UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Holding the police to account in a divided society : the Northern Ireland reforms McGovern, Clare Joanna 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2004-0555.pdf [ 5.13MB ]
JSON: 831-1.0091642.json
JSON-LD: 831-1.0091642-ld.json
RDF/XML (Pretty): 831-1.0091642-rdf.xml
RDF/JSON: 831-1.0091642-rdf.json
Turtle: 831-1.0091642-turtle.txt
N-Triples: 831-1.0091642-rdf-ntriples.txt
Original Record: 831-1.0091642-source.json
Full Text

Full Text

HOLDING THE POLICE TO ACCOUNT IN A DIVIDED SOCIETY: THE NORTHERN IRELAND REFORMS by C L A R E JOANNA M C G O V E R N B.A. Hons., The University of Oxford, 2000. A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR T H E DEGREE OF MASTER OF ARTS in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Political Science) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA August 2004 © Clare Joanna McGovern. 2004  T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A  F A C U L T Y OF G R A D U A T E S T U D I E S  Library Authorization  In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholariy purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Name of Author (please print)  Title of Thesis: SoCXErry  Date (dd/mm/yyyy)  ^ L O ^ A / ^ :  THE  T>/£  ^QUTZCE  / V ^ r n £ ^  TO  T-&^t-(\vO  Degree: Department of  Pt-CCoucJT  cAL  Pt  Qjr^PeO  RC^F-ORM^.  Year:  Po LxTl^  jr (v  J QQ 4  Stj:£<vCE  The University of British Columbia Vancouver, BC Canada  grad.ubc..ca/forms/?formlD=THS  p a g e 1 of 1  last updated: 20-M-04  ABSTRACT. This thesis uses a case study of Northern Ireland to examine the difficulties involved in holding the police to account in a divided society. Northern Ireland is a deeply-divided society where the primary political/religious cleavage is reinforced by the fact that the two communities tend to live in distinct areas and use separate educational, social, political, cultural and economic institutions. This has led to long-standing violence and instability, particularly over the last 35 years. Since 1993 there has been a peace process, which has led to an inter-party / inter-governmental agreement on the way forward, ceasefires from the main paramilitary groups, new political institutions and the devolution of government from Westminster (although the local assembly is currently suspended).  One element of the conflict has been the different attitudes of each community to the local police force (the Royal Ulster Constabulary or RUC). Unionists looked to the police for protection against an Irish Nationalist insurgency whereas Nationalists viewed the RUC as a tool of oppression. As part of the peace process an independent commission examined this issue and recommended radical change to the RUC's personnel, structure, training and operations. Police accountability was seen as particularly important - the commission made 35 recommendations in this area alone. Examination of these recommendations suggests four different approaches to accountability based on democratic, legalistic, technocratic and community based institutions respectively.  This thesis assesses these four models of accountability against three criteria, drawn from the literature concerning the management of ethnic conflict and public management. Firstly, the models are evaluated for their involvement of the public in evaluation and decision-making. The second criterion is whether accountability mechanisms can be insulated from the Unionist/Nationalist conflict, or alternatively, whether they can contribute to the management of this conflict. Finally, the type of competence promoted by each model is considered -whether the police are encouraged to focus on fiscal concerns, political responsiveness, compliance with rules or longterm strategic issues.  This analysis suggests that any one model of accountability operating alone has serious flaws. However, collectively, the four models meet all three criteria. This leads to the conclusion that a multi-stranded approach is required if Northern Ireland is to successfully hold its reformed police service to account.  iii  T A B L E OF CONTENTS  Abstract  ii  Table of Contents  iv  List of Figures  v  Introduction  1  Chapter 1: Theoretical Appraisals of Accountability  9  Chapter 2: Police Accountability: Historic and Comparative Models  32  Chapter 3: Fitting Accountability Models to Northern Ireland: Elite Preferences . . .47 Chapter 4: Fitting Accountability Models to Northern Ireland: Public Preferences . 60 Chapter 5: Police Accountability in Practice: Two Case Studies  71  Conclusions  88  Bibliography  96  iv  LIST OF FIGURES  Figure 1: Classification of the accountability mechanisms overseeing the P S N I . . . . 6 Figure 2: The interaction of the four competencies  19  Figure 3: The spectrum of political parties in Northern Ireland  48  Figure 4: Who should look into complaints about police behaviour?  61  Figure 5: Who should consider general changes in the way the police do their job? 64 Figure 6: The relationship between demographic factors and cynicism about  67  Community Police Liaison Committees. Figure 7: Have you heard of the Police Ombudsman for Northern Ireland?  73  Figure 8: Complaints received by the Ombudsman  77  Figure 9: Issues raised in Northern Ireland Policing Board Meetings  85  Figure 10: The strengths of the accountability models: assessing the evidence . . . . 91  INTRODUCTION. The debate over the role of the police ultimately rests on one's concept of the state. Policing is a particularly important state function, partly because police officers have coercive powers (including the power to question, to search, to detain, and ultimately to use force). However, whilst government entities such as tax collection agencies and the armed forces have similar powers, the police are distinctive because they combine these powers with daily contact with the public.  There are two conflicting perspectives on the nature of the state, which entail two very different views of the role of the police. The state can be conceptualized as a set of organisations which protect the prevailing regime from instability. The police are such an organisation, maintaining law and order in order to protect the regime. Although this function may also benefit the wider community, their consent to the police's role is not necessary. Alternatively, the state can be viewed as a contract with citizens whereby the state establishes institutions which protect their welfare. The police's role is therefore to serve the community rather than the regime of the day, and their legitimacy is derived from widespread support for their activities . 1  This debate is particularly relevant to divided societies , where there are divisions over the 2  legitimacy of the state and disagreement over the form that it should take in the future. Two  These models of the state are found in the writings of Machiavelli and Rousseau respectively (Caygill, 2001,76). Defined by Arend Lijphart as having cleavages of a religious, ideological, cultural or ethnic nature which are reinforced by social and political institutions (such as political parties, schools and voluntary associations) as these are organized along the lines of the cleavages. (1977, 3-4). Eric Nordlinger states that these segments become "conflict groups" when a significant number believe that their segment's identity and interests conflict with those of another segment (1972,7). 1  1  of the models posited to manage such conflicts correspond closely with these divergent 3  views of the state and the police. Under the control model described by Ian Lustick, the police help the state maintain stability through the control of potentially disruptive ethnic groups. The consociational model advocated by Arend Lijphart and Eric Nordlinger involves all groups in state operations. Cross-sectional support for the police will support consociationalism in two ways. Firstly, the police w i l l not contribute to the conflict themselves by using their powers disproportionately against one group (or being perceived as doing so), and secondly, they can contain inter-group conflict which might otherwise disrupt power- sharing arrangements.  This thesis w i l l focus on the role of policing in Northern Ireland and its contribution both to the conflict itself and to the management of conflict. I w i l l argue that historically the police in Northern Ireland have seen their role as the protection of the incumbent regime. This has caused acute problems because this regime is not accepted by all groups. The reforms which will be analysed in this thesis seek to shift the police's role so that they serve the community as a whole rather than protect the state from parts of the community.  Northern Ireland was formed in 1922 when it was partitioned from the rest of Ireland, which became independent from the U K . A t partition, Catholics formed 3 5 % of the population, although this has since risen to 44% . The Protestants form the majority group, (53% of the 4  population in 2001). It is a deeply-divided society where the religious/political cleavage is Lustick makes the point that there are other theories of conflict management (such as pluralism, theories of mass society and assimilationist models), but these seek to reduce conflict by diminishing cleavages. The control and consociationalist models are the only theories which explain stability in a society which nevertheless remains divided (1979,327). Latest figures takenfromthe 2001 UK census, cited in the Northern Ireland Policing Board Annual Report 2002-3 pl2. Available on-line at: http://www.nipolicingboard.org.uk/publications 3  4  2  reinforced by separate educational, social and economic institutions, and mutually incompatible views of the province's future. The society suffers from a "double minority" problem, as the minority group within the province forms the majority in the neighbouring Republic of Ireland. The in-province majority fears being subsumed into the Republic, particularly as this is the long-term objective of the two main Catiiolic political parties. The response to Irish Nationalism is Unionism, which emphasises the Protestant heritage of Northern Ireland and seeks to preserve its union with the United Kingdom.  As a result, there has been political tension and periodic violence in Ireland, and latterly Northern Ireland, dating back to the late 16 century (when Elizabeth I first sought to th  colonise Ireland). Over 3,000 people have been killed since the beginning of the current "Troubles" in 1969 (Farren and Mulvihill, 2000,201). Since 1993 there has been a peace process involving the Nationalists, Unionists and the British and Irish governments. This culminated in the 1998 Good Friday Agreement ( G F A ) and devolution of power to the 5  Northern Ireland Assembly (NIA) in 2000. However, the Assembly was suspended in October 2002, following allegations that Sinn Fein, the Nationalist political party affiliated with the Irish Republican Army (IRA), was using the Assembly to gather intelligence on Unionist politicians. New elections for the NIA were held in 2003 and the local parties are continuing to negotiate with the British and Irish governments over the restoration of devolution . 5  For a detailed account of the conflict in Northern Ireland and the peace process, see Farren, Sean and Mulvihill, Robert F. Paths to a Settlement in Northern Ireland. Gerrards Cross: Colin Smythe, 2000. B B C Online. Autumn talks "very important". 7 July 2004. 5  6  3  During the negotiations leading up to the GFA, policing proved to be a particularly intractable issue. The Nationalist/Catholic parties argued that the Royal Ulster Constabulary (RUC) was overwhelmingly Protestant and that it served that community's interests, while oppressing Catholics. They concluded that either the abolition or the reform of the RUC was crucial to a political settlement. Conversely, the Unionist parties portrayed the RUC as a victim of The Troubles rather than a protagonist, stating that it was a neutral arbiter which had been caught up in intercommunal conflict. This impasse was resolved by setting up an independent commission to review policing (McGarry and O'Leary, 1999, 4). 7  The Commission's report was published in 1999. The recommended reforms focused on symbolic/cultural change (emphasizing neutrality between Catholics and Protestants), human rights, community policing and the 50:50 recruitment of Catholics and non-Catholics to make the force more representative. A 10 year change plan began in November 2001, when the RUC's name was changed to the Police Service of Northern Ireland (PSNI).  However, as Mary O'Rawe and Linda Moore argue, it is exceedingly difficult to alter an organisation's culture and its members' behaviour; oversight mechanisms are needed if reforms are to be maintained over the long term. Such oversight should ensure that coercive powers are not abused, but also give the community some choice in the way in which it is policed (1997,107 and 134). This second function of oversight fits with the consociational notion that state functions should be conducted for the benefit of all groups in society.  The Commission is commonly referred to as the Patten Commission, as it was chaired by Chris Patten, former Cabinet Minister in the U K Government and former Governor of Hong Kong. The Commission also included criminologists, former police officers, corporate executives, former civil servants and a barrister. (Patten Report, 1999, para 2.1.) 7  4  The Patten Commission agreed that police accountability was crucial, and accordingly 35 of its recommendations either called for new accountability mechanisms or endorsed existing structures. The Oversight Commissioner (see Figure 1, page 6) has stated that these mechanisms will scrutinize the PSNI "more closely than any other police service among the world's developed democracies" (1 report, 2000, 7). st  The Patten recommendations are based on four distinct models of accountability. The police will be accountable to three groups of people - local or national politicians (the democratic model), community groups and the public in general (the community model) and experts in a particular field who are independent of the police and political parties. The "expert" model can be further sub-divided according to the issues examined. Under the technocratic model, experts hold the police to account for overall performance and policy, whereas under the legalistic model they are concerned with police behaviour in particular cases. The courts are an obvious example of the latter. This model also includes other organisations which do not have the power to impose criminal sanctions, but which share some of the law's traits - the measurement of conduct against pre-ordained rules, the examination of evidence to establish the facts of a case and a final decision made by an impartial arbiter.  Under the democratic and community models, the police may be held to account for both general performance and individual incidents (for example an MP or a community group may be interested in both clear-up rates and a death in police custody). Figure 1, over the page, shows how the various mechanisms endorsed or recommended by the Patten Commission fit into each of these four models.  5  Figure 1: Classification of accountability mechanisms overseeing the PSNI  Accountable to politicians  Accountable for general policy and performance  Accountable for individual incidents  Democratic model  Democratic model  Secretary of State for Northern Ireland (responsible to the Westminster parliament)  Secretary of State for Northern Ireland  Northern Ireland Affairs Select Committee (Westminster parliament)  Northern Ireland Affairs Select Committee  Northern Ireland Executive, responsible to the Northern Ireland Assembly (when the NIA is re-instated and policing powers are devolved)  Northern Ireland Executive  Policing Board (includes members of the NIA)  Policing Board  Accountable to community groups  Community model  Community model  District Policing Partnerships (at district council level)  District Policing Partnerships  Accountable to independent experts  Technocratic model  Legalistic model  Northern Ireland Audit Office (audit accounts and makes value for money assessments)  Police Ombudsman (investigates complaints against the police from members of the public)  Inspector General of the Constabulary (assesses policing strategies, techniques and performance) Oversight Commissioner (monitors implementation of the Patten reforms)  Commissioner for Covert Policing (issues warrants for wire-tapping) Legal action brought by individuals under the Human Rights Act, 1998. Criminal prosecutions of police officers by the Director of Public Prosecutions  6  Such a multiplicity of accountability structures raises several questions. Firstly, the GFA did not make the conflict between Catholics and Protestants disappear; there is still deep disagreement over the future of the state. Instead it sought to provide local political forums in which the conflict could take place, as a substitute for street-level violence. This ongoing constitutional debate may well affect preferences over policing. Furthermore, the fact that politicians could not agree on policing during the G F A negotiations demonstrates that there will continue to be different perceptions of the police and divisions over the type of policing which is best for Northern Ireland. If there is no generally agreed vision of what the police should be doing, how can they be held to account for it?  