These extracts are copyright John Darby (1986) and are included on the CAIN site by permission of the author. You may not edit, adapt, or redistribute changed versions of this for other than your personal use without the express written permission of the author. Redistribution for commercial purposes is not permitted.

From the back cover:

Visitors to Northern Ireland are often surprised by its confusing mixture of day-to-day normality and general violence. When internment was introduced in August 1971, for example, hordes of reporters were diverted from the world's other trouble spots to Belfast. They were driven from the airport through sunny peaceful countryside into a city busy with shoppers. Around the hotels favoured by visiting journalists, there were few obvious signs of disruption or violence. Yet less than a mile away as they soon discovered, people were being killed and injured and more than 2,000 families had been forced by intimidation to evacuate their homes during the month of August.

The peace and the violence were aspects of the same reality. One was as characteristic of Northern Ireland as the other. The co-existence of normality and abnormality in such a small space is one of Northern Ireland's many contradictions and is rooted in the dynamics of conflict and in the relationship between conflict and violence.

The core of this book is three communities in Northern Ireland. The experiences of people living in them are not typical. On the contrary, they have experienced much higher levels of violence, and live closer to the conflict than most people in the province. All three have suffered greatly from intimidation and the population movements which followed it. It was for this reason they were chosen, for the research aims to examine the process of community conflict through its most violent expression, and the ability of people to deal with its aftermath. What actually happens in a community which is experiencing violent disruption? What are the mechanisms and controls which enable a return to some sort of normality?

The emphasis throughout is on interactions and relationships at local level. Discussions of 'the Northern Irish conflict' often concentrate on its political and international dimensions at the expense of its operation at ground level. The intention here is to examine the relationships between local interactions and these broader dimensions. The author argues that long familiarity with community conflict in Northern Ireland has led to the evolution of effective mechanisms to control relationships between the two communities; that these mechanisms are essentially local; and that their efficiency and variety hold the key to explaining why a conflict of such duration has not produced more serious levels of violence. They amount to a major and effective safeguard against the conflict expanding into a genocidal war.

ETHNIC conflicts are distinguished from international wars because the combatants permanently inhabit the same battlefield. It is not possible to terminate hostilities by withdrawal behind national frontiers. Even during tranquil periods their lives are often intermeshed with those of their enemies. As a consequence, inter-community conflict is often characterised by internecine viciousness rather than by the more impassive slaughter of wars.

For this reason community conflicts, unless arrested at an early stage, tend to develop along predictable lines: they expand to involve a greater number of activists disputing a greater number of issues; disagreement grows into antagonism; enemies become more efficiently organised under more implacable leaders; the restraints on decent behaviour are weakened. As Coleman put it, 'the harmful and dangerous elements drive out those which keep the conflict within bounds'.

Against this general pattern, consider the conflict in Northern Ireland. Its origins, depending on one's political perspective, have been traced to various points between the twelfth century and 1969. Since 1969 the conflict has been persistently violent. There has certainly been ample time for the dangerous elements to drive out the restraining ones.

During the early 1 970s, many observers believed that the upsurge of violence could only lead to two outcomes: the belligerents would either be shocked into an internal accommodation, or propelled into genocidal slaughter. Neither occurred. More than a decade later a settlement seemed further away than ever, and the level of violence, though remarkably persistent, had not intensified.

On the contrary, there is evidence that violence between the communities has diminished rather than grown. The casualty rate, having reached a peak of 468 deaths in 1972, had dropped to sixty-four in 1984. The proportion of civilian, as opposed to admitted combatant, deaths had diminished at an even greater rate. Rioting and direct sectarian confrontation, common in 1970, had become rare.

So why has the crisis in Northern Ireland not been resolved, either by a compromise settlement or genocidal carnage?

The answer is simple. There has been no resolution because the violence has not been intolerable. By whatever calculus communities compute their interests, the price of compromise is still thought to be greater than the cost of violence.

This shifts the balance of questioning. As well as asking why the violence is so persistent, we should ask why it has not been more severe. Instead of asking why people have tolerated such sustained disruption of their lives, we need to understand how they have tolerated it.

The core of the book deals with three small communities in Northern Ireland. The experiences of the people who live in them are not typical. On the contrary they have experienced much higher levels of violence, and live closer to the conflict, than most people in the province. All three communities have suffered greatly from intimidation and the evictions which followed it. It was for this reason that they were chosen, for the book aims to examine the process of community conflict through its most violent expression, and the ability of its victims to deal with the aftermath. What actually happens in a community which is experiencing violent disruption? What happens after the worst violence has passed? What are the mechanisms and controls which enable a return to some sort of normality?

The emphasis throughout is on interactions and relationships at local level. Discussions of 'the Northern Irish conflict' often concentrate on its political and international dimensions at the expense of its operation on the ground. The intention here is to examine the relationships between local interactions and these broader dimensions.

It will be argued that Northern Ireland's conflict is remarkable for the limitations on its violence rather than for the violence itself; that long familiarity with inter-community conflict within the north of Ireland has led to the evolution of effective mechanisms to control it; that these mechanisms arise from the mundane and essentially local accommodations reached in their own localities by people whose hostility has been modified by their need to carry on living in the same 'narrow ground'; and that the efficiency and variety of these mechanisms hold the key to explaining why a conflict of such duration has not produced more serious levels of violence. They have amounted, so far, to a major and effective control against the conflict expanding into a genocidal war.

This book has been a long time in the making. The research has its origins in 1972 when the Northern Ireland Community Relations Commission, alarmed by the high level of intimidation and enforced migration in Belfast, commissioned a research investigation into its process and effects. This was carried out by Geoffrey Morris and myself. My first acknowledgment is to him and the people who assisted us.

The subsequent direction of the research, and especially the decision to return after ten years to communities examined in the early 1970s, was influenced by many people. Particular thanks are due to academic colleagues from my own and other universities: to John Hickey who helped to establish a sociological framework for the empirical data; to my colleagues at the Centre for the Study of Conflict, especially Tony Hepburn, Nicholas Dodge, Seamus Dunn, Michael Poole and Ed Cairns, whose comments on the problem of defining intimidation were particularly helpful; to Leo Kuper and John Whyte for interest and guidance.

The opportunity to present research-in-progress papers at the Centre d'Etudes Irlandaises at the Sorbonne, the UCLA Sociology department and the Institute of Irish Studies at Queen's University Belfast, as well as in my own university, was invaluable both for the comments they provoked and for those they did not.

My main thanks are to the people who live and work in the communities, and who invariably gave their time and insights for no other reason than generosity.

John Darby
Centre for the Study of Conflict
University of Ulster
1986

VISITORS to Northern Ireland are often surprised by its confusing mixture of day-to-day normality and general violence. When internment was introduced in August 1971, for example, hordes of reporters were diverted from the world's other trouble spots to Belfast. They were driven from the airport through sunny, peaceful countryside into a city busy with shoppers. Around the hotels favoured by visiting journalists there were few obvious signs of disruption or violence. Yet less than a mile away, as they soon discovered, people were being killed and injured and more than 2,000 families had been forced by intimidation to evacuate their homes during the month of August.

The peace and the violence were aspects of the same reality. One was as characteristic of Northern Ireland as the other. The co-existence of normality and abnormality in such a small space is one of Northern Ireland's many contradictions, and is rooted in the dynamics of conflict and in the relationship between conflict and violence.

Conflict and Violence
There has been dispute since the last century about whether social conflict should be regarded as a disruptive or natural phenomenon. The former view saw conflict as 'a disjunctive process' (Wilson and Kolb 1949) and 'characterised by a suspension of communication between the opposing parties' (Lundberg 1939, 275) - a process which threatened existing structures and the smooth running of society. However, sociologists like Simmel, Coser and others have argued that, not only is it impossible to conceive of a group which lacked inherent conflicts, but these apparently negative factors are essential elements in encouraging group formation and cohesion; they bind members together, define group boundaries and give people control over their own activities. Nevertheless this more positive view of social conflict has some serious limitations:

The recognition of regulated conflicts of this kind as a permanent feature of all social structures is now recognised in sociology. They do not, however, adequately explain socially disruptive conflicts.   (Rex 1968, 39)

Insofar as conflict is a joint transaction, the course of events can be understood only as a succession of reciprocating adjustments that the combatants make to one another. What each side does is a response to the actual or anticipated moves of its opponents; thus, the course of events is built up by social interaction.   (Shibutani and Kwan 1971, 135)

They constitute the chains which carry controversy from beginning to end as long as they remain unbroken, but which also provide the means of softening the conflict if methods can be found to break them. It is important to note that these reciprocal relations, once set in motion by outside forces, become independent of them and continue on their own.   (Coleman 1971, 256)

The question is, under what circumstances do some minority groups initiate a process towards apparently inevitable defeat, and why do some dominant groups, but not others, set out to eliminate a weaker enemy?

Two main factors have been suggested as likely to influence whether a conflict remains peaceful or becomes violent. One is the rigidity of the social structures within each combatant group and the level of integration within the general society. If the conflict takes place within an agreed consensual framework, the parties will be less likely to conduct it in a way which endangers their common bonds; indeed, such societies develop institutions and mechanisms to facilitate the resolution of differences. However, if the conflict reflects a fundamental lack of consensus, its spread will not be modified by a reluctance to offend common interests and values: it is therefore more likely to become violent.

