board and the Community Relations Council. Finally, we thank Roger van Zwanenberg and everyone at Pluto Press.

5 Patterns of Violence

Like many other phenomena, political violence in Northern Ireland has been distributed unevenly across space and time. Certain periods saw extreme concentrations of violence relating to particular events like the introduction of internment, Bloody Sunday or, more recently, the annual crisis associated with the determination of the Portadown Orange Lodges to march through the Catholic area of the Garvaghy Road following the church service at which they commemorate the battle of the Somme. Equally, over three decades of violence have been concentrated in particular places according to the different phases of the Troubles and the different types of conflict.

Indeed, the conflict itself has taken several forms:

• The war between Republican paramilitaries and the security forces has not merely exhibited different intensities over the years but has taken different manifestations in particular places. For example, the bombing campaign against 'economic targets' was primarily undertaken in cities and towns as was anti-security force rioting and, in the early period, attempts to establish 'no-go' areas. Such activities were associated a high rate of civilian casualties, particularly as the result of urban car bombs and the 'collateral' damage of the riot. In contrast, the 'rural' war was fought more directly between the protagonists themselves with ambushes carried out by both sides. This was a war whose success depended crucially on the intelligence each side possessed and the capacity for surgical strikes at precise targets. In this case, the profile of victims was remarkably different from that generated by urban confrontation.

• Moreover, different time cycles saw different strategies and tactics as each side developed the skills of insurgency and counterinsurgency. Crucially paramilitary actions depended on supplies of weaponry, the capacity ,to develop organisational structures that were difficult to penetrate and the ability to mobilise popular support. The overall strategic approach of the paramilitaries was to create the conditions under which their own capacity for military action was maximised while those of the security forces were politically inhibited by international pressure and legal constraints. Conversely, the security forces sought to create a terrain of maximum manoeuvrability within these restraints. Hence, Northern Ireland has seen different legal strategies (e.g. Diplock Courts) and different security tactics ('supergrasses' or the endless debate over the existence of 'shoot-to-kill' policies). All these developments provided the framework within which the war was fought, influenced how it was conducted and determined the character and location of its victims.

• A wholly different category of violence was associated with inter-communal conflict, although this has also taken different forms. The territorial segregation of the two main religions in Northern Ireland has steadily increased over the period of the conflict. In certain cases, the degree of physical separation has been substantial—the River Foyle largely divides the Catholic and Protestant populations of Derry Londonderry — in Belfast, east of the River Lagan has only 12 per cent of the population Catholic, whereas west of the river has 55 per cent. In other instances, only a few streets, frequently intersected by 'peace walls', separate the different religions. North Belfast is the primary example, containing 14 out of Belfast's 17 peace walls. This proximity factor has often been associated with extreme inter-communal violence — segregation does not convey safety. Moreover, the necessity to move outside an area for work or leisure can carry high risk. Indeed, demographic change, particularly the increasing proportion of the Northern Ireland population who are Catholics, has had spatial implications — Catholics are now the majority in places where, previously, they were a tiny minority. Some such places have been associated with 'traditional' Loyalist marching routes and thus become contested spaces, flashpoints of confrontation and violence during the 'marching' season. Dunng penods of tension, minority populations, predominantly Catholics, were frequently intimidated from where they lived or worked. Thus, patterns of contiguous territories give rise to particular forms of violence and specific categories of victims.

• Equally, the complex geography of religious segregation has also determined the terrain for sectarian assassination. Republicans legitimate their actions as anti-imperialist' struggle. The legitimate targets of this struggle are those involved in, or associated with, the British state. The definition of legitimate targets has been almost infinitely elastic —construction workers rebuilding police stations, those providing catering for the British Army or those long-time retired from the RUC reserve. The specific history of Northern Ireland has meant that those associated in some way with the state have been overwhelmingly Protestant. Accordingly, Protestants perceive Republican intentions as being directed primarily at them in an attempt to intimidate or evict from Northern Ireland altogether. Conversely, Loyalist paramilitaries had no 'obvious target' for their activities — the IRA, by definition, was a covert organisation. Nevertheless, they employed classic counterinsurgency tactics by terrorising the Catholic population which 'hid' the Republican guerrillas. Thus, particularly in the 1970s, all Catholics were regarded as 'legitimate' targets for Loyalist paramilitaries. The motivation was either revenge for the 'Republican terror campaign' or an effort to pressurise the IRA to desist or, in some cases, it was simply hate-crime. Belfast, and particularly North Belfast, was the cockpit for sectarian assassination. The fact that the two communities lived in such proximity facilitated the search for victims and ensured that escape from the scene could be quickly accomplished.