Secondly, it has been argued that too much oversight can stifle the creativity of an agency's staff (Behn, 2000,13) and lead to an emphasis on rule compliance to the detriment of effectiveness (Light, 1993, 230). Thirdly, there is a tension between the pressure for oversight, and the desire to decentralize authority. Monitoring police behaviour suggests a distrust which belies the responsibility given to front-line officers in community policing models. This ambivalence about front-line officers is mirrored in the role of the general public in policing. Three of the four models appear to make the police accountable to ehtes (albeit politicised and non-politicised elites), whereas the fourth model sees the public as being the best equipped to judge police performance.  These questions suggest three criteria, which will be used to assess the four models: i)  does the accountability model involve the public or is it elite based?  8  1 use the term "elite" to include non-political "expert' groups who are may not come from Northern Ireland and who are not particularly attached to one regime or another. 8  7  ii) Is the accountability model insulated from the Nationalist/Unionist conflict, or will it be hampered by the conflict , or alternatively, can it help to resolve the conflict as far as it 9  pertains to policing? iii) what kind of incentives does the accountability model give the police - does it promote rules compliance or encourage strategic thinking?  Chapter 1 will address these questions by applying theories from the ethnic conflict and public management literature to the four accountability models. Chapter 2 will review the history of police accountability in Northern Ireland and consider how other divided societies have dealt with these issues. Chapters 3 and 4 will situate the debate in contemporary Northern Ireland, by considering both elite and non-elite preferences for different kinds of accountability. Chapter 5 will then assess how these accountability models have worked in practice through two case-studies; examining the Policing Board (an example of democratic accountability), and the Police Ombudsman (legalistic accountability).  Ultimately, this thesis will argue that none of the models on their own can satisfy all three criteria. However, collectively, they ensure both elite and non-elite participation, provide information on policing which is insulated from the political conflict, whilst also seeking to resolve that conflict, and encourage a balanced set of competences in the police service. I will therefore conclude that although the Patten reforms are at times contradictory, together they form a coherent package which reflects the complexities of policing in Northern Ireland.  The conflict may hamper accountability in two ways. The conflict may be so severe that the institution holding the police to account is suspended, not established in the first place or has its membership changed. Alternatively, one of the groups represented in the institution may try to use accountability procedures to protect their preferred regime. 9  8  CHAPTER 1: THEORETICAL APPRAISALS OF ACCOUNTABILITY.  This chapter will use the ethnic conflict and public management literature to appraise the four different models of accountability found in the Patten Report. Five theories in particular will be used: consociationalism and the control model from the ethnic conflict literature, and bureaucratic autonomy, policy competence and the administrative-policy divide from the public management literature. After outlining these concepts, I will employ them to assess whether the accountability models are likely to meet the three criteria posited in the introductory chapter (insulation from the conflict, public involvement and balanced incentives to the police service).  1.1  Theoretical concepts Consociationalism  1.1.1  Lijphart defines a consociational system as one in which deep cleavages exist, but where the elites of each segment co-operate with each other to avoid conflict. The four elements of a consociational democracy are a grand coalition of the elites, which governs the country, a veto for each segment, a strictly proportional allocation of resources and power among the segments, and autonomy for each segment to organize its own affairs (1977, 7). The work of Nordlinger can be used to supplement this structure, as he describes the conflict regulation practices which are used by elites in such societies. These include compromises, concessions and purposive de-politicisation, where elites agree not to use sensitive issues to 10  score political points, in case they inflame inter-group hostility. (1972, 27).  However, he warns that concessions are only effective when made by a group which is "markedly stronger than the others (1972,27).  9  Both Nordlinger and Lijphart note that a tradition of accommodation between elites will greatly aid an attempt at consociationalism (1972, 55 and 1981,11, respectively). However, in the absence of such conditions, Nordlinger argues that there are several motivations which will nevertheless induce elites to co-operate. These include the risk that conflict will damage the economic interests of one group, or the whole society, die chance to acquire or retain political power and a desire to avoid violence and instability (1972, 48-51) . 11  It follows from the principle of proportionality that police officers would come from all segments of society and that policing would be for the benefit of the entire community as opposed to the dominant group. As a sensitive function of the executive, policing would be "purposively depoliticized" - controversial issues such as policing techniques and the ethnic profile of offenders would not be subject to public debate. Policing which serves the entire community not only removes a source of resentment, it can also make a positive contribution to consociationalism. As Joanne Wright and Keith Bryett note, it is easier for elites to accommodate each other when attitudes have not been polarized by street level violence (2000, 92). Furthermore, the need to balance order with space for social change and respect for individual rights is magnified in polarized societies where different sections of the community may have fundamentally different plans for the future (ibid, 114). For consociationahsm to be successful, the police need to be able to contain inter-group violence without stifling political debate and protest.  These motivations are particularly relevant to Northern Ireland, as Lijphart (writing in 1977) expressed little hope that consociationalism could be successful there, because its political culture was used to an "uncompromisingly adversarial style of governance" (1977,136). Whilst this was a reference to the local Unionist government of 1922-1972 (see Chapter 2 for more details), it was replaced by direct rule from Westminster. Lijphart has more recently described the U K system as "an exemplar" of majoritarianism (1999,247) and hence both unused and unsuited to coalition government or political compromise. 11  10  1.1.2  Control  model  Lustick describes an alternative method for stabilizing divided societies. Under the control model the dominant group seeks to manipulate the subordinate group in order to prevent it from becoming politicized and seeking autonomy or independence (which would destabilize the state). The elite of the dominant group equates society's interest with that of its own group. The resources of the state are therefore not allocated proportionately between groups; rather they are deployed for the benefit of the dominant group (1979, 330-331). When the control model operates successfully, the elite of the subordinate group (if one exists) is not in a position to bargain over resources, and can merely react to the agenda set by the dominant group. Farren and Mulvihall note that such strategies are deployed by groups which are in the majority but nevertheless feel insecure (one reason for this is that they are in a double minority situation). These groups develop a "siege mentality" as they feel they are under constant threat of attack. Domination is seen as the only alternative to humiliation at best, and extinction at worst (2000,109 and 111).  Policing is therefore seen as a state resource which should be used to protect the dominant group. Its role is to prevent dissatisfaction crystallizing into an identifiable group of dissidents which could threaten the dominance of the state (Caygill 2001, 76). Part of its job is to prevent symbolic protests which could strengthen the identity of the excluded group and lead to demands for autonomy (Smyth 2002, 302). Ronald Weitzer describes a "divided society model of policing" which essentially applies Lustick's control model to policing. The police force itself is politicized, and strongly identifies with the current regime. It therefore polices the regime's opponents more aggressively than members of the dominant group (1995, 5).  11  However, Lustick emphasises that the control model is not meant to be violent - indeed it has failed if the subordinate group becomes violent or if police brutality occurs, as both can destabilize society. The very existence of regime opponents indicates that control is already breaking down as the subordinate group has become politicized. This point is reinforced by Ralph Crawshaw, a former Chief Constable in the UK: "Those who are subjected to (police abuse) . . .become radicalized and more prepared to join or support violent, subversive opposition groups . . .such groups benefit hugely from the propaganda value of serious lapses of behaviour by state security forces "  (cited by O'Rawe and Moore, 1997, 97)  However, even once it is failing it can be dangerous to replace a control model of policing with another form of conflict management. Reforming the police can be interpreted as reducing the "security bulwark" of the dominant group (Smyth 2002, 305), which may result in a violent backlash, or demands for a quid pro quo. Alternatively, such reforms may be interpreted as a concession made by a group which is no longer sure that it is stronger than other groups. As Nordlinger notes, this may simply convince the recipient group that all-out "victory" is possible, thus increasing their intransigence and prolonging the conflict (1972, 29).  1.1.3  Bureaucratic  autonomy  Daniel Carpenter uses this term to describe government agencies which follow their own policy preferences, as opposed to those of politicians. The difficulties in holding such agencies to account need to be considered when assessing how effective oversight mechanisms are likely to be. Carpenter argues that agencies achieve autonomy by establishing political legitimacy, which has two elements. Firstly, autonomous bureaucracies will have a  12  reputation for being able to provide services to society which are not found elsewhere (reputational uniqueness). Secondly they have support from a diverse coalition which are "irreducible to lines of party, class or parochial interest" (political multiplicity)(2001, 5).  It should be noted that political multiplicity is the antithesis of the control model, as the agency courts a wide range of social groups rather than the dominant segment alone. The Patten Report endorsed the principle that the police should build links with as wide a group as possible, stating that the aim of their reforms is a police service that is "capable  of attracting  and sustaining  support from the community  as a whole " (Section 9, para 1)  Furthermore, the concept of reputational uniqueness echoes Robert Reiner's argument that policing by consent is possible when the police perform a particular role - that of neutral arbiters who manage conflict in the interests of society as a whole (cited in O'Rawe and Moore, 1997,12). This echoes consociationalism, where state resources are used to benefit all groups, thereby maintaining stability.  As outlined so far, bureaucratic autonomy appears to be compatible with consociational methods of conflict management, rather than the control model. However, Carpenter developed the concept to explain agency behaviour in the US, arguing that broad coalitions are necessary in systems with a separation of powers (1997, 15). It may be that in a majoritarian, centralised system, such as Northern Ireland before the GFA, an agency does not need to build a broad coalition in support of its programme. In such situations (or indeed, under a control model), a strategic alliance with the only power base that exists may be more effective.  13  There are three distinct periods in Northern Ireland's history. The first was the Stormont government of 1922-1972, where the power base was the local Unionist community (the dominant group in a control model). The second was the period of direct rule where the power base was the U K government. The third period, of devolution under the GFA, was the only time in which power was shared, in a coalition of four political parties. This created a situation where there were multiple power bases, which is more analogous to the system described by Carpenter . 12  1.1.4  Policy competence  Colin Campbell describes three distinct types of competence which a government agency can display - fiscal, responsive and strategic. These are often in tension with one another; achieving a sustainable balance between them is termed policy competence (2001, 260-261). Fiscal competence is concerned with the inputs into government programmes (in the case of the police, this would include the recruitment and training of police officers, intelligence systems, police vehicles and stations). Campbell notes that fiscal competence is often equated with the need for economy, i.e. a short-term concern with minimizing inputs in order to reduce government expenditure (ibid, 260).  Responsive competence concerns an agency's ability to provide medium-term outputs which are valuable to important stakeholders (either in their own right, or because they will be publicly popular and will help politicians to get re-elected). For the police, this might be achieving a target for drug seizures, which is measurable and a simple concept to convey in a However, an obvious difference remains, namely that the political parties in Northern Ireland had explicitly agreed to share power under the Good Friday Agreement whereas a Republican President who shares power with a Democrat-controlled Congress, or vice versa, has not. 12  14  political speech or manifesto. Responsive competence is therefore similar to Carpenter's explanation for bureaucratic autonomy - agencies gain legitimacy if they provide valued 13  services (outputs) to political actors (although in Campbell's formulation, these actors do not need to be part of a broad coalition).  Strategic competence is a concern for long-term outcomes which will benefit society as a whole. It involves planning for a changing environment and ensuring that the agency is able to adapt to these changes and remain relevant (ibid, 261). The Patten report is a good example of strategic thinking as it assesses how policing can best serve the whole community and how it will have to adapt to the changes in Northern Irish society brought about by the peace process.  The concept of policy competence is developed further by Michael Barzelay and Colin Campbell, who describe strategic competence as a question of "positioning for opportunities." An agency which wishes to achieve its long-term vision must ensure it is well-placed to take advantage of opportunities to demonstrate its abilities. In doing so it increases its reputation and building a political coalition behind it's overall strategy (2003,234). The three competencies thus become complementary; responsiveness helps build support for the agency's strategy, and fiscal competence becomes less a matter of economy and more a matter of efficiency - using resources wisely so that the agency can deliver outputs and ultimately fulfill its long-term goals. The term "legitimacy" may be used in two senses. Carpenter's discussion of bureaucratic autonomy suggests a legitimacy that has been earned through the agency's work. Legitimacy may also be derived from the constitutional status or legal standing of an agency. For example, the R U C were granted legitimate status by the various Police (Northern Ireland) Acts, which gave them legal powers of arrest, stop and search etc. However, they did not earn legitimacy as there was no public consensus over their role. 13  15  However, in the case of the police, a fourth type of competence is also important. If the rule of law is to be respected it is important that law enforcement agencies are also seen to comply with the law. Given their coercive powers, a respect for human rights is particularly crucial.  