The other factor which helps to determine the level of violence is the degree to which a conflict is carried on for selfish motives or for some common aspiration. In Coser's words: 'Conflicts in which the participants feel that they are merely the representatives of collectivities and groups, fighting not for self but for the ideals of the group they represent, are likely to be more radical and merciless than those that are fought for personal reasons' (1971, 119). The crusading element in the conflict allows the commission of acts in good conscience, thus helping to explain the surprise and shock frequently expressed by combatants at acts they had committed during conflict. Further, if the selfless violence on behalf of the group is rooted in an ideology - as in a holy war or a revolutionary struggle - restraints upon it will be even fewer, and a willingness to offer and accept supreme sacrifices much greater.

To these two factors might be added a third which is less easily defined but more closely related to the cultural and traditional characteristics which are peculiar to each conflict. Conflict may take the form of racial riots in American cities, passive resistance in India, assassinations in Algeria, self-immolation in Vietnam or genocide in Ruanda. Each of these expressions of grievance or aggression emerges partly from the historical antecedents of each conflict, and are as natural a response by participants in one conflict as they are alien in another. In each case, however, there is a tendency for the form of protest to recur in successive outbreaks. When agitation is resumed after a dormant period, it looks back to the previous periods of dissent for its models. The importance of this observation is that some expressions of conflict may inherently be more likely to lead to violence than others. The nature of the Algerian and Lebanese conflicts of the 1950s and 1970s, for example, made civil war a likely outcome, whereas the sort of passive resistance encouraged by Ghandi and Martin Luther King applied strong pressure for political settlement. These peculiarities affect the rate as well as the pattern of conflict. A struggle which has rapidly intensified on previous occasions is more likely to reproduce a similar accelerated process. In effect, the stages through which it progresses, having been 'learned' during earlier fights, become compressed like the bellows of a concertina, and are therefore more difficult to divert. Accelerated progress towards violence is also more likely if the conflicting groups have established their own structures and organisations for the conduct of a violent struggle. These are more easily mobilised or diverted towards a more total form of warfare, and their very existence may contribute to such a development.

As Simmel has pointed out, however, few conflicts, from personal relationships to international wars, do not exhibit some restraints on behaviour:

If a fight simply aims at annihilation, it does approach the marginal case of assassination in which the admixture of unifying elements is almost zero. If, however, there is any consideration, any limit to violence, there already exists a socialising factor, even though only as a qualification of violence.   (see Lawrence 1976, 140)

An appreciation by combatants that their conflict is limited will certainly affect their conduct of it. The realisation that one's enemy is not bent on annihilation may encourage a search for compromise. It is also true that the same appreciation may actually encourage an increase in violence, in the belief that a weak opponent will be reluctant to follow along a similar route. In both cases the direction will be determined by the dynamics of the conflict, dynamics which are peculiar to the society in which it is being conducted.

The Northern Ire land Experience
The academic disputes about the role of conflict in society are echoed in the literature on the province's 'Troubles'. In general the predominant themes emphasised by historians and political scientists are dissention, violence and dysfunction. 'To the Irish all History is Applied History', according to Stewart, 'and the past is simply a convenient quarry which provides ammunition to use against enemies in the present' (Stewart 1977, 16). McCracken claimed that 'there is no floating vote on the constitutional issue' and Mansergh argued further that the resulting stalemate 'subordinates every vital issue, whether of social or economic policy, to the dead hand of sectarian strife' (Mansergh 1936). This general position was expressed in its most trenchant and pessimistic form by Rose:

Many talk about a solution to Ulster's political problem but few are prepared to say what the problem is. The reason is simple. The problem is that there is no solution.   (Rose 1976, 139)

The scene described is fashioned by the interests of the observer and by the academic prism through which it is viewed.

Most of the studies which describe ordinary people going about their activities untouched by the conflict were carried out in rural areas or small towns. Burton's study of Anro, a Catholic community in Belfast, is the only one rooted in an urban setting in the heart of the Troubles. Arguing that the press cannot penetrate how the Catholics living in Anro constructed the meaning of events, he claimed that their social consciousness was formed from three interconnected elements: the importance of community, the consciousness and institutionalisation of sectarian identity, and republicanism in its broad, cultural sense rather than a narrow political one. Together they give the community its 'gemeinschaft', which has expanded to take in the new relationships and priorities created by the war. It is a more uncompromising image of conflict than that observed in the more tranquil rural communities studied by Harris, Leyton, Blacking and Buckley.

Which picture is more accurate? The problem is that they do not describe the same image. Some communities are exclusively Protestant or Catholic, others mixed; some have been seriously scarred by violence, others unaffected; some are urban and others rural. The conditions are greatly varied. The research findings are not at variance at all, if their aim is only to describe a great variety of different settings, but some make larger claims. Burton's assertion that Anro is 'taken as typical of the community structure of the Northern Irish Catholic' (Burton 1978, 3), for example, is not easily sustained. The main message from the rich and varied evidence accumulated through local studies is that there is no typical community structure.

Intimidation: The historical background

The Pound district has for many years been chiefly inhabited by a Roman Catholic population, while Sandy-row district has been chiefly inhabited by a population of Orangemen and Protestants. Until lately, however, there was some intermixture, a few Catholics residing in the Sandy-row district, and a few Protestants in the Pound district. Since the commencement of the late riots, however, the districts have become exclusive, and by regular systematised movements on both sides, the few Catholic inhabitants of the Sandy-row district have been obliged to leave it, and the few Protestant inhabitants of the Pound district have been also obliged to leave that locality.

These removals were often kindly enough effected on both sides: friendly notices to quit were often given; and the extreme penalty for non-compliance - namely, the wrecking of the house - was, in many instances, not resorted to until the lapse of some time after such notice.

(Belfast Riot Inquiry 1857)

The origins of this conflict considerably pre-date the riots of 1857. The conquest of Ulster, and the subsequent colonisation of parts of the province during the early seventeenth century, produced a demographic pattern where the native Irish were concentrated in different areas from those of the colonising Scots and English (see for example Beckett 1952 and 1956 and Lyons 1971). The newcomers differed significantly from the native Irish in religion, language, social customs and economic status.

Nevertheless, despite attempts to maintain distinctions between the two communities, there was considerable mixing from the start; Buchanan has described the cultural overlap between the Irish and the Planters, especially those from Scotland (Buchanan 1982), and the attempts to retain demographic segregation were informally abandoned soon after the Plantation. From the start the differences between the dominant Planters and the subordinate Gaels were occasionally expressed in violent terms.

Until the early nineteenth century the violence was essentially rural and was expressed through informal and illegal associations. In most parts of Ireland it was directed against the state and its institutions. As Townshend has demonstrated, collective rural violence was a bewildering confusion of faction fighting, grievances about land, political protest and 'elements of carnival'. Popular reactions to it were often ambivalent. Many disapproved of its forms, but few did not have some sympathy with its causes:

The better farmers are more anxious to have these parties [the Ribbonmen] put down, yet they would sooner surrender their arms and property than give any information to the authorities.   (Townshend 1983, 18)

With the growth of industrial Belfast from the early nineteenth century, sectarian violence became increasingly an urban phenomenon. Between 1835 and 1935 there were eight periods of serious rioting in Belfast - in 1835, 1857, 1864,1872, 1886, 1898, 1920-22 and 1935 - and many years during which some disturbances have been recorded (see Boyd 1969 and Budge and O'Leary 1973). There were also two serious riots in Londonderry - in 1869 and 1884. Five of these were sufficiently alarming to produce official inquiries, and their findings provide a record of urban tensions. Apart from the reports themselves, the verbal and written evidence which accompanied them gives insights into relationships between the two communities. They also permit longitudinal comparisons with similar reports from the riots of 1969-71, especially the Cameron and Scarman reports, and with data collected during this research. There were disadvantages, however, in that witnesses could not be compelled to attend, cross-examination was strictly controlled and, except in the later reports, witnesses were not sworn. The other main sources of information were contemporary reports by citizens, newspapers and visitors to the city; in some cases, these can be used to verify the evidence of witnesses.

One of the early riots in Belfast was in fact recorded by an English visitor in 1835, twenty-two years before the first official report. John Barrow visited the city on 12 July, Orangeman's day, when disturbances broke out over an arch in Sandy Row, the Riot Act was read and a woman was shot. The novelty of the incident was remarked by Barrow:

Though riots of this kind are not of unusual occurrence in the great towns of Ireland, and happen but too frequently in some of those of Great Britain; yet here, in Belfast, where every one is too much engaged in his own business, and where neither religion nor politics have interfered to disturb the harmony of society, it could not fail to create a great and uneasy sensation.   (Barrow 1835, 35-36)

Despite considerable variation in the intensity and duration of the riots, the occasions which led to them were remarkably similar. There was a sharp increase in the Roman Catholic populations of both Belfast and Londonderry during the early eighteenth century. Barrow described how 'some four or five thousand raw, uneducated Catholic labourers from the south had, within a few years, poured into the city, to supply the demand for labour' (Barrow 1835, 33). In Londonderry too, as the 1869 commissioners remarked, 'trade and commerce attracted population, much of it from the Roman Catholic county of Donegal' (Londonderry Riots Inquiry 1889, 15). In both cities the religious communities were concentrated in particular and adjoining districts. This provided the backcloth to the urban riots.