• Finally, there were internal struggles going on within both communities either as feuds amongst rival paramilitary groups or 'punishment' type activities. The former have tended to emerge out of ideological split within paramilitary organisations such as the early secession of key dissidents from the Official IRA to set up the Provisional IRA or the later conflict between Official Republicans and the Irish Republican Socialist Party. Paramilitary punishments have been viewed either as a necessary evil, a form of policing in areas where the writ of the RUC does not run and where petty criminals are 'turned' as informers or as a systematic strategy to maintain control and intimidate opposition to paramilitary power. A surprisingly high proportion of the deaths in each community has been perpetrated by organisations that claim to defend that community. Unlike the random violence associated with deprivation and urban crisis, the violence in Northern Ireland has tended for the most part to conform to certain rules, given that the protagonists keep a weather eye on the media and public opinion. Attempts are made to demonise people after they have been assassinated or attacked, in order to justify the violence, the warning codes surrounding bomb planting and the relatively few no-warning bombs that have been deployed are examples of how the violence has tended to been used according to certain codes.

WHEN DID THE DEATHS OCCUR?

The distribution of political deaths over time in Northern Ireland is described in Table 5.1.
 

* to the nearest decimal point

Just over half of all deaths occurred in the period 197 1—76. If the frequency of deaths can be regarded as a good indicator of the intensity of the conflict, then these five years stand out over the entire two and a half decades. The pivotal event was the introduction of internment in 1971. Only 32 deaths took place before 9 August. Thus, 154 people died in the remaining five months of the year —equivalent to a yearly figure of 370. The number killed in 1972 was even greater, as resentment against security policies (typified by the Bloody Sunday deaths in Derry) increased. The proroguing of Stormont in 1972 fuelled Loyalist resentment and saw the growth of their paramilitary operations. A further high point was 1974, the year the Power Sharing Executive was dramatically brought down by the Ulster Workers' Council strike. After 1976, despite equally dramatic events such as the renewed Loyalist strike in 1977 and the hunger strike deaths of the early 1980s, the number killed in any one year never exceeded 125 and in only seven years was it greater than 100. Put another way, the average annual number of deaths for the whole period was 120, for 1969—70 it was 22, for 1971—76 it was 306 and for 1977—94 it was 91 — thus, the intensity of killing between 1971 and 1976 was more than twice that for the whole period and more than three times that for all of the subsequent years. This can be seen in Figure 5.1 which takes the average annual number of deaths (120) as 100 and scores each other year relative to that. As illustrated graphically, the first half of the 1970s stands out. Indeed, in 1972, the number of deaths was over three and a half times the yearly average for the entire period.

The data themselves cannot provide a comprehensive explanation for the trend. By cross-tabulating the years in which the deaths occurred by those organisations responsible, it is possible to say something about the roles of the various protagonists over time. Just over half of all those killed by the British Army (163 deaths or 51.6 per cent) died between 1971 and 1973 compared to 15.3 per cent of deaths (7 deaths) for which the RUC was responsible. It could be said that the most offensive security posture by the military occurred over the two and a half year period consisting of the last five months in 1971 and all of 1972—73. Although Republican and Loyalist paramilitaries killed very large numbers of people during these years (502 and 216), they constituted a lower share of their respective totals (25.4 per cent and 23 per cent). Thus, the overall distribution of deaths for which they were responsible is less concentrated in the short period characteristic of the British Army. Indeed, 40 per cent of deaths (452 deaths) perpetrated by Loyalist paramilitaries took place between 1974 and 1976.

There would thus appear to be different cycles amongst the organisations principally responsible for the killings that produced the massive concentration of deaths in the early 1 970s. For the British Army, the most offensive period was 197 1—73, for Loyalist paramilitaries, it was 1974—7 6 and for Republican paramilitaries it was 1971—76 accounting for 47 per cent of all the deaths they have carried out in the entire period.