14  However, the need for compliance with the law covers many other facets of police activity from January 2005 their release of information to the public will have to comply with the Freedom of Information Act, the storage of intelligence must conform to the Data Protection Act, the working conditions of police officers must meet the requirements set out in Health and Safety regulations; the presentation of their accounts must be in accordance with Government Accounting Standards. I will term this fourth competence "regulatory competence" - the ability to comply with laws and regulations which in the case of the police particularly concern human rights standards.  15  Such requirements do not fit easily with either fiscal or responsive competence. Compliance may be expensive and may divert resources from the investigation of crime so it is not necessarily fiscally competent. Ian Loveland has detailed some of the public and political reactions to human rights judgements , which suggests that a concern for such issues is not 16  always responsively competent, particularly when the individual whose rights are being protected is suspected of criminal or terrorist activity.  This was recognised by the Patten Commission who devoted the first section of their report to the issue. Human rights and fundamental freedoms (as defined in the European Convention of Human Rights) became enforceable in U K law in 2000, when the Human Rights Act 1998 came into force. The Act applies to all public institutions in Northern Ireland, including the PSNI. H R A 1998. s. 6(1). Loveland, Ian. "The War Against the Judges". Political Quarterly. 68:2 (March-June 1997), pp 162-170. 14  15  16  16  The relationship between regulatory and strategic competence is more complex. For society as a w h o l e , the t w o are complementary. Respect f o r the rule o f l a w that ensures people are treated f a i r l y and that they can predict the consequences o f their actions, enabling them to plan for the future. The protection o f h u m a n rights respects human dignity and enables m i n o r i t y groups to both participate i n society and preserve their o w n culture, m a k i n g it less l i k e l y that they w o u l d challenge the status quo. B o t h o f these factors therefore increase the stability o f society. I t f o l l o w s that the government w o u l d v i e w regulatory competence as part o f strategic competence - i t improves the quality o f l i f e for its citizens and aids the l o n g - t e r m survival o f the state.  H o w e v e r , individual government agencies may f i n d that regulatory competence sometimes hinders their ability to p e r f o r m specific functions and meet particular social needs. I n the case o f p o l i c i n g , compliance w i t h laws and regulations m a y at times prejudice a c r i m i n a l investigation (for example b y restricting methods o f interrogation or evidence gathering), so i t m a y not always further the strategic agenda o f crime prevention. Regulatory competence is nevertheless important i f the police are to earn legitimacy - i t is d i f f i c u l t f o r the public to trust the police to enforce the l a w i f they do not obey it themselves. This f o u r t h type o f competence therefore forms a context i n w h i c h the police have to operate.  Campbell conceptualises competence as a triangle; fiscal, responsive and strategic competence are at the three corners, and the space i n the middle o f the triangle represents p o l i c y competence, where fiscal, responsive and strategic concerns are balanced. W h e n assessing the performance o f an entire government, regulatory competence w o u l d f i t into the  17  "strategic competence" corner. However, in the case of an agency such as the police, regulatory competence is potentially in tension with each of the other competencies. Figure 2 shows how I would modify Campbell's representation of policy competence in order to apply it to the police. The requirement to comply with the law is non-negotiable (or 17  should be, if the rule of law exists), whereas the otiier three competences may be forgotten in i8  some circumstances.  Furthermore, the sanctions for regulatory /«competence if caught are  severe and certain (i.e. criminal conviction, civil liability or dismissal). Being profligate with resources, unresponsive or lacking a long-term vision may lose a government the election or an agency head his or her job, but this is dependent on many other factors, such as the extent of publicity given to the incompetence, personal ties and the existence of a viable alternative government. For these reasons, I would argue that an agency such as the PSNI has to first comply with the law and regulations. The degree of discretion left is the arena in which the other three competences must compete - regulatory competence therefore forms the boundaries of the triangle. 1.1.5  The Policy-Adm in istration Divide.  In the early 20 century public management scholars attempted to distinguish policy from th  administration. Policy was the value based choices involved in allocating resources and deciding the long term goals of an agency and was the preserve of politicians. Administration was the economic and efficient implementation of these decisions and was the role of the career public servant (which would include police officers). A similar rationale was behind  The Accountable Juggler. Beryl A. Radin, CQ Press, 2002, p 18. Campbell gives the example of the 1996-1998 government of John Howard in Australia, where a supposed budgetary "black hole" and the need to keep campaign promises led to a focus on fiscal and responsive competence to the exclusion of strategic competence (2001,271-272). 17  18  18  Figure 2: The Interaction of the Four Competencies  Regulatory competence\^  Regulatory competence  Responsive competence  POLICY COMPETENCE  Fiscal competence  Strategic competence  Regulatory competence  Source (modified from the original): Campbell, Colin. "Juggling Inputs, Outputs and Outcomes in the Search for Policy Competence: Recent Experience in Australia" Governance: An International Journal of Policy and Administration. Vol. 14, No. 2, April 2001, p 260.  some of the New Public Management (NPM) reforms implemented in western democracies the 1980s and 1990s. Barzelay (citing Aucoin) states that a central claim of N P M argued the functions of policy-making and operations should be assigned to different organisations, enabling managers to concentrate on the efficient implementation of policy, rather than deciding the policy itself (2001, 8). It followed that most of NPM's attempts to increase the effectiveness of government werefocused on reforms to the bureaucracy (the implementers of policy), rather than the political systems in which they operated.  19  However, the validity of the policy-administrative divide has been disputed by many public management theorists. Graham Allison argues that how resources are distributed and policy is implemented have a substantial effect on who gets what resources, or which policies are implemented (1979,385). Joel Aberbach and Bert Rockman have warned against reforms which focus only on the bureaucracy, as the political environment in which bureaucrats operate may harm government as much as inefficient administrative structures (2000,176177). This debate echoes the different attitudes towards police reform in Northern Ireland. The Nationalist perspective stated that there were things about the RUC itself which undermined its legitimacy , which could be addressed through internal reforms of its structure, personnel 19  and practices. Unionism focused on the political environment (i.e. the Nationalist/Unionist conflict) as being the cause of the RUC's problems, and advocated "political" solutions; mainly that Nationalist leaders should accept and support the police (McGarry and O'Leary, 1999, 9-10) . 20  1.2  Appraising the accountability models  1.2.1 Insulation from the conflict Nordlinger argues that majoritarian democracy is not suited to regulating severe conflicts as it does not take account of all segments in society (1972, 33). It follows that police accountability which is purely to a ministerfroma one-party government will not be insulated  "Legitimacy" is used here in the "earned" sense (see footnote 13 on page 15), meaning that there was not a public consensus that the RUC provided a useful service. The continuation of vigilantism by paramilitaries suggests that the police still do not have "reputational uniqueness" - a reputation for addressing social problems (i.e. crime and anti-social behaviour) which cannot be solved by anyone else (Carpenter, 2001, 5). However, as will be discussed in Chapter 2, Unionist and RUC relations became strained in the 1980s and 1990s. 19  2 0  20  from the conflict. Instead, it may well exacerbate divisions as the minority group will be aggrieved that it has no say in the quality and type of policing to which it is subject. Geoffrey Marshall concurs, arguing that political accountability is "explanatory and cooperative" politicians and police officers engage in a dialogue about the allocation of resources and policing priorities. He states that in a divided society this debate needs to be as inclusive as possible, to demonstrate that minorities are not simply subject to policing; rather, they are citizens who are served by the police (cited by McGarry and O'Leary, 1999, 98).  These arguments suggest that only a consociational form of democratic accountability is appropriate in a divided society. Such a model will not be insulated from the divisions in society; indeed it will reflect those divisions. However, it is designed to manage conflict by creating a forum in which each group ensures that the police are serving its needs and where debates over conflicting needs can take place.  When assessing community models in the Northern Irish context it is important to note that the Catholic and Protestant communities share many localities. Wright and Bryett argue that it would be impossible to set up police units which deal exclusively with one group (2000, 70). Community based groups which hold the police to account at this level should therefore include representatives from both groups. The conflict might directly hamper these mechanisms if the two communities could not work together, or if one group boycotted the groups altogether.  Nordlinger discusses whether inter-segmental contacts (below the elite level) are beneficial, as he believes that they may breed antagonism. However, he concedes that isolation can also  21  exacerbate hostility to other groups, whilst working relationships with members of other segments may breed tolerance and trust. He concludes that involuntary contact, due to living close to one another is more likely to be antagonistic than contact which is a deliberate 21  choice, for example by joining a community policing group (1972,106-108). This suggests that community based structures have the potential to alleviate conflict. As with a consociational democratic model, they seek to aggregate preferences for particular types of policing and turn them into workable policies which serve all segments of the local community.  In contrast, legalistic and technocratic models do not seek to resolve a conflict. Instead they seek to insulate the accountability process from political and social divisions. Difficult issues are depoliticized by applying either neutral pre-ordained standards (the legalistic model) or the opinion of an impartial expert (the technocratic model). This fits with Nordlinger's description of "purposive de-politicisation"- the delegation of controversial decisions to independent arbiters so that the elites of each segment do not have to take a position which could cause conflict . 22  However, the debate over the administrative-political split suggests problems with this argument. It may be difficult to successfully depoliticise such an emotive subject as policing, particularly given the potential for the abuse of police powers, and the constant interaction A good example of this in Northern Ireland is the stand-offs which occur when local Orange Lodge members parade through nearby Catholic areas. The Patten Commission was itself an attempt at purposive depoliticisation, as the political parties could not agree on policing during the G F A negotiations. The Oversight Commissioner (see Figure 1 on page 6) praised the Patten recommendations as having been "developed and tested against policing benchmarks rather than political criteria" (9 Report, December 2003, p2). However, as will be discussed in Chapter 3, this did not make the Patten recommendations immune from political criticism. 2 1  2 2  th  22  between the police and the public, often in stressful situations. John Alderson (the former Chief Constable of Devon and Cornwall Police) observes that:  "when social conditions are good, insensitive use of the police will result in complain through formal channels . . .when they are bad they will result in riot" (cited by O'Rawe and Moore, 1997, 56)  This suggests that the accountability structures will only be effective if police-community relations are already good, despite divisions. If this is the case, the conflict over policing has already been defused by another mechanism, such as a consociational democratic or community model. Weitzer agrees with this analysis, stating that the public (particularly minority groups) evaluate the police not so much on their own merits (as a technocratic or legalistic model would do) but also in terms of what they symbolize, as defenders of the state and therefore the constitutional status quo. It follows that the police will always be "political" and that it is misguided to analyse them as simply an administrative agency, in isolation from the conflict.  1.2.2. Public involvement  Lijphart states that a successful consociational system rests on "structured elite predominance", with non-elites taking a passive role and deferring to the judgement of their leaders. (1977,49). If segment leaders are to be able to successfully bargain with other groups they need to have the authority to make commitments without suffering a backlashfromtheir followers (Nordlinger, 1972, 73). Nordlinger goes on to argue that non-elites are not suited to the task of negotiation (as would occur when setting police policy) as they are too numerous, scattered andfragmentedto be able to aggregate preferences or process information efficiently (ibid, 40). Furthermore, they do not have the skills to evaluate policy alternatives, or assess long-term police performance - he argues that they have a shorter time perspective  23  than elites,  their cost -benefit analyses are less rational  and they tend to take the regulation  of violence for granted (ibid, 75). 25  It follows that a consociational democratic method of accountability would by definition be elitist. Lustick has criticized consociational systems on the grounds that they constrain public participation in decision-making and suppress dissent within segments. As a result, he argues that society may be less closed (at least for the non-elite of the dominant group) under a control model (1979, 334).  Bernard Rosen has questioned whether accountability which relies on the political elite is effective, arguing that for backbench MPs the function of oversight by comes a poor third to constituency matters and passing legislation (1998, 25). Furthermore, Colin Campbell and Graham Wilson cast doubt on whether even Ministers can hold public servants to account. They cite Max Weber who argued that a "political master" was essentially an outsider to the government bureaucracy , who did not necessarily know much about how it functions. He or 26  she is therefore no match for trained officials in career positions (such as the Chief Constable of a police force) who know the agency intimately and are experts in the field (1995, 251).  