There is no discernible relationship between economic conditions and the tendency towards rioting. Violence broke out during periods of prosperity and depression. However almost all the major riots emerged from a combination of general political unrest and a particular catalytic incident. The former occasions included the 1886 and 1893 Home Rule bills, political meetings (Londonderry 1869 and 1884), the centenary celebrations for the 1798 rebellion and the introduction of partition (1920-22). The catalysts varied greatly in form: the funeral of an early victim, an argument between two navvies in the shipyard which led to a drowning, the pulling down of an Orange arch etc. Rioting was predominantly a summer activity, and the annual marches and celebrations which marked July and August were often the immediate cause of fighting or led to its intensification. The Londonderry commissioners in 1869, referring to the annual demonstrations by the Protestant Apprentice Boys on 12 August, observed that in recent years 'the character of the demonstrations has certainly undergone a change, and, among the Catholic lower classes at least, they are now regarded with the most hostile feelings' (Londonderry Riot Inquiry 1869, 15). The Apprentice Boys were involved in both of the city's riots. In Belfast the 1857 riot commissioners singled out the Orange marches on 12 July as 'the originating cause of the riots' (Belfast Riot Inquiry 1857, 8); the point was made more graphically in the cross-examination of Francis Keenan, a Catholic worker in the Belfast shipyards, in 1887. Asked whether sectarian fighting was an unusual occurrence, he replied:

Oh, it is an annual occurrence. Every year it is just the same. They stick up everywhere all through the place, 'Away with the Fenians', and I saw posted on the boiler house, 'All Fenians clear out, by order of the RBC', which I take to mean the Rivet Boys' Club.   (Belfast Riot Inquiry 1887, 444)

A major preoccupation of all the reports was the conduct of police and magistrates, and consideration of how the riots might best be controlled - also a main concern for the Cameron report in 1969. One feature was common to all the disturbances: the inability of any form of police force to establish a neutral stance in Belfast's sectarian disputes. In 1857 the main responsibility for policing was in the hands of a local force of 160 men, controlled by Belfast Corporation. As the commissioners pointed out, 'the police force are, with six or seven exceptions, entirely Protestant, and those in any command amongst them are exclusively so; a great many of them are, or have been, Orangemen' (Belfast Riot Inquiry 1857, 4). Consequently they were severely attacked by Catholic rioters. By 1886, however, the Royal Irish Constabulary (RIC) had assumed the main responsibility for policing. Many of the policemen were marked as southerners by their accents and presumed to be Catholics. Protestant grievances about the anti-Protestant bias of the RIC were a main preoccupation of the witnesses in 1887. Consequently if one side regarded the police with hostility, the other side often regarded them as comrades. In the 1886 riots, for example, Protestant witnesses described how Catholic crowds applauded the RIC as they attacked Protestant crowds. In 1857 a Protestant local policeman, Mr Bindon, described the hostility when he arrested a Protestant: 'They thought I was a bad Protestant when I arrested some of their own party' (ibid, 56).

The term 'intimidation' was commonly used by witnesses during the 1864 inquiry, replacing the more sedate 'removal' of earlier accounts. There was little sign of misunderstanding about its function. From the 1857 inquiry, the process which witnesses described was clearly designed to rid alien elements from districts, factories and services:

The separation of these two districts (i.e. the Pound and Sandy Row) into exclusive encampments appears to us to be little more than the preparation for the festivals of July, and the clearing out of the supposed supporters of the opposite classes to prepare the respective districts for the scenes which follow the celebration of these festivals. (ibid, 2)

They said they would not disturb us at that hour of the morning [3 o'clock], but they would give us until six tomorrow evening to leave, for they had orders from the authorities of the town that none of our class should live in that locality. (ibid, 138)

A woman come and told me to leave there that night, for she heard them say that if I was not out of there on Sunday night my corpse would leave on Monday morning. (ibid, 198)

Although the extract from the 1857 report quoted at the start of this chapter referred to 'regular systematised movements on both sides', the evidence itself suggests that they were relatively spontaneous. In the later riots, however, the process of intimidation became more systematic. In 1886, for example, eleven Protestant families were ejected forcibly from Argyll Street (Belfast Riot Inquiry 1887, 688). The eviction of people from workplaces was also common, and the shipyards and docks were the centres of many attacks on Catholic workmen. During the same 1886 riots, public houses became major targets for the first time and at least forty were attacked and destroyed, thirty-eight of them owned by Catholics (ibid, 7 and 9). John Riordan, one of the publicans, considered that the reason for this new concentration was the fact that 320 of Belfast's 400 public houses were owned by Roman Catholics (ibid, 496).

A Catholic priest, Fr John Tohill, presented a list of Catholic businesses wrecked in Protestant districts, and suggested that the attacks were systematic: 'The conviction in the minds of the Catholics was that there was a determination to drive Catholics, because of their religion, out of Protestant districts' (ibid, 508). The subsequent expulsion of 700 Catholics from their work during the 1898 riots (Boyd 1969, 175), of 2,000 from the shipyards in 1912 (Gribben 1982, 49) and the activities of the Ulster Protestant League in 1935, appear to suggest that some of the enforced movements were caused by organised activities as well as by the immediate anger of individuals.

During the course of the riots there were strong pressures for people to conform to the standards and behaviour of their communities. Indeed it is clear from the evidence presented to the inquiries that there were strong sanctions against crossing to the other side. Members of the Orange Order during the 1850s had been expelled for marrying Catholics (Belfast Riot Inquiry 1857, 10). One Captain Venner, a magistrate, when cross-examined about discrimination on appointments to the magistrates' bench, suggested that the low number of Catholic magistrates was partly explained by their unwillingness to join the side of the authorities.

Mr Lynch: How many Catholics are there?
Captain Venner: There is only one gentleman that I know of- one who is a Roman Catholic. I believe he is the only one competent to fill the situation, and, at the same time, willing to accept of it. (ibid, 219)

I live in Newtownards Road; but, my Lord, 'Mike Hale, the informer' is even written up on the walls in the Ballymacarret direction. That is the thanks I got for trying to save the old man that is old enough to be my grandfather, and the police must know it.   (Belfast Riot Inquiry 1887, 385)

A band accompanied the procession; but does not seem to have played any party-tunes. Here, indeed, we may interpose the remark that it is difficult to say what is not a party-tune in Derry; for the most innocent airs, if played by a particular band, assume, at once, for another section of the people, a party character.   (ibid, 17)

Black: The great majority of the small farmers in that district are Protestants and it is principally from their sons that the appointments are to be made, inasmuch as they are the strongest class of men, generally speaking.

Dowse: Do you mean physically?

Black: Yes, they are generally stronger than the lower classes of the Roman Catholics: besides, I think they are better educated.

Dowse: You look at the candidates before appointing them?

Black: In some cases I could tell a man's religion by his face, but not always. (ibid, 123)

There were a number of cases in all the riots between 1835 and 1935 which demonstrated that intimidation did not meet with the approval of all the members of intimidating groups. Instances were recorded of victims being rescued and protected, sometimes with unfortunate consequences for the protector. In 1886, for example, the Protestant neighbour who gave shelter to Patrick Cox when his shop was destroyed was subsequently boycotted and forced to sell his own business (Belfast Riot Inquiry 1886, 495). The clergy were often praised for their moderating influence and in 1912 patrolled the Pound district to prevent the celebrations for the reading of the Home Rule bill expanding into a riot (Gribbon 1982, 49). The 1864 commissioners were sufficiently moved by the evidence of a Presbyterian clergyman, the Rev. Isaac Nelson, to quote from it at length in their report:

My Protestant neighbours remained up, wandering round the houses, playing 'The Protestant Boys' and 'The Boyne Water', and using phraseology which I hope will in future be foreign to the towns of this island. Having taken possession of the highway, they maltreated, in spite of all my remonstrances, every passer-by who would not use certain language. I am speaking of a number of persons with whom I have been to a certain extent acquainted for years, and can state to be most well conducted and quiet persons. I saw that crowd come up to the houses of four poor members of the Latin Church. I did not then know myself exactly their religious denomination. I saw the furniture broken to pieces on the floor, and I saw the houses, as you express it, gutted. . . . The mobs in my neighbourhood not only hunted poor Roman Catholic neighbours out of their houses, but I had to go and beseech them to grant so many hours to these poor people to take their furniture out of the place. I had also to go and get horses and carts to remove the furniture, and I had a great deal to do to repress the violence of the mob.

Whose conflict?
The riot reports of the nineteenth and twentieth centuries chronicle a tradition of urban violence between Catholics and Protestants, but they also present a distorted picture. By definition their subject was exceptional. They were written about periods when public order had broken down and made few references to the other years when disruption was minimal. Further, they were concerned only with Belfast and Londonderry and, in the main, only with relatively small districts within the cities. While this research is also primarily concerned with communities affected by exceptional violence, it is important to provide a perspective by looking first at the literature on Protestant-Catholic relations in Northern Ireland.

The recent acceleration in publications about the Northern Irish conflict underlines the intimacy of the relationship between public interest and academic activity. In 1976 a bibliography of the Northern Irish conflict detailed more than 800 references, of which 580 had been published in the six years since the outbreak of violence in 1969 (Darby 1976). More recent bibliographies have demonstrated that the traffic has not diminished (Eager 1980; O'Dowd, Rolston et al. 1983).

The earlier lack of interest in the subject was not entirely due to the generally unfashionable nature of ethnic studies, nor to the generally mundane and non-confrontational relationships between Catholics and Protestants in Northern Ireland before 1969. It also arose from a gradual shift among historians and social scientists in their definitions of the boundaries and nature of the conflict and the combatants. The creation of the Northern Ireland unit in 1921 and the development of separate institutions north and south of the border made it increasingly difficult for historians to treat developments in the two parts of Ireland in an integrated manner. Social scientists also have become more inclined to treat Northern Ireland's community problems as peculiar. Rosemary Harris' study of Ballybeg in 1952-53 made it clear that the villagers there interacted within a fundamentally Northern Irish context. Barritt and Carter entitled their pioneering social survey The Northern Ireland Problem and presented a study of what was clearly an agreed, if divided, society. This view has been largely endorsed by subsequent writers.