Explosions and shootings were the predominant causes of death. Almost 91 per cent of victims died from these causes. Deaths caused by explosions were more characteristic of the 1970s with shootings more evenly distributed across the period.

Given that violence has also been associated with regular annual events, for example, Loyalist marches or anniversaries of internment, it is also useful to examine the deaths in terms of a monthly cycle — i.e. the number of deaths occurring in particular months over the entire period. Table 5.2 gives the absolute number of deaths that occurred in each month and the percentage of the total. It also gives a ratio that is based on the fact that if the deaths had been evenly distributed across the months, then each would have contained 8.5 per cent of the total. The ratio column gives a figure for each month with its own score relative to 8.5 (treated as 100) so as to indicate clearly the months in which a disproportionate number of deaths occurred.
 

There is some indication of a cycle corresponding to periods of particular tension. The first four months had average or low ratios and the deaths begin to rise in May, with July and August also having high ratios. Months with equally high scores, however, were October and November, both outside the calendar of tension-associated events. It may have been that the years when deaths were most frequent have affected the general monthly cycle. Nevertheless, on this basis the notion of a systematic monthly cycle is not proved.

In summary, Northern Ireland's armed conflict began as a political struggle in the late 1 960s that rapidly escalated into a violent struggle in the first half of the 1970s and which persisted at a lower intensity for the next 20 years. Interpretations vary about why an essentially political struggle about civil rights and modernisation so quickly moved into a violent phase:

• A traditional Unionist analysis contends that the initial civil rights campaign had at its core a Nationalist (and Republican) agenda designed to destabilise the state. The campaign's demands were 'transitional', designed to provoke conflict and, even when granted, fresh demands were raised to generate more friction. The Republican elements involved had a tradition of violence and were seeking a means of legitimating a new campaign against the Northern Ireland state.

• Nationalists argue that the absence of civil rights and indeed, the political centrality of organisations like the Orange Order within the state, were core elements of Northern Ireland. It had been established as structurally discriminatory and coercive. Thus, the state forces reacted violently to a programme of political protest because the state itself could not accommodate reform, thus catalysing three decades of ensuing conflict.

• It is possible, however, to interpret the situation by incorporating elements of both these approaches. The basis of Partition in Ireland lay in the uneven development of the island's economy. Given the industrialisation of the North East and the fact that its products depended crucially on British and Empire markets, the inhabitants of that area could not accept a 'home rule' (never mind independence) proposal because of the threat it posed to its economic infrastructure — Sinn Féin had openly talked of tariff barriers to protect infant industries in the South Johnson, 1981). Had these been applied to Northern industries, with the probability that Britain would retaliate, the area would have deindustrialised even more rapidly than it did. The division of Ireland was not an 'imperialist conspiracy' (in any case, the industries were owned by local, rather than 'foreign' capital), but rather the result of the political and social implications of uneven economic development within the island. Nevertheless, the process of establishing the Northern Ireland state was sustained by an ideology of division and supremacy — sectarianism (O'Dowd et al., 1980). This outlived the economic conditions that determined Partition. Such economic variables almost guaranteed that the island would be politically divided, but sectarianism determined the form that Northern Ireland would take. While Protestants were faced with a Southern state which was predominantly Catholic in its institutions and culture, and while Catholic Nationalists persistently refused to accept the validity of Northern Ireland, the well-documented levels of social and economic exclusion experienced by Catholics were more than just a defensive reaction on the part of Protestants.¹ Given the history and character of the state, it was impossible to separate out the necessity to modernise, and thus eliminate all forms of discrimination, and challenges to its very existence. Such confusion created potent conditions for a rapid transition to violence. O'Leary and McGarry (1993) suggest that Northern Ireland was:

the territorial line of the retreat of the British state in Ireland, and was testament to the failure to build a British identity which would have enabled the descendants of Catholic natives and Protestant settlers to transcend their differences. It was also proof of an Irish failure: Nationalists lacked the coercive and ideological resources to achieve a popularly supported revolution throughout the whole island between 1916 and 1925. (p. 107)

WHERE DID THE DEATHS OCCUR?

Table 5.3 describes the deaths that have taken place in each district council area in Northern Ireland comparing these figures with district populations in 1991.