As will be detailed in Section 1.2.3, Campbell argues that political elites may only have a medium term, rather than a long term perspective. This is because they place more emphasis on symbolic issues than material benefits. Applied to the police, this would mean that the public would be more concerned about RUC/PSNI names and symbols (and which country they signalled allegiance to) than about crimerates.However, the political reaction to some of the Patten recommendations demonstrates that this is a criticism which could be equally leveled at the political elites in Northern Ireland (see Chapter 3). Although he accepts that this is more likely in a society where there has not been recent political violence. The mass demonstrations for peace which occurred periodically throughout the 1990s and the endorsement of the G F A in a referendum suggest that this last point does not apply to Northern Ireland. This would be exacerbated in Northern Ireland as democratic accountability for policing is "remote" - the police on the ground are answerable to an executive based in London. 2 4  2 5  2 6  24  This last criticism of democratic accountability suggests that technocratic accountability may be more effective. Rosen argues that determining the effects of policy and then pressing for any changes that are needed requires "considerable expertise, resources and persistence" (1998,27). Evaluation should therefore be a full-time job, performed by those who have knowledge and experience of policing (or particular skills in support areas such as financial management). However, from the point of view of the public, the technocratic model merely swaps the political elite for a depoliticised elite. Alasdair Roberts warns that the use of experts and an appeal to a technocratic body of knowledge suggests that only those with the necessary training can possibly form an opinion about public policy (1996, 65). O'Rawe and Moore point to the way in which police experience was used as a "trump card" in Northern Ireland to negate all other arguments, thus preventing the publicfromparticipating in the evaluation of a public service (1997, 151).  In contrast, legalistic models do involve the public as they concentrate on individual contacts with the police. They therefore reaffirm the importance on non-expert experiences and opinions. Paul Light details how such mechanisms rely on individual rights (to information, or to a legal hearing) to provide "sunshine" on the inner workings of a bureaucracy (1997, 3132). This can go some way to relieving the information gap between accountability holders and the agency being held to account. Furthermore, if alternative dispute resolution is used the complainant is involved in the entire accountability process, rather than simply initiating it (O'Rawe and Moore, 1997,125).  25  H o w e v e r , although legalistic models i n v o l v e the public i n h o l d i n g the police to account f o r i n d i v i d u a l incidents, they are less good at i n v o l v i n g them i n the evaluation o f broad p o l i c y . 27  Rosen notes that to use a mechanism such as j u d i c i a l r e v i e w an i n d i v i d u a l has to have standing - they must have suffered some personal damage as the result o f police actions or omissions ( 1 9 9 8 , 2 9 ) . Such mechanisms can therefore not be used b y a citizen w h o i s concerned b y police methods o f restraint (for example) i f he or she has not personally experienced them. S i m i l a r l y , a court or complaints process is not the place to make decisions about the allocation o f resources or p o l i c i n g priorities - they do not therefore a l l o w citizens to make choices about the quality and type o f p o l i c i n g they w o u l d like ( O ' R a w e and M o o r e , 1997, 173).  H o w e v e r , c o m m u n i t y based models o f accountability are specifically designed to i n v o l v e the public i n such decisions. M a r t h a Feldman and A n n e K h a d e m i a n describe h o w such structures develop the decision-making capacity o f the p u b l i c , b y b u i l d i n g relationships between different demographic groups w h i c h a l l o w t h e m t o debate priorities and aggregate preferences (2004, 7-8 and 18). This can go some w a y to countering the argument that o n l y experts have the skills and knowledge to p e r f o r m p r o g r a m evaluation and make p o l i c y choices. O ' R a w e and M o o r e agree, arguing that dialogue between c o m m u n i t y groups and the police is needed before the police are held to account f o r their performance. This enables the public t o appreciate the context i n w h i c h the police are w o r k i n g and to make i n f o r m e d choices (1997, 134).  This would mean that the legalistic model is best suited to promoting regulatory competence. However, as will be discussed in section 1.2.3, particular types of legalistic mechanism may also support strategic competence.  26  Building such relationships can boost the bureaucratic autonomy of the police, thus undermining accountability to the political elite. Carpenter describes how autonomous agencies "end-run" the representative process. They bypass the political process and build networks directly with public groups. This can shift voter preferences and hence the choices of politicians (2001, 33). Similarly, Feldman and Khademian describe how including the public in policy formation creates political capital - politicians are less likely to oppose plans which have been agreed with a wide range of community groups and are more likely to provide adequate resources (2004, 31-32).  Community based accountability can therefore be praised for involving the public. However, a caveat applies. Rosen has warned that links between an agency and a particular community group may develop into "a cozy relationship with a special group of citizens", with the agency becoming accountable to them, but not to the public as a whole (1998, 16). Similarly, O'Rawe and Moore cite research in Lancashire, England that found community-police liaison groups did not do a good job of representing the unemployed, the homeless, black and working class young men and the victims of domestic violence despite the fact that these groups are more likely to have contact with the police and have need of their services (Jemphrey et al, 45, cited by O'Rawe and Moore, 1997,163). They therefore excluded the politically marginalised and socially disadvantaged who had not organized themselves into interest groups. This raises the risk that if community based accountability structures are not representative they may simply replicate the elitism of democratic accountability at a local level.  27  1.2.3 Policy  competence  By definition, legalistic accountability focuses on individual incidents. It is therefore most likely to focus on questions of law and assess compliance with rules, which constitutes regulatory competence. Rosen argues that courts (and other legalistic bodies) are effective in holding agencies to account in the particular circumstances that come before them. However, they will only see the evidence presented by a particular case and are therefore not designed to evaluate an agency's general performance, in fiscal or strategic terms (1998, 28). 28  Bozeman makes a similar argument, stating that legalistic models are handicapped as they are forced to assume that the issue in question (such as acceptable methods of covert policing) affects only two parties - the plaintiff and the defendant. This ignores the fact that matters of policy will affect multiple parties, and that an institutional plaintiff or defendant (such as a police service) is itself not monolithic in its interests or intentions (2000, 55).  However, O'Rawe and Moore make the point that legalistic models can be a valuable source of information which can be used by other accountability holders to enforce strategic or responsive competence. Complaints commissioners or the courts can conduct trend analysis of recurring problems with particular policing techniques (strategic competence) or involving particular sections of the community (an issue of responsiveness). These can then be used to change policies and procedures or re-train personnel, thus improving the general performance or responsiveness of the agency (1997,122).  As a depoliticized body, legal institutions would explicitly discount considerations of responsiveness.  28  Technocratic models take the opposite approach to legalistic models; employing their expertise to consider questions of general performance and policy. Depending on the area of expertise, this can encourage either fiscal competence (if the overseers are auditors) or strategic competence (if the overseers come from the same line of business as the agency). Campbell argues that the experience and knowledge of career officials encourages them to take a more long term view than politicians, thus suiting them to strategically planning out the agency's future and its contribution to society. This argument can be also applied to technocratic accountability holders who have spent their careers working in the field . 29  However, Behn states that judging the social value created by an agency (i.e. assessing its strategic competence) is by its very nature political - people disagree over what is valuable and who should benefit. By looking at these questions, technocrats begin to lose some of their objectivity, which was the reason they were chosen as accountability holders in the first place. (2001,13). For this reason, Light argues that accountability holders have incentives to focus on rule compliance (regulatory compliance) and economy and efficiency (fiscal competence):  "It is safer politically for monitors to remain focused on post-hoc audit where the croo easy to identify and the sanctions obvious " (Light, 1993, 234)  Light also notes that even when technocratic monitors themselves want to focus on strategy and capacity building, politicians encourage them to provide more examples of rule-breaking and waste (1997,97). This raises the question of which competences will be encouraged by  Although career officials' concern for the long-term is also due to the fact that they will be working at the agency in the future and will have to live with the consequences if the agency is unable to adapt. This is not an incentive that applies to technocratic overseers. 2 9  29  the democratic model. This is endorsed by Campbell's research in Australia which found that when decisions were made by politicians alone they focused on the trade-off between fiscal and responsive competence (i.e. cutting the deficit whilst fulfilling campaign promises in time for the next election). Long term strategic issues were left off the agenda (2001,279). Similarly, Lijphart admits that a lack of efficiency may be a particular problem in consociational democracies. Decision-making will be a cumbersome process because the entire span of preferences has to be accommodated (or at least appeased with a quid pro quo in another issue area). This is likely to make it difficult to decide on an agency's strategy and therefore to hold it to account against a commonly agreed standard. He argues that in a divided society this may be a worthwhile price for avoiding antagonism and conflict. (1977, 50).  However, Rosen notes that backbench politicians (as opposed to the executive) may well ask the "big questions". The first of these is whether government agencies are fulfilling the purpose of laws, which would indicate regulatory competence, plus responsiveness to the politicians who passed them. The second asks whether agencies are being run effectively and in the public interest (strategic competence) (1998, 30). He also notes that on behalf of their constituents backbenchers will scrutinize agency behaviour in individual cases, thus reinforcing regulatory competence.  Rosen believes that such individual cases are also the proper focus of community based accountability. This is because he does not believe that the average citizen has the "time, information or resources" to hold a public body to account, unless he or she is dealing with a straightforward situation experienced personally (1998, 27). However, in the case of the  30  police, good community relations can have a major impact on performance as local groups may well have access to information that the police need. Feldman and Khademian argue that local knowledge, "the mundane yet expert understanding of local conditions", is often discounted in public management. This is because it is not collected in a structured and systematic manner and is therefore not viewed as scientific (2004, 25). However, as Weitzer argues, such information can be crucial. The involvement of the community in policing encourages them to report crimes and co-operate as witnesses, thus improving police performance (1995,1). The Patten Report encouraged community policing for this reason, arguing that it had two advantages. Firstly, it would improve police understanding of the culture and needs of groups which were previously estranged, improving their responsiveness. Secondly, community groups facilitated a strategic problem-solving approach. Under this model, the police scan all the information they get from the community to identify patterns of crime and disorder or trends which could become problematic in the future. They then determine the causes of those patterns, and, with the help of the community, devise solutions. This information is used to secure resources (e.g. drugs education and rehabilitation programmes, if a high proportion of the area's crime is drug-related). The situation is then reassessed to evaluate the success of the solution (1999, paragraph 7.15). The focus is therefore on solving problems over the long term rather than reacting to day-to-day events, and is a good example of policy competence - the combination of strategic, responsive and fiscal competence.  31  CHAPTER 2: POLICE ACCOUNTABILITY: HISTORIC A N D COMPARATIVE MODELS  Having established the complexities of holding the police to account, particularly in divided societies, this chapter will consider how Northern Ireland has dealt with this question in the past, and how other societies with similar problems have approached the issue. The first section will outline the accountability mechanisms used in Northern Ireland during two distinct periods in the past - the 1922-1972 Stormont government and the period of direct rule, from 1972 to 2000. The system used in England and Wales will also be discussed, as 30  this was the model for the 1972 RUC reforms. The second section of this chapter will consider the approaches used in two other Western democracies which also have (or had) deep cleavages - The Netherlands and Spain.  2.1 U K approaches to police accountability 2.1.1 Northern Ireland, 1922-1972: The Stormont Government. The RUC was established after the partition of Ireland in 1922. It evolved from the Ulster Special Constabulary (USC), which largely drew its membership from loyalist paramilitary 31  groups. These groups had formed to oppose Home Rule for Ireland and to protect Protestants from IRA violence. A local, Protestant police force was seen as the only organisation that would be sufficiently motivated to protect Ulster (Wright and Bryett 2000,4-5). O'Rawe and Moore argue that the RUC inherited this function, as Unionists continued to fear those within the province and in the Republic who did not support the existence of Northern Ireland. A As noted earlier, direct rule was reinstated in 2002. However, 2002-2004 also saw the implementation of most of the accountability mechanisms recommended by Patten. It would therefore be misleading to analyse these years as part of the direct rule period; two of these recent changes will be assessed in Chapter 5. The term "loyalism" was coined to describe those who are loyal to the UK; it is generally used to denote hard-line unionism. 31  32  strong police force which suppressed these groups was therefore seen as a necessary part of nation building (1997,268). This situation corresponds closely to the control model, where the dominant group seeks to use state resources (including the police) to protect its own position.  The function of policing was devolved to the Northern Ireland government . The police were 32  accountable to the Minister for Home Affairs for matters of general policy. Individual complaints were dealt with internally and there was no independent body to review police performance. The courts did deal with allegations of police misconduct, but these were criminal rather than civil cases. This meant that the case was only initiated if the AttorneyGeneral decided there was enough evidence, which was collected by the RUC themselves. The individual complainant was thus unable to initiate legal proceedings on his or her own behalf (Weitzer, 1995, 51-53). Northern Ireland therefore eschewed the technocratic model and had only a weak legalistic model.  Accountability was instead centered on the local democratic model. The flaw was that democracy in Northern Ireland at the time was majoritarian. As Protestants formed the majority of the electorate, and as they voted in a block for Unionist parties (as did the Catholic community for Nationalist parties), Unionists always had a majority in the local parliament. The cabinet from 1922 to 1972 was exclusively Unionist and Protestant.  