This publication is copyright Democratic Dialogue 1998 and is included on the CAIN site by permission of the publisher and the author. You may not edit, adapt, or redistribute changed versions of this for other than your personal use without express written permission. Redistribution for commercial purposes is not permitted.

Northern Ireland appears to face a hopeless dilemma on policing. Unionists stand for defence of the policing status quo; nationalists reject piecemeal reform of the RUC, while republicans call for its disbandment. Government measures-the Police Bill and proposals for enhanced accountability-have won support from neither.

But this Democratic Dialogue paper, kindly contributed by Mike Brogden, professor of criminology at Queen s University and UN adviser to the South African police, argues there is a positive way forward. Prof Brogden presents the case for two-tier policing as a global trend with particular potential for Northern Ireland.

Crudely put, two-tier policing would sustain a specialised RUC to deal with continuing challenges of terrorism, intercommunal offences and other serious crime. But, alongside, it would introduce a community policing service, unarmed and locally accountable, addressing the quality-of-life concerns at street level using techniques of mediation and problem-solving.

Prof Brogden believes this approach can reassure unionists as to their security while offering to nationalists now a service they can endorse and participate in. Without miimising the practical difficulties, he contends that the 'downsizing' of the RUC and the establishment of this new (and cheaper) service could be managed within a 'peace dividend' over a transitional period.

While calling for further research and discussion, Prof Brogden stresses that this would not mean handing policing power to paramilitaries, colluding with localistic sectarianism or creating avenues for corruption.

On the positive side, his proposal would not only provide an avenue out of a political impasse. It would also offer a sensitive and effective response to the profound desire from disadvantaged communities in Northern Ireland for a means to address the concrete problems that often loom much larger in people's lives.

There is no simple resolution to the quandary of police reform in Northern Ireland. Every option has its opponents and would be difficult to implement.^[1]

But a recent survey reported by the Belfast Telegraph suggests there is a common denominator in the plurality of views across the two main religious communities: a two-tiered policing structure could attract broad acceptance.^[2] This is, moreover, an inexorable resolution: cosmetic changes, such as in recruitment patterns (the issue the Northern Ireland Affairs Committee has addressed) solve little.

Indeed, failed attempts to alter the composition of the police in the north of Ireland have a long history. As long ago as 1863, the then Belfast City Police were disbanded by a commission of inquiry, following major disturbances and the sectarian bias displayed by a predominantly Protestant force. One recommended solution was the recruitment of more Catholics …

There is no point either in attempting to change the Royal Ulster Constabulary to a more effective, 'crimefighting', force or to a 'community-police' service.^[3] The history, culture and current priorities, in training and practice, of the RUC reflect the tradition of centralised, continental, gendarmerie forces, not the model of the London Metropolitan Police. The ancestor of the RUC officer was not the 'tythingman' (parish constable) of medieval England but the gendarme of a European Ministry of Police-a model that co-existed with a second tier of local policing.

'Slimming-down' will disgruntle many-inside and outside the RUC-while solving none of the major challenges facing the organisation. And for those who still believe the RUC can be abolished, logistical as well as political problems seem insurmountable.

The answer is reform - but structural, not cosmetic.

Northern Ireland possesses over three times the police (man)power per capita of Great Britain. The 8,500-strong RUC is backed by some 3,000 full-time and 2,000 part-time reservists. This police-to-population ratio of 1:106 compares with 1:350 in Britain, and 1:512 in Sweden.^[4] It costs the Northern Ireland Office some £660 million per annum to police some 1.6 million people.

The force's composition^[5] is skewed by religion, gender and race.^[6] The religious pot-pourri is about 88 per cent Protestant and 8 per cent Catholic. A markedly higher proportion of senior officers are Catholic (presumably having joined before the 'troubles') than of the lower ranks. There is evidence that the composition of the civilian support staff is similarly skewed, although not so much.