This does not imply that both groups had a similar attitude towards the state. On the contrary, while the Northern Ireland state came to acquire a positive reality for its Protestant majority, for its Catholic minority the reality was a reactive one. Protestant nationalism, deeply suspicious of Britain, had become primarily attached to Northern Ireland; Catholics, on the other hand, continued to regard it with suspicion. However the passage of time had produced among many some measure of reluctant acceptance of the six-county unit, if only because alternative accommodations would present greater problems. Every po1l and survey since the 1960s has shown a substantial level of Catholic support for a settlement within Northern Ireland, with certain conditions attached. In 1968, for example, Richard Rose found that thirty-three per cent of Catholics approved of Northern Ireland's constitutional position, and a further thirty-two per cent did not express an opinion; in another po1l in 1972, forty-one per cent of Catholics said that they would vote against the unification of Ireland, and six per cent that they would abstain. Sixty years of existence have given Northern Ireland a level of de facto recognition, even from its opponents. In 1972 JC Beckett summed it up:

A fifty-years experience has not given Northern Ireland either unity or stability: but it has defined the political, economic and social problems of the area in specifically local terms. Even if Northern Ireland, as a separate political unit, were to be abolished, it would, in a real sense, survive, and carry into the foreseeable future the characteristics that have developed from the settlement imposed in 1920.
(Beckett 1972)

The Home Rule struggle is a struggle between two nations, the Protestants and the Roman Catholics, or as, to avoid the semblance of ministering to religious bigotry, they had better be called, the Unionist and the nationalist.
(Moneypenny 1912)

The consciousness of group differences is most tellingly illustrated by the novels, poetry and autobiographies of Ulster people. McCann, who grew up in the Catholic Bogside district of Derry, remarked: 'We were never taught to hate Protestants. Rather we were taught to accept that it was for the best that we did not know them. We resented them, of course, in a generalised way. We told one another to "just ignore them"' (McCann 1974, 21). Robert Harbinson, growing up in the Protestant Sandy Row district of Belfast, recounted the fear of Papal conspiracy, and describes his Orange Order excursion to Bangor, where 'from the safety of the passing train, we could boldly hurl abuse at the Mickeys' houses and their papish murals' (Harbinson 1960, 130). Linked to this antagonism was a firm conviction, infrequently tested, in the universal ability to recognise the other side. As John Hewitt put it, in a powerful poem comparing Northern Ireland's Protestants with the Romans during the declining days of empire:

You may distinguish,
if you were schooled with us by pigmentation,
by cast of features or by turn of phrase,
or by the clan-names of them which are they,
among the faces moving in the street.
They worship Heaven strangely, having rites
e snigger at, are known as superstitious,
cunning by nature, never to be trusted,
given to dancing and a kind of song
seductive to the ear, a whining sorrow.
Also they breed like flies.... (Hewitt 1950)

So this is the first point to make: That virtually everyone in Northern Ireland feels himself under threat and reacts accordingly. There is no inclination for reason or compromise simply because the most urgent need is to combat a threat which may seem small or non-existent to outsiders, but looms obliteringly over those locked into the situation.
Jackson 1971, 6)

Despite the common hostility and apprehension, it is clear that there are significant differences in mutual group perceptions, although not all researchers have agreed on the nature of the differences: 'Catholics see discord in nationality terms whereas Protestants see it in religious terms', Rose claimed; 'Politics in Northern Ireland involves ideologically unrelated conflicts' (Rose 1971, 216). Heskin supported this view, and went on to suggest that, 'whereas Protestants objected primarily to Catholics as representatives of a religious group, Catholics objected primarily to Protestants as people' (Heskin 1980, 47). Jackson, on the other hand, believed that Protestant stereotypes of Catholics - as feckless, dirty and sexually unrestrained - were essentially racial, and almost indistinguishable from English views of racial immigrants or Israeli views of Arabs. He went on to make the point that general stereotypes of the out-group were not applied in practice to its individual members:

At individual level one never finds that these faults are singled out. No matter how viciously a man may attack the other religion as a group, he will always make an exception of the members he knows personally.
(Jackson 1971,9)

The most common characteristic attributed to Roman Catholic schools by Protestant teachers was the role of the church in the school's affairs; on the other hand, the description of Protestant schools by Catholic teachers frequently included such terms as 'cold', 'rigid' and 'more academic'. These descriptions were used with great consistency.
(Darby et al. 1977, 67)

The distinctions between the two religious groups did not exist merely in their perceptions of each other. There were also real differences in terms of economic prosperity, demographic location, institutional arrangements and social relationships.

Birrell has demonstrated, in a useful synthesis of research on the subject, that Catholics not only perceived themselves to be economically deprived in relation to Protestants, but that their perception was well founded (Birrell 1972): they were more likely to be unemployed, to be paid lower wages and to live in inferior housing. Like many others, he suggested that there was a direct relationship between deprivation and minority discontent (see de Paor 1970; Boserup 1972).

The demographic location of the two groups, in broad terms, also marked a difference between them. Generally, Protestants predominated in the eastern counties of Antrim and Down and in Belfast, while Catholic majorities were more common in the areas closest to the border with the Irish Republic. In some areas, notably the working-class districts of Belfast and Londonderry, polarisation was high. Even before the current violence started, two-thirds of Belfast's families lived in streets in which more than ninety per cent of the households professed the same religion (Poole and Boal 1973, 14).

There were also institutional links which bound together each community and excluded the other. Political allegiance and religious adherence were highly correlated, and various attempts to establish bi-confessional political parties - notably the Northern Ireland Labour Party during the 1950s and 1960s, and the Alliance Party since 1971 - have only confirmed that the political middle group does not attract the support of more than ten-fifteen per cent of the population. As the historian McCracken expressed it, 'there is no floating vote on the constitutional issue' (McCracken 1967). Schools were also highly segregated by religion, especially at primary level. In 1976, more than seventy per cent of schools participating in a survey had no children from the religious minority group; a further twenty-six per cent had religious minorities of less than five per cent. Among teachers the level of polarisation was even higher: only 1.5 per cent of the teachers in the survey taught in schools where the predominant religion was different from their own (Darby et al 1977, 25-28).

Social interaction between the two groups reflected these differences. 'In general the social network of the individual is based overwhelmingly in most fields on ties with his co-religionists', wrote Rosemary Harris; 'All social relationships are pervaded by a consciousness of the religious dichotomy' (Harris 1972, x). The dominance of the central schism has been a major theme in many other studies (Barritt and Carter 1962, Rose 1971, Leyton 1974). More detailed studies of a Protestant community in Belfast (Nelson 1982) and of a Roman Catholic community in the same city (Burton 1980) confirmed the denominational exclusiveness of some working class urban districts, and supplemented the research by Harris and Leyton in rural areas of counties Tyrone and Down.

Within these polarised districts both groups exhibited a high level of self-sufficiency. Boal examined a wide range of social relationships between two highly segregated urban districts, including school attendance, shopping patterns, social visiting and newspaper readership: he concluded that 'the cumulative evidence indicates the presence of two very distinct territories' (Boal 1971, 8). Harris, describing shopping patterns in the small village of Ballybeg, also confirmed that 'the advantages offered by one shop over its rivals had to be very considerable before a Protestant owner could attract Catholic customers, and vice versa' (Harris 1972, 6).

Leisure activities were also highly segregated in Ballybeg, partly because dances, film shows and sports were often organised by the churches. When cross-religion contact occurred, it was mainly in middle-class settings like golf or tennis clubs, or at formal dances at the local hospital - an observation supported by Kirk's study of Lurgan, a town in County Armagh. Kirk found similar middle-class mixing, but a high level of separate and parallel leisure provision among workers: a Protestant Mechanics' Institute and a Catholic Working Men's Club; the Irish National Foresters (Catholic) and the Masons (Protestant); two religiously segregated old people's homes.

Such a high level of segregation between the two groups depended, to some extent on a system of recognition, or 'telling', as Burton called it, by which the group membership of strangers could be assessed. Cairns and others have pointed out that this discrimination is based 'not upon perceptual cues but upon stereotyped cues' (Cairns 1980, 117); consequently, without obvious indicators like skin colour, the cues were more difficult to acquire. According to Cairns, the most frequently used cues were, in order, area, school, name, appearance and speech, while Leyton found that the elaborate code of recognition used in Kildaragh depended most on schools attended and games played (Leyton 1974). At this point, however, the research dries up, leaving a number of unanswered questions: How accurate is the process of telling? What are its functions - to avoid or to facilitate conflict? On what occasions is it called into play? There has been virtually no research on these and allied questions.

Another under-researched theme is the incidence of mixed marriages between Protestants and Catholics - the interface between the two groups. The delicacy of the subject in Northern Ireland has made enquiry difficult. In Rose's survey two per cent of the married respondents had partners from across the religious divide, but twenty-six refused to answer the questions (Rose 1971, 508). Harris claimed that in Ballybeg there was an 'almost universal refusal to recognise kinship across the divide' (Harris 1972, 143), and that mixed-marriage couples often ceased church attendance - a conspicious act in the 1950s when her research was carried out. 'Intermarriage bridges no gaps,' she concluded. Leyton too has referred to 'the total prohibition of marrying Catholics' among Protestants in Aughnaboy (Leyton 1975, 57). However, research by Hickey and others suggests that these conditions are far from universal, and that cross-religious relationships may be more common in other places (Hickey 1984).