The table records deaths in two ways: the first gives the number of violent incidents in which someone died (fatal incidents) occurring within the district; the second records the number of district residents who have been killed (victims). Neither covers the total number who were killed. This is because a number of deaths occurred outside Northern Ireland, a large number of non-Northern Ireland residents were victims and, in some cases, there are difficulties in assigning incomplete addresses to particular districts. The number of deaths in each category has been divided by the 1991 figure for the resident population to produce a comparable death rate per 1000 population and these have been ranked. Clearly, the death rates are exaggerated by the fact that the accumulated number of deaths is being compared with the population for a single year. True yearly death rates for each district would be considerably lower.
 

In terms of absolute number of incidents, Belfast, Newry and Mourne, Derry, Armagh, Dungannon and Craigavon stand out. These constitute four areas worthy of greater attention, though the last three can be aggregated to constitute a 'Mid Ulster' (which is distinct from the electoral constituency of 'mid Ulster'). In both absolute and relative terms, Belfast has seen the greatest intensity of violent deaths. Indeed, the rates per 1000 population have been almost twice as high as the next district, Armagh. In some districts, the number of incidents and the number of resident deaths almost tally. As the last column of the table shows, the two lists were within plus or minus 10 per cent of each other in ten districts. In others, the number of incidents was substantially greater than the number of resident deaths, for example, Newry and Mourne. It may be hypothesised that this reflects the 'rural war' referred to earlier in which a high proportion of casualties were members of the security forces who did not live in the area. Equally, in still other areas, the number of residents who died was greater than the number of incidents, for example, Castlereagh and North Down. While such areas were relatively untouched by violence, their residents were less insulated reflecting either membership of the security forces or perhaps being victims of bombs in city centres.

A similar ranking of postal districts provides the next table (Table 5.4), which points to particular parts of Belfast, Derry, South Armagh and 'Mid Ulster' as locations for a high intensity of violence.
 

With respect to local victims, postal districts BT14, BT12, BT13, BT15 and BT11 stand out. All are in Belfast and, indeed, define the North and West of the city. BT47 and BT48 define Derry. BT34 and BT35 cover Newry and South Armagh while BT60, BT61 and BT71 define 'Mid Ulster'. All these areas illustrate the spatial intensity of violence.

Belfast is the most extreme case. Here the impact of violence is revealed by the ward analysis in Table 5.5.
 

The wards have been assembled under the headings of East, North, South and West of the city to provide an indication of the spatial concentration of deaths within it. The North and West of the city have the highest averages for both incidents and resident deaths. In East Belfast, only two wards had scores in double figures and only this area contained a ward in which there were no violent incidents resulting in death. It can be seen that some wards were virtually unaffected on this measure whereas some, like Falls, had more than the total for East Belfast. The number of incidents in St Anne's was even greater but this was a city centre ward affected by the violence taking place there. Accordingly, the number of resident deaths is considerably lower than the number of incidents.

These figures accord with everyday experience, i.e. there are parts of Northern Ireland which have only been marginally touched by the conflict. Within Belfast, the intensity of violence has been skewed towards the North and West of the city. One of these has the highest proportion of Catholics in its population and the other has the highest concentration of sectarianly segregated wards.

Another matter of debate has been the relationship between violence and deprivation — has deprivation been the breeding ground for violence? The Irish Congress of Trade Unions argued that 50,000 jobs would do more to end the conflict than 50,000 guns (Carlin, 1979). In the early 1 990s an exercise in plotting spatial deprivation was undertaken by the Micro Statistics Centre at the University of Manchester (Robson et al., 1994). Thus it is possible to compare the death rates for particular places with their deprivation scores. Table 5.6 depicts the six district council areas that were ranked highest for fatal incident per 1000 population and gives the corresponding ranks for the rates of resident deaths and its deprivation score.
 

Of the six districts ranked highest on fatal incidents, four were among the six worst deprived — the exceptions being Armagh and Cookstown. Strabane ranked highest on deprivation in the region and was ranked eighth on the rate of fatal incidents. There would appear to be some relationship between deprivation and violence, but it is not a simple one. The correlation coefficients of the two violence scores with the deprivation score across all 26 districts were .76 and .52, respectively. It would thus appear that there was a higher level of association between fatal incidents and deprivation than resident deaths and deprivation. The former is a better measure of the intensity of the violence that affected these areas. Many areas untouched by violence saw their residents die elsewhere. Thus, intensity of violence and deprivation do seem to be positively associated.