About 13 per cent of RUC members are women, who comprise 14-16 per cent of forces in Britain. Women are concentrated in the part-time reserve, more than half of whose members are female.^[7] Only one woman has reached superintendent rank, significantly short of the norm in Britain.

Fewer than 1 per cent of RUC members are from ethnic minorities. While this is little different from the proportion of minorities within the larger population, it is not however a satisfactory figure, given migration in the UK and Ireland and the critical need to create diversity in policing.

Information on support for the RUC is contradictory. A variety of polls suggest the RUC is no less favourably perceived than forces in Britain and elsewhere. There are suggestions, however, of major methodological flaws-for example, in the failure evident in a recent survey to receive responses from those in areas like the lower Falls in Belfast who are likely to be the most antagonistic.

Comparability with Britain, the Republic of Ireland and abroad is meanwhile highly problematic: policing may have a quite different meaning to respondents in Northern Ireland. The experience of armed, paramilitary style policing is distinct from that of the population in Britain; expectations of the RUC as a community-style police may be correspondingly low. And where the RUC is perceived antagonistically, consent to policing may mean little more than a consent to its non-interference in local practices.

Thirdly, poll evidence of RUC support flies in the face of social reality. It is hard to understand how the republican military campaign of the last three decades could have been sustained without considerable minority opposition to the RUC - out of all proportion to public reaction to the British police services and the Garda Síochána.

In any case, polls commonly confuse attitudes to the RUC with views on 'policing'. It is important to distinguish between those who are anti-RUC and those who are anti-police. Communities need and support policing; the question is whether they need and support the policing identified with the RUC.

The RUC, as Graham Ellison has recently documented,^[8] never enjoyed the supposed 'golden age' characteristic of British forces. In particular quarters, it has always been perceived as an armed occupying agency-and has acted reciprocally, as an armed patrol force, especially in border areas. The RUC (like its predecessors) never was a British-style police.

Finally, training in the RUC has of necessity given primacy to self-protection in a manner atypical of British forces. The rank-and-file culture of front-line officers has inevitably been conditioned by a security-conscious environment. Any attempt to transform the RUC to a British-style service ignores not only history and culture but especially the alternative, distinctive policing model from which the RUC derives.

Whatever happens to the 'peace process', the new' RUC must retain a major security as well as 'crime-fighting' capacity. The eventual phased release of some 600 convicted paramilitaries - belonging to clandestine but still organised groups - could be a destabilising element even in a future peaceful society. Such releases, and continuing minority support for paramilitary organisations, could to lead to an increase in organised crime, such as racketeering, in certain communities.

Northern Ireland is an armed society. Well over 100,000 weapons are held legally - albeit mostly airguns and shotguns - and a large if inherently countless number illegally. Offences involving armed force will be a problem for many years. There must always be a trained and armed police force to meet that threat.

According to official statistics, Northern Ireland has the lowest crime in western Europe. But there are reasons for doubting the accuracy of these figures. Them is likely to be much under-reporting in Catholic areas. Many criminogenic activities by young people may have been displaced, albeit temporarily, into activities against the 'security forces'. Many violent offences (such as firebomb attacks on the police or army) are grossly under-recorded, as is community intimidation of minorities. Crimes in the household, white-collar crimes and offences by state personnel are equally badly reported, though likely to be high in Northern Ireland.^[9] Generally, there is no reason to believe Northern Ireland is an especially peaceful society-although in particular contexts, such as drug abuse, there appears to be less of a problem.^[10]

There is little point in RUC reform which pays no dividends in social order, quality of life and crime control. But reform advocates should not delude themselves that making the RUC apparently better at crime-fighting or community policing will actually have beneficial effects.