Barriers to polarisation
There is a general consensus in the academic literature, therefore, that Northern Ireland is the arena for a fundamental and dysfunctional social conflict, and that it is between Catholics and Protestants. Even such broad generalisations, however, are subject to a number of important qualifications. Their common premise is that a crude picture of Protestants and Catholics in stratified conflict is misleading and inaccurate. Barritt and Carter, for example, having described some of the major divisions, went on:

There is, however, another side to the matter. The two communities in Northern Ireland live side by side, generally at peace. Their people are linked by many ties of personal friendship, and by occasional significant ventures in cooperation.. . On the whole, people manage to adapt themselves very well to ancient and continuing differences, and they use those differences to add to the richness and humour of life.   (Barritt and Carter 1972, 2)

Following an analysis of anthropological research in Northern Ireland, Donnan and McFarlane drew attention to the wide variations in the nature of Catholic-Protestant relations throughout the province: 'As our review shows, there seem to be some contradictory findings even within the body of literature discussed' (Donnan and McFarlane 1983, 134). Some of the contradictions are contradictions only if it is assumed that Northern Ireland can be regarded as a unitary social entity. In some important respects this might be questioned. Protestants are more concentrated in the eastern part of the province along the Lagan valley and counties Antrim and Down, while Catholics predominate in the border areas, although even this gives a false impression. It is rather more accurate to regard Catholics and Protestants as living in a series of denominational enclaves. In some cases, like the working-class districts of Belfast described by Boal, there is a very high level of segregation; elsewhere, especially in rural areas, the enclaves may vary in size from small groups to individual families. Poole's studies of the small towns which are such a common feature throughout Northern Ireland, show convincingly that the highly segregated ghetto is untypical, and that in most parts religion is not a major determinant in housing choice. Total absence of cross-religious contact, therefore, is very rare; contact takes place in every part of the province.

Community studies confirm this fragmented picture. Certainly Harris found that the closeness of kinship ties in Ballybeg, and the high correlation between religion and kin, reduced the incidence of cross-religion contact. However other researchers like Kirk found specific conditions in their communities which encouraged visiting and social contact. Yet another social pattern was described in Burton's study of urban Anro, which was adjacent to almost exclusively Protestant districts, but which housed virtually no Protestants. Leyton's Aughnaboy presented a Protestant mirror image, but in a rural setting. In effect all the major studies of particular communities described systems of relationships between Protestants and Catholics which were apparently quite distinct in their operation.

Because of the longevity of conflict and violence in Northern Ireland some commentators have been persuaded that one of its major features has been its immutability. The most direct recent expressions of this view have come from political scientists. Richard Rose has suggested that the basic lack of consensus about the legitimacy of the state is an almost insurmountable barrier to settlement. Aunger argued that the major splits in Northern Ireland were all congruent with the central divide between Protestants and Catholics, that this made cross-cutting allegiances and compromises difficult, and that this situation has altered little through time. The main challenges to these views have come from historians and social scientists. Historians have pointed out that relationships between Protestants and Catholics have not always been sharply dichotomous, and that allegiances have occasionally shifted in the past. During the last two decades of the eighteenth century, for example, Catholics and Presbyterians found common cause in the United Irishmen, until their attempted rising in 1798 failed. In 1932 Catholics and Protestants joined in a protest hunger march.

Although these instances were rare, there have been a sufficient number of similar illustrations (the 1965 cavalcade from Derry to protest about the location of Northern Ireland's second university, for example) to demonstrate that exceptional mutual feelings of deprivation can overcome sectarian suspicions. Community violence in Northern Ireland appears to have been cyclical rather than cumulative, and adjustments made during violent stages in the cycle have often been re-adjusted during more peaceful periods.

So the relationships between Northern Ireland's two religious groupings, rather than remaining static, have constantly changed. These changes have accelerated during periods of violence, and are more volatile in urban areas. Their effects may be altered again during peaceful periods, and have hardly affected many parts of the province at all.

In general, the importance of understanding the process of change - in determining the timing of social reforms, for example - has been underestimated in the social science literature.

So has the amount of internal division in both Catholic and Protestant communities. The wide variation among Catholics about their constitutional preferences is well established and has already been noted. Among Protestants, antipathy to Irish unification is a strong unifying factor, although opinion is more divided on the question of the political role which Catholics should have within Northern Ireland. The occasional tensions between the different Protestant denominations, it has often been suggested, were infrequent and unimportant. There is evidence that, before the First World War at any rate, 'the Protestant denominations as a whole shared a common mind, and the laity in particular.. were Protestants first and members of their own church after that' (Megahey 1969, 63). His opinion was shared by Emrys Jones:

Any discussion of religious groups in Belfast is dominated by the fact that the population is sharply divided into Protestants and Catholics. Differences between denominations within Protestantism tend to be overlooked or dismissed; comparisons and contrasts are constantly framed in terms of the two major groups.   (Jones 1960, 172)

Class was another factor which cut across the main Protestant-Catholic division. Jackson expressed this bluntly, in reference to both parts of Ireland:

Let us be quite clear, the problem is one of the working class. There have been no riots in the prosperous areas of Belmont or the Malone Road in Belfast.   (Jackson 1971, 3-4)

THE TERM 'intimidation' was used frequently by witnesses during the inquiry into the 1864 riots in Belfast. It was also a feature of Irish rural violence during the nineteenth century. Townshend has described the extent to which it was ingrained in Irish society at the time:

A whole range of retributive actions was available, and the system of terror was so efficient that extreme violence was not necessary: the anticipation was enough to ensure compliance.   (Townshend 1983, 22)

The term 'intimidation' was not confined to Ireland, of course, and has been used in social science literature, especially in the American research about labour disputes and the Ku Klux Klan, as well as in research into secret societies like the Mafia (e.g. Hardman 1937). Despite this it has not been elevated to the status of a phenomenon with an agreed definition in either psychological or sociological literature. Most psychologists have been content to subsume it under such characteristics as 'fear' and 'threat', where the former is considered as arising from an immediate perception of danger and the latter as an anticipation of future danger. Within the social sciences the only definition discovered was in the 1937 edition of the Encyclopedia of the Social Sciences (Seligman 1937, 239-42), and the category was omitted in the subsequent edition (Sills 1968). The 1937 definition, by E. Hardman, was clear:

Intimidation as a means of achieving desired ends is a feature of behaviour where power or authority is based primarily and essentially on force. The less the public feels bound by standardised ethical norms of conduct, the larger are the opportunities for using intimidation. (Hardman 1937, 239)

Working within this background, and from the use of the term in the Irish context over the last century, it is possible to move towards a tentative definition of intimidation. The main problem is whether it should be defined by the intention of the intimidator, the action of the intimidator or the effects on the victim. Certainly intimidation suggests an intention to produce certain responses from the victim, but if the determination is not manifested in some form of action - a threat or an act of violence, for example - it remains fixed among such psychological concerns as prejudice and stereotyping. A prejudiced attitude may not lead to prejudiced behaviour, some sociologists believing that, on the contrary, 'behaviour typically shapes and alters attitudes' (Raab and Lipset 1971, 35). Further, unless the threat or action leads to some response from the victim, it is indistinguishable from the everyday abuse common in divided societies. So it is a premise of this research that intimidation requires action rather than intention, and should be defined by both the actions of the perpetrators and their effects on the victims.

In Ireland, on the basis of the history of violence since 1835, intimidation has been confined to two main areas. The first was pressure on people to move from their workplaces. The attacks on Catholic workers in the docks and shipyards, and on factory workers from both religions, have been features of all major riots since 1857, and pubs and shops in hostile areas were prime objects for attack in 1886 and 1969. However, this has been a subordinate theme to the extensive and recurrent intimidation of people from their homes. This predominant form of intimidation involved a much greater variety of settings - minority families forced to evacuate hostile streets; mixed-marriage families pressured in both ghettoes; islands of families departing from alien neighbourhoods; even entire districts or housing estates changing in their religious composition, where demographic trends appeared to endanger the future security of whole communities.

In all these settings it is clear that the victim's perception of what was happening was as important in determining responses as the intention of the perpetrator. What was perceived as a direct threat by one person was sometimes dismissed as a chance occurrence by another.

For the purpose of this research, intimidation is defined as the process by which, through the exercise of force or threat, or from a perception of threat, a person feels under pressure to leave home or workplace against his or her will.

It can be considered within a framework of three categories, acknowledging that they are not initially exclusive, or discrete:

(1) actual physical harm;
(2) actual threat;
(3) perceived environmental threat.

Physical harm has taken a variety of forms during the Belfast riots from 1857 to 1969. Workers ejected from the shipyards or factories, or attacked while going to their workplace; pets and children beaten by other children because of family religion; husband or wife jostled or beaten; stones, bottles petrol bombs or bullets directed through windows or doors; houses and public houses ransacked whether vacant or occupied; furniture piled up in the street or burned; petrol bomb attacks resulting in damage or destruction to buildings; a member of the family shot, injured or killed. The list is almost endless, and has altered little since the events described in 1857.