33  However, the group which had the worst relations with the police, and the most reason to be  Commonly known as the Stormont Government, as the local Parliament met at Stormont Castle. With the exception of one minister who served briefly in 1969 (McGarry and O'Leary, 1999, 29).  33  dissatisfied  with policing was Cathohc/Nationalist -they had very little political power and  thus limited recourse.  The reasons for forming the USC and then the RUC could also be interpreted as an early attempt at community policing. It relied on local recruits to maintain order and protect citizens, on the grounds that they would be motivated to do so by their personal ties to the community. However, this aim is distorted in a divided society if the police only form links with one part of the community (especially if they do so out of fear of the rest of the population) . 35  The Stormont Government is a good example of how Carpenter's concept of bureaucratic autonomy can operate in divided societies under the control model. The police developed a reputation (particularly during the 1920s) for being able to provide an important service protecting Protestants from a Nationalist insurgency. This enabled them to gain support from the politically dominant group. It was this, rather than "political multiplicity", that allowed them to operate under weak accountability structures for 50 years. The UK government did not intervene until violence reached an unacceptable level in 1969. This was partly due to Unionist resentment of outside interference; although they valued their union with the UK, this was under the condition of local autonomy. There was also a fear that London's  The partisan actions of the R U C during this period and their heavy-handed policing of civil unrest in the late 1960s is well documented - see Wright and Bryett pp 6-8, Weitzer pp 27-58, McGarry and O'Leary pp 25-43, Farren and Mulvihill, pp 42-44 and Baxter, pp 142-178. When the R U C was first formed, a third of the recruitment spots were reserved for Catholics, an attempt to avoid this problem. However, it was dropped by the Stormont Government in response to pressure from the Protestant electorate (Farrell, 1983,190-1) and the intimidation of Catholic recruits within their own community (Wright and Bryett, 2000, 6). J 4  35  34  intervention would be perceived by the Nationalist community as a colonial action, providing propaganda for the I R A (Farren and M u l v i h i l l , 2000, 35, and 45-46).  2.1.2 England and Wales: The Template for Northern Ireland. The street violence of 1969-1972 and the re-introduction of direct rule eventually prompted reforms to the R U C , based on die structures used in England and Wales . Police 36  accountability in the rest of the U K is based on a tri-partite structure, in which each police service is responsible for a particular region. The Chief Constable of a police service is responsible for day to day operations. A local Police Authority  37  agrees policy and long-term  plans with the police, which include measures of overall performance such as crime rates and clear-up rates. The Authority then holds the Chief Constable accountable for his or her delivery against the plan. The Home Secretary, a central government minister, sets the statutory framework for all police services in England and Wales (for example setting out police powers and liabilities). He or she also issues guidance on matters of national and international importance (such as terrorism or drug smuggling).  This structure seeks to balance local and remote democratic accountability. It is also somewhere in between the majoritarian and the consensus versions of democracy. A t the  The mechanisms used in Scotland at the time were similar, although its Police Authorities are entirely elected. (McGarry and O'Leary, 1999,100). Since 1999, however, policing has been devolved to the Scottish Parliament. (Scodand Act 1998. Schedule 5). Similarly, the Mayor of London and Greater London Assembly control the Metropolitan Police (Greater London Authority Act 1999. sections 310-313). For the rest of the UK, policing is a reserved function of the Westminster Parliament, (Government of Wales Act 1998. Schedule 2; there has been no legislation devolving central power to England (with the exception of London) so the status quo remains). The full text of all these statutes is available at: http://www.hmso.gov.uk/acts.htm Each Authority has 17 members; 9 (i.e. a bare majority) are councillors elected locally; 3 are local magistrates and 5 are chosen by the rest of the Authority from a shortlist drawn up by the Home Secretary. fPolice Act 1964. section 3 as amended by the Police and Magistrates' Courts Act 1994. section 3 and Schedule 2.) 36  37  35  national level, only the political party with a majority in the House of Commons will get any control over policing. However, the councillors on the Police Authority have to reflect the balance of political parties in the local council(s) . While this will give the advantage to the 38  largest party, smaller parties will still get a say in policing at the local level.  However, there has recently been a shift of power away from Police Authorities towards central government. David Blunkett, the current Home Secretary, has set policing priorities with which local policing plans must comply. He has also established the Police Standards Unit, under the aegis of the Home Office, which monitors the work of local crime units and which has the power to intervene in individual investigations (Easton, 2004). Finally, the High Court has recently held that the Home Secretary had the power to suspend or dismiss a Chief Constable for failings that cause national problems, despite the objections of the local Police Authority . 39  England and Wales also use the other three models to hold the police to account. Individuals who are mistreated by the police have two routes through which they can initiate legal action. If the police violate a human right which is protected under the European Convention of Human Rights the victim can seek damages under the Human Rights Act 1998. If victims believe an action or omission by the police was either ultra vires , unreasonable or 40  Police and Magistrates' Courts Act 1994. Schedule 2. section 4(1). R v Humberside Police Authority ex parte Secretary of State for the Home Department. The Chief Constable was suspended because his force had failed to pass on the details of an individual's criminal record to another police force. The individual concerned took a job as a school caretaker and went on to murder two children. The case appeared to be a classic example of an administrative error developing into a high profile incident which has the potential to cause political damage to the Minister concerned. The case has not yet been reported. See "Embattled police chief suspended" (BBC On-line, 2 July, 2004) for a news report of the judgement. The action was beyond the powers delegated to the police and was therefore unlawful. 3 8  3 9  nd  4 0  36  procedurally unfair they can apply for judicial review in the High Court, which can reverse the police decision (e.g. to close an investigation) or order damages if the claimant has suffered harm. (Treasury Solicitor (UK), 2000,4, 11 and 40).  41  The technocratic model can be seen in the Police Standards Unit, discussed above. Her Majesty's Inspectorate of the Constabulary (HMIC) is another technocratic body which examines the efficiency of a police force. They consider issues of fiscal and strategic competence and are staffed mainly by senior police officers (i.e. people who are considered 42  experts in law enforcement). These organisations report to the Home Secretary and to Police Authorities . Technocratic mechanisms therefore also strengthen democratic accountability 43  as they provide politicians with information which they can use to assess police performance.  Finally, there are weak examples of a community based model. After the inner-city riots of 1982, consultation committees were set up to give feedback to the police. As O'Rawe and Moore note, these committees only have advisory powers, in order to preserve the ability of the Chief Constable to take operational decisions independently (1997, 137). The Home Office is currently consulting on its proposals to increase the involvement of neighbourhood representatives in policing . It views the traditional notion of consent for policing (as 44  advocated by Reiner; see section 1.1.3) as giving the public too passive a role. Instead, it  A police officer may also be liable to criminal prosecution for their actions, but such decisions rest with the Director of Public Prosecutions (an officer of the U K government) rather than the individual affected. Two examples are the 2004 Thematic Inspection Reports on workplace modernisation and the effect of guns on communities. Available at: http://www.homeofrice.gov.Uk/hmic/pubs.htm#2004 HMIC, The Role of Her Majesty's Inspectorate of the Constabulary. H M S O , 2004. Available at: http://www. homeoffice.gov. uk/hmic/hmicrole. pdf Home Office. Policing: Building Safer Communities Together. (Consultation Document). HMSO, 2003. Available at: http://www.policereform.gov.uk/docs/policing_bsct.pdf 41  4 2  43  4 4  37  exhorts the police to obtain in-depth knowledge of local needs and wants communities to "actively support" law enforcement. In return, the public will have a say in the way their area is policed . However, these proposals are still at the formative stage. The philosophy is 45  community based, but specific mechanisms to involve the public in police accountability have not yet been proposed.  As Easton argues, this approach does not fit easily with the Home Secretary's attempts to increase his own powers to hold the police to account. This tendency to centralize control can be linked to the traditional Westminster notion of democratic accountability. In this model, a minister is responsible for all events falling within the purview of their department, which in the Home Secretary's case would include the failings of individual police forces. However, Campbell and Wilson describe how this form of accountability has weakened as a result of the political-administrative split endorsed by New Public Management. On several occasions Ministers have claimed problems are the result of operational decisions made by managers, and therefore outside their (political) remit (1995,278).  The decision in the Humberside Police Authority case (see footnote 39, p36) may reverse this trend - if the Home Secretary is able to discipline senior police officers for systemic errors it is more difficult to argue that administrative issues are not his concern. As Easton concludes, if the Home Secretary is held responsible for such errors, the tension between localized, community based mechanisms and "the government's need to retain control so that they can step in if things go wrong" can only increase (2004).  Citizen-Focused Policing (Briefing Note), Home Office, 2003. Available on-line at: htrp://vww.policereform.gov.uMmple^  4 5  38  2.1.3 Northern Ireland 1972-2000: Direct Rule from Westm inster The beginning of "The Troubles" in 1969-1972 led to the suspension of the local Stormont Parliament. Responsibility for all government functions, including policing reverted to the London government. The tri-partite structure used elsewhere in the U K was introduced. In the rest of the U K , the majority of Police Autiiority members are elected, but in Northern Ireland they were appointed by the Secretary of State for Northern Ireland. (Patten Report, 1999, paras 5.5-5.6)  This model explicitly moved from local democratic accountability to remote democratic accountability, in an effort to depoliticise policing. The rationale was that the British Government would be a neutral arbiter between the two communities in Northern Ireland and would be able to reverse the "Protestantization " of the police which had occurred under 46  local government. As the Patten Report notes, the tendency for central direction was exacerbated by the fact that the Northern Ireland Secretary is responsible for only one police force, the RUC/PSNI. This, coupled with the security situation, encouraged the Secretary of State to intervene in day to day policing decisions in a way which is not feasible for the Home Secretary vis-a-vis other police forces (Patten Report, 1999, para 5.6). This has reduced the 47  impact of the local Police Authority, further limiting the possibilities for local accountability.  In addition, Patten argues that the democratic link between the Northern Ireland Secretary and the Northern Irish population is tenuous as Northern Ireland has a separate party system from the rest of the U K . As a result, the Northern Ireland Secretary always belongs to a party which This is Wright and Bryett's term (2000,6). Although since the Patten Report was published, the Home Secretary has developed the capacity to intervene in investigations via the Police Standards Unit. See section 2.1.2, above.  4 6  4 7  39  does not stand for election in Northern Ireland. This model therefore does not fit with either a democratic or a community based concept of accountability. Its attempt to provide accountability through independent oversight is limited by the fact that it is Northern Ireland's union with the U K which goes to the heart of the conflict. It follows that the British Government is not viewed as a neutral arbiter by Nationalist parties.  The Northern Ireland Office itself shares these concerns and has stated that it wishes to devolve responsibility for policing. However, for this to occur, there has to be a political institution to devolve power to . This area of policing accountability is therefore an example 48  of how political crises (i.e. the suspension of the Northern Ireland Assembly) affect administrative matters such as accountability mechanisms. This supports the argument that government agencies cannot be reformed in isolation from the wider political system. McGarry and O'Leary concur, arguing that the 1972 reforms did not increase Nationalist support for the R U C . This was because they did not accompany a political settlement, and therefore did not address the underlying problem (1999, 35).  2.2 Police Accountability in Divided Societies  2.2.1 The Netherlands Until the 1960s, Dutch society was "pillarised" into 4 groups - Catholic, Protestant, Socialist and Liberal. However, political violence was avoided by a consociational agreement between the elites of these four "pillars". (Brants, 2001,30).  Northern Ireland Office. Policing in Northern Ireland: Current Arrangements. (Briefing note). Available online at http ://www. n i o. go v. uk/pdf7nipbcurrent. pdf.  40  Policing traditionally relied on a community based model of policing. Prior to 1993  there  were 148 municipal police forces (Brants, 2001,44) and one central police force. This 50  meant that the police were drawn from their own communities, fitting Lijphart's observation that in consociational systems each segment has the autonomy to organize its own affairs. As they were working with small areas, the police were able to develop close relationships with local groups, identifying the needs of the area and potential threats to law and order.  This structure fits with the New Public Management principle that public services will be more effective and more responsive to public needs if decision making is devolved to the front line. Wright and Bryett also believe that this arrangement explains the lack of conflict during pillarisation. The police were not identified with any one group to the exclusion of the others, and were thus able to defuse conflict (2000, pp 90-92). This reinforces the importance of legitimacy, supporting Carpenter's argument that in a society with many power bases a government agency has to build a broad coalition of support. However, Wright and Bryett also note that the police were greatly aided by the Dutch culture of consensus-building. Contrary to the concept of the political-administrative split, there appears to be a symbiotic relationship between policing and the wider political system. Consociational structures helped to build public support for government activities including policing. Conversely, a police service which was sensitive to the different needs of the four "pillars" helped to prevent violence which could have threatened power-sharing. "Getting the politicsright"(ibid, p90) is therefore as important as the structure, performance and behaviour of public agencies. The police force was restructured in 1993 as the religious/political cleavages in the Netherlands had become far less important, and it was felt such a degree of localization was not efficient when dealing with national/international problems such as organized crime. For a population of approximately 15 million; contrast this to 52 regional police forces in the UK, which has a population of nearly 60 million. 