Doctrinal reforms emphasise mobile patrols, emergency response and investigation. But in practice, as the evidence shows, such orthodox changes generally do not work.^[11]

The conventional tactic for 'crime-fighting', and fighting fear of crime-the random mobile patrol-is largely a waste of resources, which inhibits community contact. Quickening the response to emergency calls has no perceptible effect on arrests (the chance of one drops below ten per cent one minute after an offence) and what most victims want is predictable, rather than rapid, response. Nor are crimes normally solved-in the sense of offenders being arrested and prosecuted- through criminal investigations by detectives. Generally, crimes are solved because offenders are immediately apprehended or someone identifies them; if neither obtains, the chance that a crime will be solved falls below one in ten.

Enhanced clearance rates are not related to changes in the quality or quantity of police. As one north American expert has put it, "If criminals noticed the increased efficiency of the police, they certainly didn't seem to care." Neither the size of police establishments nor their budgets has any clear impact on recorded or 'cleared up' crime. The crime effectiveness of the RUC is determined by other factors than whether armoured Land Rovers are replaced by thin-skinned patrol cars.

Traditional police reforms are most fallible in dealing with so-called 'victimless' crimes-such as drug abuse and prostitution. Most police reform ignores these and other misdemeanours which are often perceived as a communal priority - rather than the serious crime stressed by police culture, organisation and command. Most calls to the police relate to problems other than serious crimes which police departments tend to see as garbage work. As the experience of police liaison committees indicates, quality-of-life issues dominate communal demands-not high - level crime-fighting.

Conversely, foot patrols are important because-while there is some doubt about how much they help to reassure citizens and, especially, to 'catch criminals' - the activities encountered by patrol officers (often banal incidents) include the deviant behaviours that concern most people, most of the time. While a foot patrol does not reduce crime, it can increase public satisfaction with the police and can also reduce the fear of crime. If there is one clear demand from Northern Ireland communities, it is for a return to the (albeit mythical) 'bobby on the beat'.

Converting the RUC in its present framework into a community-policing institution is, however, equally fraught. Community-policing practices have a quite different lineage from that of the RUC. Commonly, they develop from small-town rural forces (as in the US) or stem from the tradition of the unarmed patrol constable (as in England and Wales). In countries which have similar gendarmerie traditions to Northern Ireland- Austria is the obvious case-community policing within a monolithic structure is perceived as the antithesis of appropriate police work. In Austria, community problem-solving is not even regarded as the proper province of the police.

There is no short-cut - as the South Africans have most recently found^[12] - through which a unitary policing institution focused on dealing with anti-state crime can be transformed into a crime-fighting force or a community-police service. A metamorphosis in structure, not simply of practices and recruitment, is the only resolution.

Police reform without changing structures means devoting resources to traditional, bureaucratically safe, approaches that no longer work-if they ever did. Cosmetic change has been tried many times before. Transforming policing in Northern Ireland entails taking a fundamental look at the policing organisation and process.

The civilian structures related to the RUC also require reconstruction. Despite the continuation of innovations from David Cook's period as chair of the Police Authority (PANI)-especially in transparency, the radical entrepreneurship of certain new members and the recent implementation of 'Nolan-style' controls on membership-the authority remains (with minor exceptions) a publicity body for the RUC.^[13] Its message is one-way: explaining the RUC to communities, not holding the police to account on behalf of the community.

There is no local accountability, a crucial matter given the fragmented character of society in Northern Ireland. And it is difficult to understand in what ways PANI holds the RUC accountable regionally.^[14] Despite compositional changes, it is an unanswered question as to what function PANT members play in this regard.

Despite serious attempts to broaden membership, there is no significant representation of the minority community's views on PANI (in part, a function of the current opposition to membership by the SDLP and Sinn Fein). Membership has changed considerably but the degree of representation of minority viewpoints remains questionable. Increased Catholic membership is not the same as increased nationalist or working-class inclusion. Nor is the Police Authority unique as a non-elected body: questions of its accountability must be linked to the non-accountability of other quangos to the wider community.

Related bodies such as Police Liaison Committees and Lay Visitors seem to be similarly skewed in membership (especially in class terms). Inevitably, given the history of Northern Ireland, the Lay Visitors mirror other lay aspects of the criminal justice system - such as the Prison Visitors - including their inadequacies.