2. Actual threat
The threat of physical harm falls just short of the use of force as an effective form of intimidation. It is designed to induce fear of physical danger. Although no violence or destruction has taken place, the threat is sometimes sufficient to persuade its victims to leave their homes or jobs. This type of psychological threat can take many forms, from a poison pen letter or anonymous telephone call to masked men calling at the door. Bullets have been sent in envelopes to minority families; in North Belfast in 1970 some Catholic families had mass cards posted to them; Protestant families have had slogans painted on their doors. This sort of pressure, like actual physical harm, is directed against specific individuals. It was one of the most common factors which caused housing movement.

3. Perceived environmental threat
The two categories of real physical harm and real threat are clearcut cases of intimidation, in that one person or group is trying to force an unwilling response from another. Beyond these lie the complex cases in which the perception of the victim plays a more important part than any objective personal threat. As with the other categories, this contains a wide spectrum of conditions. For the purpose of this research, they may be considered, however, in two main groups: specific environmental pressures and general environmental pressures.

Specific environmental pressures include a variety of conditions within one's immediate community which create a feeling of unease or an impression that intimidation might occur, even though no specific threat has been made: neighbours becoming unfriendly; children finding it more difficult to find friends; an increase in the number of political or sectarian slogans on pavements or walls. In their mildest forms these changes can convince parents that their district is no longer suitable for rearing their children. They can, however, also take forms which, while not directed at individuals, are symptoms of a more general violence in the community - families caught in crossfire between paramilitaries and army; houses in areas subject to frequent bombings. The desire for a more peaceful neighbourhood can be a powerful motivation for a family to leave its home.

General environmental pressures result, not from changes within a community, but from pressure on the community itself. These pressures have been strong enough to produce enforced population movements, not because individuals have been attacked or threatened, but because the communities to which they belonged have themselves become isolated and vulnerable. There are many examples of communities becoming detached from their broader religious heartland through gradual demographic changes. The Catholic enclave of Willowfield in east Belfast, or Protestant New Barnsley surrounded by Catholic estates, provide two instances where communities eventually disappeared as the result of perceived vulnerability within the broader urban setting. There is evidence that Catholics and Protestants have regarded their communities in such strategic terms since the 1835 riots, and have carefully monitored the shifting patterns of religious demography.

Both these latter groups are distinguished from the earlier categories in one important way. Intimidation which involved actual physical harm or actual threat is designed to force an unwanted course of action on their victims. In the case of perceived threat, the individual is not threatened, but feels insecure as a result of general factors operating within or outside the community. This is here described as environmental threat, and falls within the definition of intimidation in this research.

It is more difficult, however, to deal with the importance of the perceptions of victims, as distinct from observable external factors. In reality this distinction is not so easily sustained. There is substantial research evidence that the people most affected psychiatrically by civil violence are those with a history of mental disorder (e.g. Lyons 1971 and Heskin 1981), and clearly a person's decision to leave a house or job may be explained by perceptions of reality as much as by the reality itself. No sharp line exists between objective and perceived threat. Nevertheless, for the purpose of this study, individuals who leave their homes or jobs without evidence of physical attack, real threat or environmental threat are regarded as belonging to the province of psychiatric study, and are not included in the definition of intimidation.

The complex processes of intimidation, ranging from physical attack down to environmental intimidation, can be presented in the form of a composite model.

These categories, while useful for the analysis of intimidation, are far from discrete and may be regarded as a progressive process. The time likely to elapse between each stimulus and the decision of the victim is incorporated in the model; hence physical threat would seem more likely than environmental threat to lead to a decision to move. Whether or not the stimuli actually produce evacuation depends on the character and personality of the victim, as well as on his or her domestic circumstances. There have been many cases of neighbouring families, apparently intimidated to equal degrees, responding in quite different ways, some remaining and some departing.

There is a difference in kind between the first two categories and the third. In the first two a direct action against an individual target - whether violence or threat - is likely to produce fear, the reaction to an immediate danger. In the case of environmental threat, on the other hand, the response is more accurately described as anxiety, which is essentially anticipatory in nature, an attempt to cope with the future, however ineffectively.

So should environmental threat be considered as a form of intimidation? Is it possible to have intimidation without the intention of the aggressor to intimidate? The problem in answering these questions lies in measuring intent. What can be said in the end is that environmental threat was indisputably a major reason for some families leaving their homes, and that it was regarded as such by many evacuees. Any classification which rejects it ignores a factor which provides a better explanation of enforced population movement than an accumulation of individual acts of aggression. If it was not intimidation, it was a potent form of threat.

A STRONG underlying theme in the analysis of social conflict is its tendency towards unqualified violence between the participants. 'To introduce the principle of moderation into the theory of war itself,' Clauswitz wrote in relation to international conflicts, 'would always lead to logical absurdity.' In internal conflicts too, according to Coleman, 'the harmful and dangerous elements drive out those which keep the conflict within bounds'; he described this as a 'Gresham's law of conflict' (Coleman 1957). Even Simmel, the apostle of sociality, believed that 'every war in which the belligerents do not impose some restriction on the use of possible means.. . becomes a war of extermination' (Lawrence 1976, 140). A major element in this process, at both international and community levels, is a growing willingness to view one's opponents as a distinct and malevolent species against which the most extreme measures are justified. Geographical segregation and a low level of contact between the antagonists greatly enhance this view.

So the duration of a conflict and increased levels of segregation are often important factors in intensifying inter-group hatred and violence. Both conditions appear to apply in Northern Ireland, and particularly in those communities which have suffered from intimidation. In some respects it could be claimed that the conflict between Catholic and Protestant has gone on for more than three centuries; its latest period of prolonged violence endured from 1969 to 1986, longer than ever before. During these years the levels of segregation and political polarisation have certainly increased in the intimidated districts, and probably in general. Nevertheless there is strong evidence that community violence has diminished rather than risen in intensity at both provincial and local levels.

Having reached a peak of 468 deaths in 1972, there was a decline in the annual casualty figures to sixty-four in 1984; the proportion of civilian - as opposed to admitted combatant - deaths diminished, and deaths from direct violence between the communities almost disappeared. By 1984 it was difficult to find any examples of the sectarian rioting which had been the main form of violence in 1969 and 1970. It was as if the conflict had reached a peak around 1972, like 'the wall' in a marathon race, and subsequently settled down to 'an acceptable level of violence'. The question is: why did it develop in this way rather than drawing the two communities into more violent confrontation?

One reason is to be found in the dynamic of the violence itself. Since 1970 its main form has changed from sectarian rioting into a guerrilla war between republican paramilitaries and the British army. From then the conflict was increasingly fought through surrogates, and became somewhat ritualised. This allowed many people from both sides to withdraw from direct involvement in violence, supporting but not participating directly in the battle between their champions. In addition, at least one major cause of mass rioting during the early 1970s in Belfast - the attempt to remove strangers from one's community - had been resolved in many parts of the city by the mid-1970s; it is an irony that 'successful' intimidation in 1971, by reducing the number of minority families - and hence the occasions for tension - in urban communities, helped to account for its virtual disappearance by 1975. However, these are superstructural reasons for the decline in violence. The fundamental causes lie within the communities.

Shibutani and Kwan believed that the most important factors in determining the course of each conflict were the peculiar interrelations between the combatants: 'What each side does is a response to the actual or anticipated moves of its opponents; thus the course of events is built up by social interaction' (Shibutani and Kwan 1965, 135). This process often intensifies the conflict by creating a spiral towards unrestrained violence. In Northern Ireland, however, the same reciprocal process has controlled rather than stimulated the spread of violence. The long duration of the conflict between Catholics and Protestants over three centuries has led to the evolution of social mechanisms to regulate and control their relationships. These were the consequence of two hostile groups inhabiting what A. T. Q. Stewart called the same 'narrow ground'; unable to remove each other and unwilling to assimilate, they gradually evolved forms of relationships which regulated rather than resolved their antagonisms. The mechanisms were each appropriate to particular settings, varying between urban and rural conditions, and in accordance with the religious ratio peculiar to each locality. They were not always successful.

These mechanisms have been an effective control on the spread of violence, and the most significant of them are avoidance, selective contact and functional integration.

Avoidance
In some parts of Northern Ireland like the Upper Ashbourne, Catholics and Protestants have effectively avoided conflict by avoiding each other. 'We were never taught to hate Protestants,' wrote Eamonn McCann about his childhood in Derry. 'Rather we were taught that it was for the best that we did not know them' (McCann 1974, 21). The desire for physical barriers is particularly understandable in communities which have experienced high levels of violence.

The principal advantage of segregation is the reduction of inter-community violence. In Kileen/Banduff, for example, attempts to maintain a mixed community having failed, the two groups settled for segregation as a means of protection. From 1976 when the two communities had become clearly defined and separated, the number of disputes between Protestants and Catholics diminished greatly; at the same time the number of disputes between groups within each community increased. The internalisation of conflict was even greater in the Upper Ashbourne estates, where polarisation was much greater. Effectively Protestants and Catholics, in two exceptionally tense communities, had reduced the level of local violence by turning their backs to each other.

Avoidance is a less appropriate response to local violence in more integrated localities. The relatively high level of demographic integration in Dunville created greater loyalty to a common community, and this has been a significant control on sectarian violence in the town. Within this 'common community' the two groups were conscious of their separateness. However this was expressed in controlled and formal settings. Annual processions in Dunville regularly led to open conflict, but they generally took place on a limited number of set occasions, which meant that arrangements could be made to limit their effects. Similarly, the virulence and extravagance of sectarian exchanges during council meetings contrasted with the generally good personal relationships between citizens. At times it appears that the hostilities have become ritualised and limited to the letting off of surplus steam under conditions which were highly conventional.