4 9  5 0  41  Although the Dutch police have not exacerbated the segmental divide in their country, they have not always had harmonious relations with all groups. During student riots in the 1980s there were allegations of police brutality, but no institutional mechanisms to deal with complaints. Law students established informal bureaux to mediate between complainants and the officers involved. During the 1993 reforms these were formalized to become municipal Police Complaints Commissions, supported by an Ombudsman at the national level. The Commissions seek to resolve individual grievances and ensure that the police have complied with human rights standards. However, they also consolidate the lessons drawn from each case into recommendations on effectiveness (O'Rawe and Moore, 1997,220 and 266). They therefore serve as an example of a legalistic model of accountability which involves the public and promotes both regulatory and strategic competence.  2.2.2 Spain  In Spain, the campaign for greater autonomy (or independence) for the Basque Country has been subject to political violence, and the Basques were a group which were particularly oppressed during the Franco dictatorship of 1936-1975 (Kohler, 1982,24). Under the Franco regime, the police were militarized and were responsible only to the Army Ministry. They lived in their own barracks, separated from the community and usually did not come from the region that they policed (Heywood, 1995, 66). Oversight was therefore not based in the local community, nor was it democratic (the Army Minister was not elected). The police were solely answerable to the Minister and there was no independent monitoring, nor a mechanism for individuals to make complaints or gain information about police performance.  42  In democratic Spain, the police report to the Minister of the Interior, who is elected. The model of accountability is therefore democratic - elected politicians are answerable to Parliament and the public for the performance of public organisations. In the mid 1990s, evidence emerged of widespread human rights abuses committed by the police in their "dirty war" of 1983-1987 against E T A , the Basque paramilitary group (Heywood, 1995, 68). Two police officers were jailed, but there were no ministerial resignations, despite the findings of the investigating judge that several cabinet ministers were aware at the time of the police actions.  Jimenez (writing in 1999) has argued that this case demonstrates the inadequacy of the democratic accountability model. He argues that Spain has not yet developed workable traditions of accountability. This, combined with the alienation of the Spanish public from the Basques (and in particular, their antipathy towards ETA) meant that there was no public demand for political accountability over the "dirty war". He concludes that the public expects abuse of public services to be dealt with by legal sanctions, rather than political sanctions 51  (1999, 85).  However, the reaction to the Madrid bombings on March 11 2004 suggests that democratic th  accountability is now functioning healthily. A parliamentary commission is investigating whether the bombings could have been prevented, and what effect they had on the March 14  th  national election, which was lost by the incumbent Partido Popular (PP). Two senior police officers have testified to the commission that PP ministers stated that there was forensic Although this may appear to be an example of the control model (i.e. using the police to suppress groups which threaten the unity of the state), Lustick would argue that once political violence has begun, the control model has failed to manage the conflict and is no longer operating. 51  43  evidence linking E T A to the attack, despite the fact that this evidence had not yet been collected, and despite the police view that Islamic militants were responsible . This suggests 52  that the police are less politicized than they were in the 1980s - they are not prepared to collude with the government in manipulating public attitudes towards subversive groups. Furthermore, the reaction to the PP's statements suggests that Spaniards are now very ready to inflict "political sanctions" - seen not only in the establishment of the parliamentary commission, but also in the results of the election.  A community based model of accountability has also been introduced in the Basque region. A regional police force was established in the Basque Country in 1982. The regional police are responsible for public safety and "ordinary" crime (the national and regional police share responsibility for countering political violence ). The regional force is accountable to (and 53  funded by) the Basque Parliament. Wright and Bryett note that this arrangement has not become "politicized" (i.e. sectarian) as happened in Northern Ireland. This is because there is not a split within the Basque country. The group is ethnically and politically homogenous and support for autonomy from Madrid is high (Wright and Bryett, 2000, 25). However, they note that relations between the national and regional police are not good (precisely due to Madrid's opposition to further autonomy for the Basques). As they have overlapping functions this has reduced efficiency as the forces tend to conduct parallel investigations rather than collaborating (ibid, 83).  Spain MPs investigate train bombings, B B C Online 20 May 2004 and Madrid "rushed to blame Basques" B B C Online 7 July 2004. It was national police officers who were involved in the dirty war against ETA. 5 3  44  2.3  Summary  The societies surveyed use a wide variety of mechanisms to hold the police to account, covering all four of the accountability models. Democratic and community structures clearly interact - democratic oversight tends to perceived as more legitimate, and better at involving the public, when employed by a local institution. However, the experience in Northern Ireland prior to 1969 shows the limitations of the (majoritarian) democratic model when the local community is divided - the public service ends up being responsive to the majority group, at the expense of minorities.  Spain and Northern Ireland are also examples of a failure of the control model - in both cases the subordinate groups (Basque Nationalists and Irish Nationalists respectively) are highly politicized and have called for more autonomy. These experiences suggest that it is difficult 54  to insulate democratic and community models from a conflict. However, The Netherlands shows that if these models are consociational, a consensus can be built around the role of the police. As Carpenter argues, such broad agreement that an agency provides a useful service will provide that agency with legitimacy. A police service which is widely trusted and respected, as it is in The Netherlands, is better placed to contribute to conflict management.  Most of the accountability models surveyed in this chapter have been concerned with strategic and fiscal competence (the technocratic and democratic structures in England and Wales, and the legalistic complaints commissioners in the Netherlands) or responsive competence (the The Basque regional government has developed a plan for "free association" with Spain which the Basque Parliament will vote on in autumn 2004 (BBC Online, Spain to block Basque plan, 31 October 2004). 5 4  st  45  majoritarian democratic model in Northern Ireland until 1972 and the community models in The Netherlands and the Basque region). Unsurprisingly, it is the legalistic models which focused on regulatory competence. Judges and Complaints Commissioners monitor police compliance with the law in general and human rights standards in particular. What is more notable is the way these models involve the public. Any individual can use them to hold the police to account - one does not need to be a politician, have any particular expertise or even be a member of a local police liaison committee. A less formal mechanism, such as that used in The Netherlands may be even more successful in this regard. Mediation enables the complainant themselves to reassess the incident, consider the police response to it, and negotiate a remedy, whereas in a strictly legal model these would be the functions of an independent arbiter . 55  Obviously, the extent to which formal legal remedies are open to the public depends on whether legal aid is available to those unable to afford the cost of litigation. This is another reason why legalistic mechanisms such as complaints bureaux, which are free, may be better at involving the public than the law itself.  46  CHAPTER 3: FITTING ACCOUNTABILITY MODELS TO NORTHERN IRELAND ELITE PREFERENCES This chapter will assess the preferences for different types of police accountability among the four main political parties in Northern Ireland. These are the Democratic Unionist Party (DUP), Sinn Fein, the Social Democratic and Labour Party (SDLP) and the Ulster Unionist Party (UUP). During the period that the Northern Ireland Assembly (NIA) was operational, these four parties formed the political executive, under the First Minister, David Trimble (leader of the UUP) and the Deputy First Minister, Mark Durkan, leader of the SDLP. Figure 3 shows where these parties fit in the Northern Irish political scene.  In considering a party's reaction to particular accountability mechanisms, it is important to note that Northern Ireland's religious divide is reflected in its political party system. Alice Brown et al state that elections are contested within Nationalist and Unionist communities rather than across Northern Irish society as a whole (2002, 82). It follows that although the two Unionist parties share similar political objectives, as do Sinn Fein and the SDLP, they are also electoral rivals. Much of the rhetoric about policing can therefore be understood as a concern not to be outmaneuvered by the other party or as an attempt to out-bid the other party by demonstrating that they are the most capable of protecting their segment's interests . 56  A good example is a speech given by Ian Paisley, leader of the Democratic Unionist Party, during the 2003 Election. He said that Sinn Fein was likely to become the largest Nationalist party in the NIA and that Protestant voters should therefore reject the UUP who would appease them and "destroy the foundations of unionism". (Voters can "move process forward", B B C Online, 11 November 2003). There are also many examples of Sinn Fein accusing the SDLP of "selling out" by joining the Policing Board.  47  •3 &  I  e o  £  o  "3  Q  t.  00  CN  2  =«5  •S  OH  ~ H  so  .2 42  S  •2  *r> *. 5  5$ O  ^. K.  o Q a, (i, o a, o o,  o  B  < .2  3  5^  s  R  E CD •S 2  .9  V-  o  o  O  m O O CN CD  *  CO  R O C3 O  H  I  bo a  o  iw  ON  3  •5  1)  •a  o -a  « s U  s;  &o "C O.H-1  CD  c ca  ca  w 5  (L>  X!  M  W  S * fr  •a ; | c ftj  a o O Co °0 CN  r\  c CD  .a R  a.  I-I  T3  £ -2 g  T3  *  3 O  •S "3  CD  T3 T3 r_  S  s  OH  oo  c  <L>  1  C O CS  a .o  3  C3  •s c OO  CO  o  CO  T3 C  00  O  >  O 00  J3  to  •s CD  s  O  'I  CD  ^ >; s c/a  C3  •5 o c .S O  •^3  o  CD  2 « •<? C •S CM  e sgg  o  CD _D  E 3  8  oo  3.1 Political Appraisal of the Technocratic Model The Patten Commission was concerned that the changes it advocated were radical and complex, taking up to ten years to complete. It therefore recommended a new accountability mechanism, the Oversight Commissioner, to monitor the implementation of the reforms (Patten Report, 1999, paras 19.1-19.6). The Oversight Commissioner is appointed by the Northern Ireland Secretary on the basis of his or her experience in law enforcement. In order to ensure the neutrality of the Commissioner, only individuals who are neither British nor Irish are considered for the position. The incumbent, A l Hutchison, is the former Commanding Officer of the Ontario Division of the Royal Canadian Mounted Police and his predecessor, Tom Constantine, was the former head of the US Drug Enforcement Agency.  There has been little political debate about the use of independent experts to oversee the police reforms; what comment there has been comes from the Unionist parties. When the Patten Report was published, David Trimble described the recommendation for an overseas Oversight Commissioner as an insult to the people of the U K  5 7  This echoes the  general complaint of Reg Empey, a UUP Member of the Legislative Assembly (MLA) that:  "we have been spectators in our own country, watchingfrom the sidelines as somebody flies in to determine policy". (Official Record, Northern Ireland Assembly, 18 January 1999.)  More recently, the DUP has attacked the actual work produced by the Oversight Commissioner. Ian Paisley Jnr, M L A (and a Member of the Policing Board) described  5 7  B B C Online, Unionists: Patten Report "flawed", 13 September 1999.  49  one of the Commissioner's reports as "uninspiring, predictable Patten rhetoric" and criticized its focus on issues such as human rights, 50/50 recruitment, and the involvement of political parties on the Policing Board, which he viewed as "nationalist hobby-horses" . Nigel Dodds, a DUP MP, has described the fees paid to the 58  Commissioner as an abuse of public money, on the grounds that they were spent:  "bolstering the various institutions . . .which derive from the Belfast Agreement " 59  (quoted in B B C Online, MP critical of "wasted money ", 27 May 2003)  These reactions suggest two things. Firstly, there is frustration with Northern Ireland's lack of autonomy and a resentment of the implication that oversight needs to be insulated from the conflict (because this implies that accountability mechanisms need to be insulated from Northern Irish people). Secondly, it confirms Aberbach and Rockman's point that it is difficult to divorce politics from questions of administration. Both Unionist parties disagree with parts of the Patten Report and the DUP do not 60  support the Good Friday Agreement. They are therefore unlike to be overly concerned with the efficient implementation of reforms which they oppose. This supports the arguments put forward in Chapter 1, namely that in a severely divided society it is difficult to rely on expertise or independence as insulation from conflict.  B B C Online, Policing "needs Sinn Fein support" 27 April 2004. The formal name for the Good Friday Agreement. In addition to their views on accountability issues, which are discussed in this chapter, the Unionist parties objected to changes to the RUC's name and symbols and the 50.50 recruitment of Catholic and non-Catholic police officers (as this is slightly disproportionate to the province's population). B B C Online, Unionists: Patten Report "flawed", 13 September 1999. 5 8  5 9  6 0  50  3.2 Political Appraisal of the Legalistic Model Despite this distrust of independent, depoliticized accountability, at the time of the Patten Report there was all party support for an Ombudsman. There has been 61  disagreement over her day to day work (as will be discussed in Chapter 5), but this does not detract from the consensus that an independent complaints mechanism is needed. There is also some agreement over the need for a Bill of Rights (supported by both the DUP and SDLP ). 62  For nationalists, this is because they see the RUC as an instrument of control, of the type described in section 1.1.2. An example of this argument is the speech given by Martin McGuiness, M L A (and member of the Northern Ireland Executive) in response to PSNI raids of Sinn Fein offices:  63  "They (the police) sympathise with and are loyal to rejectionist Unionists and they are beavering away continuously to undermine the Good Friday Agreement" Official Record, Northern Ireland Assembly, 8 October 2002.  Prior to the beginning of the police reforms, the SDLP also expressed this view stating that the RUC had been an "instrument of the Unionist government and a defender of the Unionist ethos" . Interestingly, hard-line Unionists make a similar analysis, although 64  the position of victim and oppressor are reversed. The UK Unionist Party's  6 5  submission to the Patten Commission argues that the Protestant perception of the RUC has changed from one of support to: Wright and Bryett, 2000, 50-51, 55 and 60. DUP, 1997, 3 and SDLP, 1999, 4. Three Sinn Fein staff members were charged with possessing information likely to be of use to terrorist organisations. The raids led to the suspension of the NIA in October 2002. SDLP conference position paper, cited by Wright and Bryett, 2000, 53. The U K U P is a one-MP/MLA party which is in alliance with the DUP. It seeks the further integration of Northern Ireland into the U K (ibid, 61). 