The mainstream paramilitary ceasefires represent a major opportunity. This is not so much to start policing afresh in Northern Ireland as to reconstruct policing - to meet progressive professional policing requirements and civilian demands for a qualitatively enhanced police service which is equitable and locally accountable. There is a congruence between inexorable developments in western policing generally and public demands.

Northern Ireland has a substantial civil society. The 'troubles' brought into many self-help agencies, unconnected with the paramilitaries, executing functions which would be carried out by government bodies in other societies. The tradition of 'sorting things out' is one of the major strengths of Northern Ireland's civil institutions. In many urban areas, people have managed to survive amidst the more banal misdemeanours, reasonably independently of a police force which has necessarily had other priorities.

There is a reservoir of voluntary groups (such as Victim Support and Mediation Network, and many smaller organisations), who use lay personnel able to conduct the local mediation and related tasks which are critical to community policing. They have a significant contribution to make to a reconstructed police service. This civil tradition (which reflects the core philosophy of community policing) is a strength that needs reinforcing.

There is a documented demand for this 'ordinary policing'. Opinion polls, for example, regularly emphasise public support for the presence of reassuring foot patrols. Beat patrolling does not require the heavy commitment in technology and training which characterises the present RUC, and enjoys overwhelming public support.

Independently of the 'troubles', the RUC is facing major pressures for change. A 'managerial', consumer ethos is becoming the dominant philosophy of most police forces in western Europe, north America and Australasia. In many ways, the RUC still epitomises what E P Thompson once referred to as the (now-defunct) police feudal baronies of England and Wales.

A different career trajectory is, relatedly, emerging. Out has gone the academic qualification in law and a background in CID and Special Branch; in are coming degrees in social science and career progression through community relations. Managerialist ideologies are replacing feudalism in police command, but the RUC is lagging in that evolutionary process.

Government pressures (via the Sheehy report^[15] and UK-wide expenditure restrictions) are likely to have major effects on the RUC - for example, in the implementation of performance indicators. Value-for-money requirements will have an inexorable influence. To date, the RUC has been somewhat cushioned from the 'belt-tightening' that has affected police services in Britain since the Home Office Circular of 1984.^[16] But with expenditure currently running at £1,000 per annum for every man, woman, and child in Northern Ireland, something has to give.

The RUC has also been relatively immune from pressures to civilianise and privatise. In Britain (as, indeed, in most European forces), many ancillary tasks (such as police station reception) are conducted by less expensive civilian staff. Similarly, functions such as static guard duty and especially the transport of remand prisoners have been handed to cheaper private security organisations. In a peaceful society, the value-for-money pressures for civilianisation and privatisation will be irresistible.

No one model of policing dominates the western world, but unitary models such as the RUC are an exception. Tiered policing is the west European and north American norm.

Most policework on the ground in normal societies involves tasks for which much current technical training is not required: barking dogs, responding to burglary victims, dealing with neighbour disputes and so on. Officers may be over-trained for the banality of beat policing.

In recognition of this, two western societies in particular, France and Holland (both historically having much in common with the old Irish Constabulary), have elevated the importance of the local beat patrol, by developing a separate tier. The Dutch system-often regarded by other professionals as a centre for new ideas in policing-encompasses a second tier of local patrols, with restricted legal powers^.[17] In those countries with unitary forces, some aspects of tiered policing already exist-for example, in the employment of civilian wardens to deal with humdrum traffic matters.

Yet police training (especially in Northern Ireland) pays little attention to those mediation, problem-solving, referral and diversionary functions which traditionally characterised street patrols. While much of the history of the apocryphal village constable is mythological, he (it was always he) was often able to 'sort things out' without resort to arrest and detention. Street patrols do not require qualifications in all the paraphernalia of professional police training.

There are clearly many matters - especially violent offences (some ten per cent of matters dealt with by the RUC), inter-community offences, more serious property crimes and political offences - which can only be dealt with by a professional police officer.