Selective contact
Complete avoidance between Catholics and Protestants is impossible in most parts of Northern Ireland. Most people may 'mingle with a consciousness of the differences between them' (Beckett and Glasscock 1967, 188), but they do mingle. It is possible for members of the two groups to develop relationships without abandoning their separate basic allegiances. In Ballybeg, for instance, while they were not prepared to attend the film shows in the Catholic church hall, Protestants often drank in the same pubs as Catholics (Harris 1972). Similarly the novelist Robert Harbinson described how, although he shunned and feared Catholics during his childhood in Protestant Sandy Row, it was possible for him to become friendly with a Catholic family while on holidays in another part of the province (Harbinson 1960).

This position of 'co-operation - but only up to a point' is well illustrated in Kileen/Banduff. Since 1976 demographic polarisation has been almost absolute, and contact cautious. Catholics in Kileen were prepared to welcome visiting Protestant old people for meals in their community centre, but not for prayer meetings. Protestants in Banduff shared in the management, teaching and training in Shamrock Youth Training Centre, but the contacts thus made had rarely extended into broader friendships. Even more graphically, cooperation between two local schools, although successful, was stopped short of a point where the closeness would be seen as a threat to the existing relationship.

So the sharp divisions of the early 1970s had begun to ease in some areas, but the circumstances in which re-engagement could happen were carefully controlled. This sort of periodic reintegration is not new. There is evidence of a similar process in the nineteenth century, when minority families were removed from the same districts after each bout of intimidation. The conditions in which cross-community relationships may be resumed are not constant; they change as local circumstances become more violent or tranquil.

The advantage of these situational variations, when contact may take place in some respects but not in others, is that they permit changes to evolve at their own speed. Even during periods of tension, it is possible for a Protestant and Catholic who would regard each other with suspicion if seen in the other's district at night to suspend animosities and continue to work on the same work bench during the day. Two people who would be suspicious of each other in one setting may be able to develop a co-operative relationship in another. On the other hand this did not extend to exempting policemen from the penalties of war when they were off-duty - even then they were regarded as 'legitimate targets'. Indeed it could be argued that the IRA decision to attack not only off-duty policemen and soldiers, but also men who had actually left the forces, illustrated the tendency for the controls of conflict to weaken as violence continued. The limits to the relationships were determined by the context, and the context itself varied through time.

Functional integration
While relationships with members of the other group may vary with circumstances, certain groups in any community are more disposed towards co-operation. Kirk, in his study of Lurgan, found that Protestants and Catholics from the middle classes were much more likely to be in contact than those from the working classes through common membership of clubs and societies (Kirk 1967). Harold Jackson went further:

Let us be quite clear, the problem is one of the working class. There have been no riots in the prosperous areas of... the Malone Road in Belfast.  (Jackson 1971)

There is also evidence that common material or social interests can overcome sectarian suspicions. The main reasons for cross-religious contact in Kileen/Banduff, for example, were shared interests between teachers and shared concern about unemployment in the Shamrock workshop. The Gingerbread group in Dunville provides another example. It included among its members women whose backgrounds cut across social class and religious divisions, but who had a problem in common. The group continued to flourish despite the strong divisions in the town during the hunger strikes, and despite the fact that it included people who had been intimidated from their homes. In all these cases the motivation was a common interest or concern, and the fact that the resulting relationships included people from both groups was incidental. 'Reconciliation groups', whose objective was to bring together Catholics and Protestants, were much less successful in both Kileen and Dunville. In districts which had experienced sectarian violence, the argument that people should get together because they were hostile towards each other was not persuasive.

All these mechanisms act as restraints on the conduct of the two conflicting communities. Rather than presenting a single model which applies to Northern Ireland as a unit, they demonstrate that intergroup relationships should be regarded as a spectrum. At one end is a highly polarised, potentially violent relationship; at the other a high level of co-operation and interaction. Different communities throughout Northern Ireland can be found at every point. Individuals too do not take a consistent position within the spectrum; rather their position may vary with the setting or situation in which the cross-group contact takes place. Nor is the spectrum itself static; it has altered through time, and the alterations have accelerated most dramatically in times of community violence.

The main function of the mechanisms is to reduce and manage community violence at local levels. They are more successful in some areas than in others. But the cumulative effect of so great a variety of micro controls also constitutes a macro control. In effect they are obstacles to absolute group cohesion for both communities, and therefore to a more extreme and genocidal form of conflict. The variety of local relationships has produced within both religious groups a gap between the members at the centre and those at the periphery.

At the centre is the person who lives in the heartlands and whose contact with the other religious group is minimal. His political and cultural attitudes are not contaminated by the views or desires of his opponents, because he does not know what they are. He is relatively untouched by either the ties or the fears of his co-religionist who lives among members of the other group. The problem seems uncomplicated and the solution - 'Smash the IRA' or 'Brits out' - simple. All that goes on around him, even the internal disputes, confirms him in the uncompromising purity of his position.

On the periphery is the person who lives in a more-or-less integrated area. He shares cultural connections with the other side which, if they do not amount to a consciousness of sharing what Leyton called 'the one blood', create an obligation of decent behaviour. His friends and workmates are likely to include members of the other religious group. If he belongs to the local minority, he may be aware of his vulnerability if violence should spread to his locality. Consequently his inclinations and apprehensions are both more likely to urge him towards accommodation and compromise. In the view of his coreligionists at the centre he is a trimmer.

The Northern Ireland problem has often been described as a conflict between two communities. In another sense it is a conflict between two different concepts of community. On the one hand most people in Northern Ireland, whether they wish it or not, recognise that they are born into a Catholic or Protestant community which shares beliefs, culture and problems, but not a geographical base. They are also born into a geographical community, a small localised territorial group defined as a unit by its members. Both carry obligations and loyalties.

Both concepts of community might be regarded as magnets which exert pulls on the loyalty of the individual. In less dramatic times the strength of the ethnic magnet is determined by a number of factors, including local religious ratios, previous experience of violence and distance from the ideological heartlands. In times of extreme tension and violence, the attraction of the ethnic magnet becomes more powerful. During the UWC strike in 1974, for example, many Protestants felt the need to testify to their membership of the Protestant community by demonstrating or manning barricades, even at the price of antagonising or frightening their Catholic friends. Seven years later the hunger strikes had a similar effect on many Catholics, and Protestants were alarmed by the number of previously nonpolitical Catholics who attended the funerals of dead hunger strikers. Even districts which took pride in their long history of good community relations were affected by the pull. Ethnic identities, like seeds, could lie inert for decades and still retain their fertility. They are activated, not by the duration of conflict, but by its periodic eruption into spasms of intense violence. In the final analysis, however, the more violent the conflict becomes, the more likely are the outlying members of the group to be pulled towards positions defined by the centre. The controls provide no guarantees.

R. Ardrey, The Territorial Imperative, Collins 1967

P. Arthur, The Government and Politics of Northern Ireland, Longman 1980

E. Aunger, In Search of Political Stability: a comparative study of New Brunswick and Northern Ireland, McGill-Queens Press 1981

D. Barritt and C. Carter, The Northern Ireland Problem, OUP 1962

J. Barrow, Tour Round Ireland, John Murray 1836

A. Beals, Cutture in Process, Holt Rinehart and Winston 1967

J. C. Beckett, A Short History of Ireland, London 1952

J. C. Beekett, The Making of Modern Ireland, Faber 1966

J. C. Beckett and R. Glasscock (eds), Belfast: the Origin and Growth of an Industrial City, BBC 1967

J. C. Beekett et al., The Ulster Debate, Bodley Head 1972

Belfast Riots Inquiries, British Parliamentary Papers, XXXVI, 1857; XXVIII, 1865; XVIII, 1887

D. Bell, 'Ethnicity and Social Change', in N. Glazer and P. Moynihan, (op. cit.) 1975

J. Bernard, 'The Conceptualisation of Intergroup Relations with Special Reference to Conflict', in G. Marx (op. cit.) 1971

B. Berry and H. Tischler, Race and Ethnic Relations, Houghton Muffin 1978

D. Birrell, 'Relative Deprivation as a Factor in Conflict in Northern Ireland', Sociological Review, 20, 3,1972

R. Black, The Growth and Role of the Whiterock Tenants' Association, Internal paper, Northern Ireland Community Relations Commission 1971

F. W. Boal, 'Close together and far apart', community Forum, 4, 3, 1972 F. W. Boal, P. Doherty and D. Pringle, The Spatial Distributionof some Social Problems in the Belfast Urban Area, Northern Ireland Community Relations Commission 1974