61  6 2  6 3  6 4  65  51  " a view that the RUC is policing a government policy which threatens expression of the majority community's political, national and cultural identity . . .the progression of a policy aimed at a united Ireland and the use of the RUC to police and enforce if (UKUP, 1998,10, emphasis added)  These examples illustrate the belief of both groups that the Northern Irish police were either maintaining or helping to instigate a regime which discriminates against large swathes of the population. Nationalists believe that the RUC sought to suppress both the Catholic culture and demands for the re-unification of Northern Ireland. Conversely, the Unionists fear that the British government will relinquish Northern Ireland to the Republic of Ireland, which would discriminate against Protestants (Farren and Mulvihill, 2000,12-13). By controlling protests against the Anglo-Irish Agreement (see footnote 66) the RUC were supporting this possibility.  67  Such responses suggest that prior to the Patten reforms the RUC were not following the political preferences of either group. This is similar to the agencies described by Carpenter as bureaucratically autonomous, except that the RUC clearly did not have a broad coalition of support. However, it could be argued that the support of either group was not necessary as they had no legislative power and did not control resources. The RUC therefore retained independence from local politicians through the narrow support The (UK) government policy referred to is the Anglo-Irish Agreement of 1985, in which the U K stated for the first time that it was willing to renounce sovereignty over Northern Ireland, if that was the desire of a majority of the Northern Irish population. This led to Unionist protests and riots and therefore conflict between Unionists and the RUC. Further clashes have resulted during the marching season when the R U C have sought to enforce bans on Orange Order parades passing though Catholic areas, (ibid, 7-8 and 63-64). A similar analysis could be could be made of Franco's use of the police to suppress Basque and Catalan nationalism (see section 2.2.2), the police treatment of Afro-Americans in the US South prior to the Civil Rights movement and policing in apartheid South Africa (Wright and Bryett, 2000,27). The unusual feature in Northern Ireland is the "double minority" situation which leaves both groups feeling insecure; as a result both argue that the police discriminates against their community for the benefit of the other. 6 7  52  of the only entity with any political power - the U K government. This would explain the widespread support (at least in principle) for legalistic models - neither side trusted the RUC to treat individual members of their community fairly, but had no political power with which to hold them to account, so instead looked to legal alternatives.  3.3 Political Appraisal of the Community Model Given this background, it is unsurprising that by the late 1990s (the time of the Patten Report and the Good Friday Agreement), there was widespread agreement that links between the police and the community had to improve. However, there was disagreement over how this should be achieved, with Nationalist parties supporting the devolution of police command to local units , and Unionists favouring more power for 68  Community Police Liaison Committees (which only had advisory powers). This split 69  reflects the basic disagreement over policing. Sinn Fein saw localized policing as a precursor to an all-Ireland force, whilst the SDLP saw it as part of a major reform programme and part of a consociational settlement (the logic behind localized policing 70  was similar to Lijphart's description of segmental autonomy) . At the other end of the 71  spectrum, Unionists were dissatisfied with the RUC but wanted only minor changes to existing structures, arguing that the resolution of the conflict would solve policing problems (UUP, 1998,15).  Wright and Bryett, 2000, 56 and M c Garry and O'Leary, 1999, 88. UUP, 1998, 15-18 and Edwin Pools (DUP MP), Official Record, Northern Ireland Assembly, 16 February 1999. McGarry and O'Leary, 1999, 8 and 88. Although it would be difficult to implement given the proximity of the two communities in cities such as Belfast and Portadown.  6 9  th  7 0 71  53  These differences over the desired outcome (between Nationalists and Unionists) and over strategy (between Sinn Fein and SDLP) can be seen in the reaction to District Policing Partnerships (DPPs), which were recommended in the Patten Report. DPPs operate at district council level. They assess the performance of their local police command and which, in conjunction with the District Commander, decide die local policing plan. Although both Unionist parties have allowed their members to sit on DPPs, they have been extremely critical of the fact that the governing legislation permits ex-prisoners to join the partnerships. They argue that Unionists will be forced out of these groups, leaving DPPs vulnerable to control by paramilitary groups . The 72  result would be local policing which was controlled by groups that wanted to undermine 73  the RUC , which would hardly be conducive to strategic competence.  Whilst there is accord between the Unionist parties on this issue, Sinn Fein and the SDLP fundamentally disagree over the efficacy of DPPs. Gerry Kelly, MLA, Sinn Fein's spokesman for policing, has stated that DPPs have little power , whereas the 74  SDLP spokesman, Alex Attwood, M L A told a party conference that the DPP is: "a powerful vehicle to create an accountable and responsive police service; addressing .. .the common problems of all our community" Policing body "powerful vehicle ", B B C Online, 19 March 2003.  Unionists "will be driven off boards", B B C Online, 3 December 2002. In fact, the provision permitting ex-prisoners to participate will only come into force when a commencement order is laid before Parliament. Paul Murphy, the Secretary of State for Northern Ireland, has said he will not do this until paramilitary groups have decommissioned their weapons (Ex-prisoner clause "likely", B B C Online, 24 March 2002). DUP, 1997,1 and interview with Ken Maginnis, M P (UUP spokesperson on security at the time, now Lord Maginnis), cited in Wright and Bryett, 2000, 58. The unionist parties predicted before the Patten Report was published that local community groups would eventually be opened up to ex-prisoners. Call to join NI policing partnerships, B B C Online, 6 September 2002. u  7 3  74  54  Sinn Fein has not allowed its members to join DPPs and has staged several protests outside DPP meetings . In County Deny/Londonderry, the party's councillors even 75  managed to get a DPP meeting banned from a community hall, on the grounds that the town in question was "99% Catholic" and therefore not an "appropriate place" for a body associated with the police . 76  3.4 Political Appraisal of the Democratic Model This disagreement is echoed in the debate over the Policing Board , which the SDLP 77  has joined and Sinn Fein has not. Martin McGuiness had previously rejected the idea of giving elected politicians control over policing on the grounds that such arrangements would be dominated by Unionists (McGarry and O'Leary, 1999, 111). However, this comment was made before the Good Friday Agreement was finalized, which envisaged consociational democratic structures rather than majoritarian ones. Sinn Fein now advocates the devolution of the Secretary of State's policing powers to the Northern Ireland Executive, with Gerry Kelly arguing that  "transfer. . . is crucial because it is the only way that control ofpolicing... can ultimately be wrested out of the hands of British securocrats in London .. .without transfer, policing andjustice will remain unaccountable and a tool of repression " (The Scotsman, 29 Feb 2004) th  This suggests that Sinn Fein no longer have an objection to the police being answerable to local politicians per se. Indeed, Gerry Adams, the leader of Sinn Fein, has stated that he can see circumstances in which they would join the Board. However, he is under Policing protest "intimidation ", B B C Online, 22 May 2003. The quotes are from a local Sinn Fein councillor, Martin McGuidy, taken from DPP Meeting "banned", B B C Online, 23 October 2003. The Policing Board replaced the Police Authority of Northern Ireland (see Chapter 2, Section 1.3). It has 9 independent members, appointed by the Secretary of State, and 10 elected members, drawn from the Northern Ireland Assembly. When the N I A was suspended, the Secretary of State re-appointed these 10 members. 75  7 6  7 7  55  pressure from hardliners in his party to concentrate on a unified Ireland, which precludes support for institutions rooted in Northern Ireland alone. At the 2003 conference, a motion was put down to link joining the Board to an end to any British jurisdiction in Northern Ireland. Although the motion was defeated it demonstrates the 78  risk Sinn Fein leaders would face if they endorsed the police beforetiieirfollowers were willing to follow suit.  This explains Sinn Fein's regular attacks on the SDLP for joining the Board, stating that the latter has "jumped too soon", thus destroying the chance of gaining further concessions through a united nationalist boycott . More emotively, they also accuse the 79  SDLP of endorsing human rights abuses against its own community: "Will they continue to back a clearly politically motivated police force that can brutally assault a father and terrorise a mother and children, raid homes, collude with Unionist terrorists and cover the backs of the "no " (to the GFA) camp in this Assembly? " Mary Nelis, M L A , Official Report, Northern Ireland Assembly, 8 October 2002.  The SDLP itself argues that to be on the Board is not necessarily to endorse the police; rather it entails critically reviewing police performance, and challenging what it does not like. They criticize both Sinn Fein for abstentionism and Unionism for blind support of the police. Mark Durkan (the SDLP leader) analyses Sinn Fein's position as being 80  one of strategic calculation. On the one hand, it delays joining the Board in order to get further concessions but still hints that it may do so soon in order to keep the U K government engaged.  81  78 79 8 0 81  On the other hand, it uses its non-participation to prevent itself  SF "hedging bets over policing", B B C Online, 31 March 2003. SF "close to biting policing bullet", B B C Online, 25 March 2003 Alex Attwood, M L A , Official Record, Northern Ireland Assembly, 10 September 2002. B B C Online, Durkan jabs Sinn Fein on policing, 25 February 2002  56  being out-bid by hard-line republicans; Durkan refers to it as "their remaining totem of subversion" . 82  The SDLP's own strategy is to emphasise the changes to policing that have occurred since 2001, and link this to its own presence on the Policing Board. This theme essentially argues that participation in democratic accountability will have a greater impact on police responsiveness than the politics of protest . Such arguments were 83  repeatedly used by the SDLP in the campaign for the 2003 Northern Ireland Assembly election . However, the results of the 2003 NIA elections (in which Sinn Fein won 84  more seats than the SDLP) suggest that this argument has limited appeal.  These reactions to the democratic model, and to the Policing Board in particular, are best understood in the context of the 2003 NIA election. The electoral campaign tended to focus on the future of the peace process rather than substantive issues, such as health and education. This was because the NIA was suspended and the British and Irish governments had announced that a review of the GFA would follow the election. Voters were therefore not electing an executive; instead they were choosing who would represent them in future negotiations (BBC Online, 28 November 2003).  SF "hedging bets over policing", B B C Online, 31 March 2003. An example is Mark Durkan's speech to the party conference; see Chief Constable addresses SDLP, B B C Online, 1 March 2003. See the 2003 manifesto, (SDLP, A Manifesto for Equality. Justice and Prosperity. 2003,25) and Devenport, "Battle within nationalism", B B C Online, 11 November 2003 for further examples. 82  8 3  8 4  57  Both the DUP and Sinn Fein argued that the hard-line party in the opposite community would win the most seats and would take advantage of a "moderate" party (i.e. the UUP or SDLP) in the subsequent negotiations. Each group was therefore encouraged to vote for a hard-line party in order to protect their interests. The result was that the DUP and Sinn Fein became the largest parties in die Unionist and Nationalist communities respectively. This suggests the mutual security dilemma described by Farren and Mulvihill (2000,105-106) persists in the political arena - whereas paramilitaries argue that they cannot disarm until their opposite numbers do, voters appeared to feel that they could not elect moderates because they feared the opposite community would not do likewise.  The UUP and SDLP had different strategies to respond to this security dilemma, neither of which was successful in the 2003 election. The UUP tried to occupy the same ground as the DUP. On policing, the two parties coordinated their response to the Patten Report, one example being their joint announcement that they would join the Policing Board (BBC Online, 21 September 2001). More generally, the UUP broke off talks with Sinn Fein in October 2003 (one month before the NIA elections) and campaigned on the basis that it had "stood up to the IRA" (Devenport, 2003). The SDLP approach was to embrace the "moderate" label, citing the Policing Board as a positive example of engagement with Unionists, rather than confrontation.  3.5 Summary This survey of political reactions to the various accountability models in the Patten Report suggests that only the legalistic model has universal appeal. The technocratic  58  model attracted little interest, and comment was critical. Given the suspicion of oversight by non-political experts, it is ironic that most of the debate over the community and democratic models has been over the fact that they are "politicized" (i.e. not insulated from the conflict). Unionists have tended to worry about who is influencing die agenda of these accountability bodies (particularly in the case of DPPs). Meanwhile, Sinn Fein and the SDLP have been involved in an electoral struggle over the merits of participation as opposed to protest as a means of holding the police to account.  59  CHAPTER FOUR: FITTING ACCOUNTABILITY MODELS TO NORTHERN IRELAND - PUBLIC PREFERENCES  This chapter will consider public attitudes to the various models of police accountability and discuss the degree of agreement between communities and their political leaders. The data analysed is from the Community Attitudes Survey (CAS). This is a survey of the Northern Irish population which examines public attitudes to the criminal justice system. It has been administered yearly since 1992/1993 by the Northern Ireland 85  Statistics and Research Agency (an agency of the U K government). There are as yet no questions in the CAS about technocratic accountability holders, such as the Oversight Commissioner, Her Majesty's Inspectorate of the Constabulary or the Northern Ireland Audit Office. This chapter will therefore focus on public attitudes towards democratic, legalistic and community based institutions.  4.1 Legalistic Model Figure 4 shows the responses given when people were asked who they thought should look into complaints about police behaviour (i.e. matters of regulatory competence). There is clear support for one type of legalistic institution - a body which is independent of the police and which investigates complaints from the public. However, this is only support for that particular form of the legalistic model - very few people saw this kind of issue as being one for the courts. Although an independent Commissioner or Ombudsman is by far the most popular choice for both communities, there is a 9 point gap between the two groups; with Catholics being more likely to As at July 21 2004, only summary data has been released from the 2003 CAS. As the analysis in this chapter requires the original data set, all figures cited will correspond to the 1992-2002 period, unless stated otherwise. 85  st  60  support this model. Conversely, non-Catholics were more l i k e l y than Catholics to t h i n k the C h i e f Constable should investigate complaints, although this was the clear second choice f o r Figure 4: W h o should l o o k into complaints about police behaviour? Cath 19922000 %  Cath Total 2001Cath 2002 % (after police reforms) %  NonCath 19922000 %  1  0  1  2  2  2  2  C h i e f Constable  12  11  Independent C o m m i s s i o n /Police Ombudsman  50  Local community groups  NonCath 20012002 (after police reforms %  Total NonCath  TOTAL POP'N %  %  2  2  1  6  4  5  4  11  16  14  16  14  61  53  42  55  44  48  1  1  1  1  1  1  1  Courts  1  0  1  0  0  0  0  District/local council  2  1  2  1  1  1  1  N I A member  0  1  0  0  1  0  0  N o - o n e i n particular  1  1  1  1  1  1  1  N o r t h e r n Ireland Office/Secretary o f State f o r N o r t h e r n Ireland N o r t h e r n Ireland P o l i c i n g Board/Police Authority 8 6  87  88  1. A l l figures to nearest integer. Percentages refer to the % of people in that demographic group who chose a particular institution - i.e. 1% of Catholics between 1992 and 2000 thought that the Northern Ireland Office should look into complaints, as opposed to 2% of nonCatholics over the same period. 