The essence of new policing must therefore be a two-tier structure-a low level police dealing with quality-of-life issues and a high-level police service dealing with serious and inter-communal crime.

The metaphor of traffic wardens, as symptomatic of a two-tiered system, is intended to convey three notions. Certain low-level law enforcement in Northern Ireland is already conducted by non-police agencies. The skills required of such low-level policework are quite different from professional policing involving criminal investigation and related professional tasks. And the community tier deals mainly with local quality-of-life issues.

While traffic wardens are not universally popular, they are not perceived by minority groups as reflecting any coercive or symbolic state functions. Any antagonism towards them is a consequence of their exercise of their specific duties.

A community police service committed to problem-solving practices is first of all a transitional device-a desirable way of developing a working relationship between Catholic communities and the RUC. An individual from a republican background who spends a period in a local community police force will develop working relationships with the larger RUC. One resolution of the RUC's lack of an adequate supply of nationalist recruits is to allow such individuals to prove themselves and their relationship with the RUC (and vice versa) in the half-way house of a community police service.

Mutual trust is developed gradually and incrementally:

despite official statements to the contrary, there is no short-cut to developing professional police legitimacy. A community which does not accept the RUC in its totality may do so when it complements a community service. The assumption that a force obtains consent by conducting low-level work itself is misconceived: the Royal Canadian Mounted Police (RCMP) is a typical high-level force, with an undoubted popularity owing nothing to mundane duties.

Similarly, there is no way that the RUC can fill the current dangerous policing vacuum in certain communities, in the short to medium term. In dealing with low-level policing, a community beat service fills that void.

Community policing by a local service is also much cheaper:^[19] training and equipment requirements are much lower than for a professional police. In particular, it makes use of 'common-sense' skills already existing in the community and in Northern Ireland's many voluntary groups.

Not merely do they not require the training or technology of professional police; they also need no exceptional legal powers. Community police possess only the powers of arrest and prosecution of ordinary citizens. Formal powers of interrogation and detention remain with the professional police. The style is of the reassuring foot patrol - 'sorting things out', rather than utilising the law as the major resource.

Community police act as a referral service: problems beyond their immediate remit and skill are conveyed to an appropriate specialist agency. In the case of serious offences, this would be to the RUC; inebriated, mentally ill and homeless individuals would be referred to relevant statutory or voluntary agencies. They are also unarmed: where coercion is required, the responsibility becomes that of the higher tier of police.

Community policing is about mediation, restoration, problem-solving, referral to specialist agencies and diversion. Beat patrolling is its key practice. Through diversion and mediation it truncates criminals careers, preventing the youth and magistrates courts being inundated by suspects whose misdemeanours are better resolved in the community than dealt with by the full paraphernalia of the law. It deflects young people from descending the slippery slope into full-time crime through judicial processing.

RUC officers would benefit from a different policing agency conducting quality-of-life work. Many officers joined the service to do what they conceive of as 'real' work - serious crime - not to sort out the problems of the drunk-and-disorderly and the misdemeanours of urban street life.

The accountability problem for a community police service is straightforward. Like the RUC, it would be bound by the law and police regulations. It would be accountable to elected community councils-the Northern Ireland equivalent of an English parish council - but it would enjoy the same autonomy from such elected bodies as did English constabularies vis-avis watch committees and police authorities before the 1964 Police Act.

The composition of these Community Policing Councils should depend on an electoral process which allows clear representation of minority groups. Local accountability would ensure face-to-face contact between elected representatives and the police on whom they depend, rather than the latter appearing to be anonymous functionaries recruited from and resident outside the district and accountable only through a long chain of command to a distant authority.

Community policing emphasises consumer choice. In Canada, for example, communities can choose whether they form their own service to conduct local tasks or hire the RCMP (recognising the inappropriateness of much RCMP training for these purposes). A tiered structure does not make the second tier compulsory but gives citizens options. In Northern Ireland, no community should be required to develop a community police force: the emphasis is upon citizen/community choice.