F. W. Boal and J. Douglas (eds), Integration and Division, Academic Press 1982

K. Boehringer, 'Discrimination: Jobs', Fortnight, 14 May 1971

A. Boserup, 'Contradictions and Struggles in Northern Ireland', Socialist Register, 1972

A. Boyd, Holy War in Belfast, Anvil 1969

R. Buchanan, 'The Planter and the Gael', in F. W. Boal and J. Douglas (op. cit.) 1982

I. Budge and C. O'Leary, Belfast: Approach to Crisis, Macmillan 1973

F. Burton, The Politics of Legitimacy: Struggles in a Belfast Community, Routledge and Kegan Paul 1978

S. Buschschluter, 'ETA froth on top of Basque ferment', Guardian, 20 February 1984

E. Cairns, 'Socialisation in children', in Harbison and Harbison (op. cit.) 1980

Cameron report: Disturbances in ,Northern Ireland, HMSO Belfast Cmd 532, 1969

B. D. Clark and M. B. Cleave, Social Patterns in Cities, London 1973

J. Coleman, 'The Dynamics of Conflict' (1957), in G. Marx (op. cit.) 1971

Commissioner for Complaints, Annual reports, Belfast

P. A. Compton, Northern Ireland: a Census Atlas, Gill and Macmillan 1978

P. A. Compton and F. W. Boal, 'Aspects of the intercommunity population

balance in Northern Ireland', Economic and Social Review, 1,4,1970

L. Coser, The Functions of Social Conflict, Routledge and Kegan Paul 1956

L. Coser, Georg Simmel, Prentice Hall 1965

L. Coser, 'Some sociological aspects of conflict', in G. Marx (op. cit.) 1971

E. S. Cowan, The development of a rural newspaper, Dissertation, Department of Communication Studies, University of Ulster 1984

R. S. Cunningham, St Columba's Church: Historical sketch, Castle Press, Belfast 1978

R. Dahrendorf, 'Integration and values versus coercion and interests', in G. Marx (op. cit.) 1971

J. Darby and G. Morris, intimidation in housing, Northern Ireland Community Relations Commission, Belfast 1974

J. Darby, Conflict in Northern Ireland, Gill and Macmillan 1976

J. Darby, D. Murray, D. Baits, S. Dunn, S. Farren and J. Harris, Schools Apart?, NUU Coleraine 1977

J. Darby (ed.), Northern Ireland: The Background to the Conflict, Appletree 1983

L. de Paor, Divided Ulster, Penguin 1970

M. Deutsch, 'Conflicts: productive and destructive', Journal of Social issues, XXV, 1, 1969

H. Donnan and G. McFarlane, 'Informal social organisation', in J. Darby (op. cit) 1983

S. Dunn, J. Darby and K. Mullan, Schools Together?, Centre for the Study of Conflict, University of Ulster, Coleraine 1985

E. A. Eager, A Guide to Irish Bibliographical Material, Methuen 1973

M. Esman (ed), Ethnic Conflict in the Western World, Cornell University Press 1977

Field Regiment, British army, Community Relations Report, 2 July 1972

N. Gist and A. Dworkin, The Blending of Races, Wiley 1972

N. Glazer and P. Moynihan (eds), Ethnicity, Harvard University Press 1975

R. Harbinson, No Surrender, Faber and Faber 1960

J. and J. Harbison (edo), A Society under Stress, Open Books 1980

E. Hardman, 'Intimidation' in Seligman (op. cit.) 1937

R. Harris, Prejudice and Tolerance in Ulster, Manchester University Press 1972

S. Heaney, North, Faber and Faber 1975

A. C. Hepburn, 'Catholics in the north of Ireland, 1850-1921: the urbanisation of a minority', in Hepburn (op. cit.) 1978

A. C. Hepburn (ed.), Minorities in History, Edward Arnold 1978

K. Heskin, Northern Ireland: A Psychological Analysis, Gill and Macmillan 1980

M. W. Heslinga, The Irish border as o cultural divide, Van Gorum, Assen 1962

J. Hewitt, 'The colony', (reprinted in Commun ity Forum, 4,1,1974), 1950

J. Hickey, Religion and the Northern Ireland Problem, Gill and Macmillan 1984

Independent Television Authority (ITA), Religion in Great Britain and Northern Ireland, ITA 1970

H. Isaacs, 'Basic group identity: the idols of the tribe' in Glazer and Moynihan (op. cit.) 1975

H. Jackson, The two Irelands: a dual study in inter-group tensions, Minority Rights Group, London 1971

R. Jenkins and P. Smoker, Northern Ireland. a case study in polarisation, Peace Research Centre, London 1971

R. Jenkins, Hightown rules: Growing up in a Belfast housing estate, National Youth Bureau 1982

R. Jenkins, Lads, Citizens and Ordinary Kids, Routledge and Kegan Paul 1983

S. Jenvey, 'Sons and haters', New Society, 20 July 1972

E. Jones, A Social Geography of Belfast, Oxford University Press 1960

M. Keenan, Moyard: the Case for Demolition, Moyard Housing Action Committee, Belfast 1982

T. Kirk, The Religious Distribution in Lurgan, M. A., Queen's University Belfast 1967

L. Kuper, 'Theories of revolution and race relations', in J. Stone, (op. cit.) 1977

L. Kuper, The pity of it all, University of Minnesota Press 1977

L. Kuper, Genocide, Penguin 1981

R. La Piere, Sociology, McGraw-Hill 1946

P. Lawrence, Georg Simmel, Nelson 1976

E. Leyton, 'Spheres of influence in Aughnaboy', American Anthropologist, 72,1970

E. Leyton, The one blood: Kinship and class in an Irish village, Social and Economic Studies, Memorial University Newfoundland 1975

S. Lieberson, 'A societal theory of race and ethnic relations', in C. Marx (op. cit.) 1971

Londonderry Riot Reports, 1869 and 1884

C. Lundberg, The Foundation of Sociology, Macmillan (NY) 1939

F. S. L. Lyons, Ireland Since the Famine, Weidenfeld and Nicholson 1971

F. S. L. Lyons, Culture and Anarchy in Ireland 1890-1 939, Clarendon Press 1976

E. McCann, War in an Irish Town, Penguin 1974

J. L. Mc Cracken, in F. McManus (op. cit.) 1967

E. McEldowney, 'The Vere Foster affair', Irish Times, 17-18 June 1972

N. Mansergh, The Government of Northern Ireland, Allen and Unwin 1936

F. McManus (ed.), The rears of the Great Test, Mercier 1967

C. Marx (ed.), Racial Conflict, Little Brown 1971

A. Megahey, The Irish Protestant churches and social and political issues 1870-1914, Ph.D. thesis, Queen's University Belfast 1969

G. Mitchell, A Dictionary of Sociology, Routledge and Kegan Paul 1968

M. Moneypenny, The Two Irish Nations, London 1912

Moyard social survey, described in Sirockin and Keenan (op. cit.) 1982

J. Mullan, 'Christmas crime control in Andersonstown', Fortnight, Belfast, 21 January 1985

K. Mullan, Research draft on Cookstown, unpublished, 1984

D. Murray, 'Schools and conflict', inJ. Darby (op. cit.) 1983

R. Murray and R. Osbourne, 'Segregation on Horn Drive - a cautionary tale', .New Society, 21 April 1977

S. Nelson, Ulster's Uncertain Defenders, Appletree 1984

Northern Ireland Community Relations Commission, Flight, NICRC, 1971

Northern Ireland Housing Executive, Annual Reports, NIHE, 1970-84

Northern Ireland Housing Executive, Moyard Estate, Internal paper, NIHE, n.d.

Northern Ireland Housing Executive, Highfield estate, Internal paper, NIHE, 1973

Northern Ireland Housing Executive, Suffolk estate Belfast, Internal paper, NIHE, 1983

Northern Ireland Housing Executive, Housing plan for Cookstown, NIHE, 1984

E. F. O'Donnell, Northern Irish Stereotypes, College of Industrial Relations Dublin 1977

R. Osborne, 'Fair employment in Cookstown? A note on anti-discriminatory policy in Northern Ireland', Journal of Social Policy, 11, 4, 1982

R. Park, Race and Culture, Free Press of Clencoe 1949

J. Peel, 'Ethnicity in Nigeria', paper read at New University of Ulster, November 1980

M. Poole, Evidence submitted to the Cameron tribunal, typescript, 1969

M. Poole, 'Riot displacement in 1969', Fortnight, August 1971

M. Poole and F. W. Boal, in B. D. Clark and M. B. Cleave (op. cit.) 1973

E. Raab and S. Lipset, 'The prejudiced society', in C. Marx (op. cit.) 1971

B. Rolsion et al., A social science bibliography of Northern Ireland 1945-83, Queen's University Belfast, 1983

D. Rea (ed.), Political co-ape ration in divided societies, Gill and Macmillan 1982

J. Rex, 'Conflict', in C. Mitchell (op. cit.) 1968

R. Rose, Governing without consensus, Faber 1971

R. Rose, Northern Ireland: A time for change, Macmillan 1976

J. Russell, 'No solution in North', Irish Times, 14 April 1980

R. Schermerhorn, These our people, Heath 1946

R. Seligman (ed.), Encyclopedia of the Social Sciences, London 1937

T. Shibutani and K. Kwan, 'Changes in life conditions conducive to interracial conflict' (1965), reprinted in C. Marx (op. cit.) 1971

C. Sills (ed.), Encyclopedia of the Social Sciences, London 1968

C. Sirockin and M. Keenan, 'Moyard - the tenants are angry', Science for People, 53, 1982

S. Starling, The sociology of voluntary organisations in Belfast, Geography

department, Queen's University Belfast 1971

J. Stephen, 'On the suppression of boycotting', The nineteenth century, cxviii, 1886

A. T. Q. Stewart, The Narrow Ground: Aspects of Ulster, Faber and Faber 1977

J. Stone (ed.) Race, Ethnicity and Social Change, Duxbury Press 1977

D. Timms, The Urban Mosaic, London 1971

P. Townshend, Political Violence in Ireland, Oxford 1983

B. White, 'A city divided', Belfast Telegraph, 12 October 1979

A. Williamson, Community integration in a west Belfast housing estate, Geography department, Queen's University Belfast 1970

L. Wilson and W. Kolb, Sociological Analysis, Harcourt Brace 1949