Community policing draws on local skills and knowledge, providing relatively high-status employment for unemployed people who can display the relevant common-sense skills. It makes use of the particular skills and attributes of female members of the community, enabling the recruitment of more women to the police service as a whole. Research demonstrates that women are particularly well-equipped to deal with the mediation problems required of street policing.

One orthodox response has been that the present RUC Reserve constitutes a second tier-why change it? Or why not appoint special constables, as in Britain? The police reserve is perceived by many in the minority community, rightly or wrongly, as an inheritance of the old 'B' constables. Perhaps even more than the full-time RUC, it is infected by the prevailing security culture and would be extremely hard to retrain in new community-style policing requirements.

Both special constables and the RUC Reserve are centrally organised and directed, and trained in generic police duties - the opposite of what is required of second-tier policing. They are not selected in terms of their potential as community constables. Apart from by accident, they have no local roots and, especially, no local accountability.

Secondly, despite rhetoric to the contrary, community policing is certainly not a question of giving power to the paramilitaries. Appointment to the community policing service would be subject to the normal criteria of police recruitment. Conviction for violence offences would preclude membership, with recruitment subject to combined community and statutory requirements.

Following the successful South African precedent, former association with a paramilitary organisation should not however disbar a potential recruit of itself.^[20] Many people drawn into paramilitary structures because of the 'troubles' would otherwise have led law-abiding lifestyles. But membership of a community police could not be open to those whose primary allegiances lay elsewhere.

Accusations of potential corruption against a community police force are also unjustified. Local policing is no more open to corruption, or 'favours', than is national policing. Safeguards must be built into the composition of the local policing authority and the organisation of the community police to prevent control by a sectional or factional group. Legal bulwarks must prevent the tyranny of a majority over a local minority.

Policing must remain an autonomous organisation, subject not to community control but accountability - the requirement regularly to justify actions taken in the vicinity. Local policing must be monitored by an ombudsperson with effective independence from all parties and wide powers of investigation.

Assuming a normal ratio of one community officer for every 340 members of the population, a complement of 150 officers serving some 50,000 people (less in a rural area) provides an optimum size.^[21] It is critical, however, that the areas selected are not heavily dominated by one religious group. Community policing depends upon a heterogeneous mix of peoples: one police force for Catholics and one for Protestants would risk reinforcing schismatic tensions.

The second tier would not usurp the powers of the RUC. Officers of the latter would retain their normal powers. Problems of overlapping jurisdictions have been encountered in many other countries;^[22] there are tried and tested means of dealing with them. Some are laid down by statute and regulation, others by local agreement subject to external monitoring. No police institutions have completely resolved these problems, but a modus vivendi has usually been established.

In essence, the 'new' RUC would remain as the primary tier, committed to dealing with more serious crime, inter-community violations and terrorist offences. It would, however - in line with other western police agencies - adopt the rubric of 'service' in its title. A service embodies the reality of modern progressive policing. The present title is perceived by the minority community as containing an historical note of fealty. The RUC has undergone several name changes in the past; retitling it as either the Northern Ireland Police Service or the Ulster Police Service meets both modern policing and communal requirements. Many individual officers already accept the inevitability of such a retitling.

Symbols such as the badge and flag have particular meanings in Northern Ireland. There is no easy way of dealing with this issue. But it is critical that police stations are seen as 'politically-free' zones. If, for example, the flying of the Union flag on a police station is perceived by the minority as a symbol of communal subjugation, the British practice must be adopted of ensuring no such emblems are displayed in the normal course of events.

Developing a community-policing tier inevitably means reducing the RUC in size and changing its composition, independently of other pressures. Some of this decrease will occur by natural attrition. Some must be secured by large ex gratia payments to younger officers from the financial dividend from the 'peace process' - the South African solution. Some officers will transfer to forces in Britain where establishments are incomplete.

The major loss will be the abandonment of the reserve - approximately one-third of the existing RUC establishment. The bulk of officers will be retained and become more highly skilled in focusing on the central crime-fighting tasks. Recruitment must be stopped immediately- there are many precedents for abandoning recruitment for a fixed period.^[23]

Some will be absorbed in 'selling' police services: the RUC represents a major source of overseas revenue in terms of its counter-terrorism expertise. Third-world police, especially in schismatic societies, frequently seek advice from the west. Traditionally, they have sought training from Metropolitan-type forces, but the paramilitary policing skills of the RUC are far more germane to these overseas forces. For a century, the Irish Constabulary and its successors represented the major training agency for overseas forces, a role which ceased in the 60s. Many such police services - from South Africa to India-owe their origins to Irish policing.

The RUC must be given a higher profile of addressing serious crime, with an increased emphasis on drug-related offences, sexual abuse and violent crimes, as well as continuing its security role. There is no reason it should not continue to be armed: the Dutch police, for example, have hardly alienated the public through the possession of handguns.

Increasing Catholic representation may be less of a problem than is perceived. Recruitment to the second tier will constitute a partial resolution. Recruitment of more Catholics directly to a reduced RUC is more complicated, especially given the composition of the higher ranks. There are two devices utilised elsewhere: lateral entry for individuals with equivalent qualifications in related fields (who would be limited in terms of their policing role) and 'fast-track' promotion.

Both are problematic and could only be developed on the basis of research evidence.

But recruitment of members of the minority community is more important for its symbolic value than its practical implications. Overseas research shows that at the end of the day people do not worry unduly about the composition of the police service as long as it deals with them effectively, fairly and transparently, in relation to critical quality-of-life issues. However, the overall policing structure must include a higher proportion of women - not only for reasons of equity but also because of the evidence that women bring particular skills to many police tasks.

The Police Authority Northern Ireland must be reconstituted in several ways. It must be democratic in composition in order to establish its legitimacy. Despite the recent undermining of the democratic composition of the police authorities in England and Wales, PANI offers an exceptional case for greater community accountability if it is to achieve the necessary legitimacy.

It must also have its own expert administrative staff who can understand and check the minutiae of police budgets and practices. An informed police authority is critical to democratic accountability. Though counter to the current trend in Northern Ireland (where control of police finances has recently been transferred from PANI to the RUC), the fact remains that legitimacy depends in part on effective accounting for police finances. The evidence from other western countries - from Canada to Norway - is that the new mode of accounting for the police is to replace political forms of accountability (which have largely failed) with financial controls.^[24]

Principally, the members of PANI must be tutored to understand that the one-way street of police accountability - explaining the police to the public - must take two-way traffic: requiring the police to take account of community requirements, including minority views.

Civilian structures such as Police Liaison Committees and Lay Visitors can only function properly, and be accorded communal legitimacy, if they are representative of diverse communities. Appointments to them should be via the Community Policing Councils.

The RUC must lose its formal prosecuting role. Northern Ireland (together with the republic and New Zealand) has become anachronistic in lacking an independent prosecuting body, such as the Crown Prosecution Service in England and Wales. The Director of Public Prosecutions' office has in practice moved a considerable direction towards a de facto CPS. But this needs affirming in statute, especially where matters of 'public interest' are concerned-such as the controversy over prosecutions in relation to the blocking of highways during the 1996 Drumcree protest.

Given that a major impetus towards removing the power of police prosecution in England and Wales was the 'Guildford four' and 'Birmingham six' cases, it is anomalous that Northern Ireland has not been directed to develop the same safeguards against wrongful prosecution. Whatever the criticisms of the CPS in England and Wales, it has brought that domain - Scotland, through the Procurator Fiscal, has always possessed such an intermediary office-into line with the rest of the western world. The development of a CPS (or equivalent) in Northern Ireland would remove any suspicion of police partisanship in determining prosecution practices.

Generally, the changes proposed in the Hayes report on police complaints in Northern Ireland-in particular, a new complaints ombudsperson - are welcome. But recent changes in Britain highlight the need also for an independent body of investigators, including former police officers familiar with the nuances of policing - a practice becoming the norm elsewhere. Complaints systems work most effectively when they consist of combined teams of trained civilian investigators and senior former officers.

The Police and Criminal Evidence Act (PACE) has gradually been introduced into Northern Ireland. While there have been several criticisms of such legislation - especially in the way it substitutes internal accountability (in terms of professionalism) for external oversight - this is a welcome development. But it must be extended to cover 'terrorist' or scheduled offences, currently excluded from parts of the PACE remit. And legal loopholes need closing. For example, the police practice of preventing oversight of stop-and-search and detention by using alternative legislation to PACE-such as the Misuse of DRUCs Act or the Prevention of Terrorism Act, which contain no such transparency safeguards-must be stopped.

A recent unresolved controversy relates to the priority of functions of the RUC: to defend the state or to defend the rights of citizens. In law, this is obscure and open to partisan debate. Legislation is required to clarify that the primary function is the second: defending the state is a matter for the armed services, not the RUC. The RUC requires 'normalising' in terms of its legal policing mandate.

It is a cliché that you cannot change policing without changing political structures. Success in the 'peace process' is intertwined with proposed changes in policing. Transforming policing is part of the larger mosaic-it is neither a prerequisite nor a postscript.

This paper sets out such a policing strategy. Many Protestants identify the maintenance of the present RUC with the security of the state. Many Catholics identify the maintenance of the present RUC with a threat to their political status. What the two communities have in common is the need for an effective policing organisation.

For the majority community, this can be ensured in two ways: first, by maintaining a primary tier of 'high' policing to deal with major problems of public order and inter-community offences; and, secondly, by providing a second tier to meet their wish for a local, visible, beat-patrol service. For the minority community, the development of two tiers removes any perceived threat of partisanship within the normality of everyday life and, like the majority community, they support a local, visible, beat-patrol service. For them the character and composition of the higher tier may not be perfect. But it is not just a political compromise: it potentially furnishes a qualitatively enhanced police service.

Some politicians and figures associated with current policing structures will misrepresent this proposal - indeed some already have - as giving power to the paramilitaries. It thus bears repeating that the proposed second-tier force would (a) exclude current paramilitary members, (b) exclude those with convictions for violent offences, (c) be recruited by regulated procedures, (d) be organised to represent diverse communities and (e) be unarmed.

Some senior RUC officers, contending with an internal audience of juniors who feel their jobs are threatened, will express more sincere reservations. But the former-especially the growing number with experience beyond these shores - will also be aware that the tide is against unitary policing.

Finally, some people will oppose the proposal reasonably, on the basis of the problems inherent in any such transitional process. But no resolutions are easy. All change processes involve difficulties and risk.

Universal principles are one thing; applying them to the complexities of Northern Ireland is another. The devil is in the detail. The broad principles of two-tier policing are easy to establish, but much research and discussion is required on specific procedures and practices. Especially important is ground-level consultation.

There is no easy solution to the dilemmas of transition in Northern Ireland, to structures more conducive to the policing of a peaceful society. Minor organisational changes are unlikely to satisfy any major constituency or to improve the quality of relevant policework. Experiments such as broadening the recruitment base have been tried (and failed) many times before.

The two-tier model of policing which increasingly characterises other police services seems the obvious way forward. It would resolve the internal problems of providing an economic and effective police service relevant to the year 2000.

It would also be the most effective mode of police organisation for Northern Ireland's particular problems. The people of Northern Ireland desire an effective police service. More than anything else - rightly or wrongly, in the eyes of police professionals - they want to see the 'bobby on the beat'.

Two-tier policing is the only way to achieve that common demand, in an effective, economic, equitable - and locally accountable - manner.