2. Percentages will not add up to 100 as "don't know" and "other" responses are not shown. Source: Analysis of Community Attitudes Survey datasets, 1992 to 2002. The Northern Ireland Policing Board replaced the Police Authority of Northern Ireland in 2001. This option was posed hypothetically until 2000, when the Office of the Ombudsman was established. District Policing Partnerships were not established until early 2003, after this data was collected. 8 7  8 8  61  both groups. This fits with the arguments of political leaders who challenge the integrity of the police and champion the need for independent oversight. Although there is widespread support for the legalistic model (among both political elites and the public), it is Catholics who desire it the most. This may be due either to their own experiences or the message they get from their leaders on the prevalence of police abuse against members of their community.  The public not only endorsed the legalistic model for its regulatory functions; they also thought that it would help the police to perform better (i.e. it would encourage strategic competence). When asked if they thought the Ombudsman would "help ensure the police do a good job" 74% answered yes . The religion, employment status and age of 89  the respondent had no significant effect on how likely he or she was to think this, although women were 24% more likely than men to think so . This suggests that the 90  main cleavage in Northern Irish society does not affect confidence in the Ombudsman. It also shows that three of the groups which are most likely to come into conflict with the police (Catholics in Northern Ireland, and the young and economically disadvantaged in any society) believe her role is useful. She is therefore well placed to be a resource to the entire community, including those most likely to need her help.  This question was asked for the first time in 2002 so this figure only refers to that year. Figure 5 shows, however, that only 2% of the general population would go to the Ombudsman first with general concerns about the police. However, this does not detract from the current finding. As a legalistic model which was designed primarily for individual cases, one would not expect it to be the first port of call for those with policy concerns. What is clear, however, is that the public does not feel that the Ombudsman's focus on regulatory competence detracts from the PSNTs ability to perform strategically. The p value is 0.034, meaning that there is only a 3% chance that this result from the sample is not reflected in the entire Northern Irish population. Age, sex, religion and employment status were regressed on the dependent variable (belief that the Ombudsman would help the police do a good job), using S T A T A software. N=l,430. 8 9  9 0  62  4.2 Democratic accountability Figure 4 also shows little support for the concept of politicians holding the police to account for individual incidents. Less than 5% of any group thought either the Police Board, the Secretary of State, a member of the NIA, or a district councillor should be responsible for looking into allegations of police misbehaviour . It is interesting that 91  this is true no matter how local or remote the democratic institution is (i.e. whether it operates at the local, provincial or national level) and whether it is consociational in structure (the Policing Board, which has Nationalist, Unionist and independent members) or essentially majoritarian (the Secretary of State, who comes from a oneparty cabinet).  However, Figure 5, below, shows that the picture is more complicated when the public are asked to consider who they would go to if they wanted a "general change in the way the police do their job." The formal democratic accountability mechanisms still do badly - only 1% would go the Policing Board (or PANI when it existed) and less than 1% would contact the Secretary of State (or more realistically, the Northern Ireland Office). This is despite the fact that these two institutions form two of the links in the tri-partite structure of accountability in Northern Ireland. However, the public would turn to politicians - their local representatives. A Member of the Westminster Parliament is the most popular choice, with 26% of the general population picking them as their first preference. Very few, however, mentioned Members of the Northern Ireland Assembly, which suggests that the public understand that the division of powers  6% of non-Catholics in 1992-2000 thought that the Police Authority of Northern Ireland (PANI) should investigate such matters; but PANI members were not elected during this period, so this is hardly an endorsement for accountability through representative democracy. 91  63  Figure 5: Who would you contact the police do their job? Cath 19922000 %  first if you wanted to see a general change in the way Cath Total 2001Cath 2002 % (after police reforms) %  Local police station  21  19  20  27  NonCath 20012002 (after police reforms % 27  Police HQ or Chief Constable  5  4  5  6  6  6  6  Northern Ireland Office/Secretary of State for Northern Ireland  0  0  0  0  0  0  0  Northern Ireland Policing Board/Police Authority  1  1  1  2  1  1  1  Independent Commission /Police Ombudsman  1  5  2  1  3  1  2  Local community  1  2  1  1  2  1  1  13  17  14  9  11  10  12  25  20  23  28  25  27  26  NIA member  0  4  2  0  0  0  1  Wouldn't contact anyone  8  9  8  6  6  6  7  1  NonCath 19922000 %  TOTAL POP'N %  Total NonCath  27  24  %  92  93  94  groups District/local council/councillor Member of (Westminster) Parliament  1. A l l figures to nearest integer. Percentages refer to the % of people in mat demographic group who chose a particular institution - i.e. 1% of Catholics between 1992 and 2000 would contact local community groups first to discuss a concern over policy, as opposed to 2% of Catholics in 2001 and 2002. 2. Percentages will not add up to 100 as "don't know" and "other" responses are not shown. Source: Analysis of Community Attitudes Survey datasets, 1992-2002.  The Northern Ireland Policing Board replaced the Police Authority of Northern Ireland in 2001. This option was posed hypothetically until 2000, when the Ombudsman was appointed. District Policing Partnerships were not established until early 2003, after this data was collected.  9 3  9 4  64  between Westminster and the N I A ; p o l i c i n g is a reserved power, meaning that the N I A w o u l d n o t be able to b r i n g about a "general change i n the w a y the police do their j o b " .  Those w h o w o u l d go to their M P are more l i k e l y to be non-Catholic, w h i l s t those w h o w o u l d go to their local c o u n c i l or councillor are more l i k e l y to be Catholic (although the gap is not large). This m a y be due to the different electoral systems used; M P s are elected under a first past the post system, a majoritarian system w h i c h w i l l favour Unionists. L o c a l and district councilors have been elected v i a proportional representation since 1973 (Farren and M u l v i h i l l , 2000, 84). This m i g h t suggest that Catholics are more amenable than non-Catholics to consociational democratic accountability models, as they ensure Nationalist representation. H o w e v e r , this conclusion is tenuous as more Catholics still chose an M P as their first choice ( 2 3 % ) than chose a councilor ( 1 4 % ) , and there is n o data o n why whose w h o mentioned councillors d i d so. I n addition, only 1 % mentioned the P o l i c i n g B o a r d , w h i c h is e x p l i c i t l y set up as a consociational structure.  4.3 C o m m u n i t y based accountability Figures 4 and 5 do n o t show public support f o r the role o f c o m m u n i t y groups, either i n responding to i n d i v i d u a l incidents, or i n matters o f general p o l i c y (they were mentioned by o n l y 1 % o f both communities). H o w e v e r , the n e w D i s t r i c t P o l i c i n g Partnerships  95  recommended b y the Patten Report o n l y began operating i n 2003 (and i n the case o f  Half of the members of a DPP are local councillors, nominated by their political parties. The remainder are selected by the Policing Board as independents. Any local resident who does not have a criminal record may apply for these positions.  65  South Tyrone, are still being established)  . The c o m m u n i t y groups that existed before  were ad hoc and only had advisory p o w e r s , so it is not surprising that they have a l o w profile.  There is skepticism about h o w m u c h input these committees had - only 9 % o f those surveyed thought that the police took " a lot o f n o t i c e " o f the views o f C P L C s . A s Figure 6 shows, a respondent's religion, sex, e m p l o y m e n t status and b e l i e f that the police understand local problems all have a statistically significant impact o n their views o n the efficacy o f CPLCs. B e i n g Catholic, male, u n e m p l o y e d and b e l i e v i n g that the police do not understand local problems all make a respondent more cynical. I n contrast to the cross-community support given to the O m b u d s m a n , it was d i f f i c u l t to convince " o u t groups" (those most l i k e l y to have problems w i t h the police) that C P L C s were effective accountability holders. H o w e v e r , the practical effect is not large - these four factors combined make a respondent move o n l y 0.888 o f a p o i n t along the four p o i n t ordinal scale measuring c y n i c i s m . This means they are n o t quite sufficient to m o v e a respondent f r o m t h i n k i n g that the police take "some n o t i c e " o f C P L C s to t h i n k i n g they take " n o t m u c h notice".  H o w e v e r , Figure 5 shows that the local police station was the second most popular choice f o r people w h o wanted a general change i n police practice ( 2 4 % o f the w h o l e p o p u l a t i o n ) , far ahead o f the P S N I / R U C headquarters or C h i e f Constable (chosen b y 98  Minutes of the 31 Meeting of the Northern Ireland Policing Board. 20th May 2004. Available on-line at. http://www.nipolicingboard.org.uk/word docs/public session/minutes Mav2004.doc These were called Community Police Liaison Committees or CPLCs.  9 6  st  9 7  66  Figure 6: The relationship between demographic factors and cynicism about Community Police Liaison Committees. How much notice do vou think the police take of what C P L C s " have to say? (Dependent Variable)  Independent Variables  Estimated Co-efficient (Robust Standard Error) p value (all to 3 d.p.)  Age  -0.001 (0.001) 0.289  Sex  -0.068 (0.024) 0.005*  1= female, 0 = male Religion 1= Catholic, 0 = non Catholic  0.133 (0.024) 0.000**  Employment Status 1= employed 0 = unemployed  -0.116 (0.025) 0.000**  Year of data collection 1= after the beginning of police reforms, 0 = before police reforms  -0.055 (0.030) 0.065  Belief that the police understand your area's problems 1= yes, 2= it varies 3 = no  0.285 (0.014) 0.000**  Intercept  1.998 (0.065) 0.000**  N R  3,473 0.144  2  * statistically significant at the 5% level ** statistically significant at the 1% level Source: Analysis of Community Attitudes Survey datasets, 1992-2002.  However, there is a 7% gap between the two communities, with Catholics being less likely to go to their local police station than Protestants. This gap actually increased to 9% in the two years after police reforms began, which shows that the police still need to improve their relations with the Catholic community. The whole question was worded "In some areas, groups of local people meet regularly with the police to discuss matters to do with policing the community. How much notice do you think the police take of what such a group has to say?" The respondent therefore does not have to recognise the term " C P L C " in order to give an answer. The possible answers were "a lot of notice", coded 1, "some notice", coded 2, "not much notice" coded 3 and "no notice at all", coded 4. The higher the value, therefore, the more cynical a respondent is about the influence of CPLCs. A negative co-efficient indicates that as the value of the independent variable gets higher, the respondent gets less cynical. 9 9  67  6%). Furthermore, a substantial group of people said they would approach their district council about a question of general policing policy. This bodes well for the DPPs, which follow council boundaries (as do the PSNI's newly established District Command Units) and draw 50% of their members from the local council. They should tiierefore be well placed to build on the reputation of local government as an effective accountability holder.  These findings suggest an appetite for localized policing which is matched by a desire for community based methods of accountability. When asked "do you think it is important for local people to have a say in the way in which they are policed or do you think decisions about policing are best made just by the police themselves", 72% opt for community based policing over operational independence for the police. The support for community policing has been steadily rising, from 69% before the Good Friday Agreement, 73% after the GFA, and up to 78% in the two years after reforms began. Furthermore, when Catholics were asked about their priorities for the police the most popular choice was "more police who understand the needs of the community" (42%) with the second most popular being "closer links between police and local community groups" (21%). Catholics therefore rated issues of responsiveness more highly than crime prevention, in contrast to non-Catholics, whose first choice was "more police on the streets, on foot" (37%).  CAS Survey data from the winter of 2003 (when most DPPs had been operating for six months) support the finding that there is a desire for more active community accountability. 72% of respondents had either some or total confidence that DPPs  68  would help address local policing problems, and 64% said they would personally contact their DPP if they had a question about or issue with local policing . However, 100  these figures have not been broken down into demographic groups, so it is not clear if both communities viewed DPPs in a similar way.  4.4 Summary The public (particularly Catholics) support both the legalistic model of the Ombudsman and the new community based model recommended in the Patten Report. However, they showed little interest in either the courts as a mechanism to deal with individual incidents or in community-police groups which are purely advisory. This suggests a public preference for mechanisms which are informal in character (not requiring litigation for example) but which involve them in decision making processes - either about local policing priorities or the way in which an individual incident will be dealt with. Public attitudes are therefore in accordance with the political elite's view of the legalistic model. However, they appear to be more optimistic about the effectiveness of the community based model than either Sinn Fein or the Unionist parties.  The public put more faith in their personal representative in Parliament or the local council than in the formal political models of accountability (the local Policing Board and the Secretary of State). The Patten Commission argued that the current tripartite arrangement was flawed as it gave too much power to a minister who was remote from the local population - he or she was not from Northern Ireland, and would never come  This information was released by the Policing Board to the press. (BBC Online, Continued confidence in NI Policing, 11 December 2003). 1 0 0  69  from a local political party (1999, paragraph 5.7). However, if the public were concerned simply about remoteness they could embrace the Policing Board and Police Authority, both of which consisted) entirely of local members. This may therefore be an area where the public share the uncertainties of the politicians. Prior to the Patten reforms, the old Police Authority was denounced as powerless by the Nationalist parties but was supported by Unionist politicians, (some of whom were appointed to the Authority by the Secretary of State). The situation is now more complicated, with a split among the Nationalists, and a delay of several months before the Unionist parties joined the new Board (see Chapter 5). Public ambivalence may therefore be a reflection of the mixed messagesfrompoliticians on democratic accountability.  70  CHAPTER 5: POLICE ACCOUNTABILITY IN PRACTICE - TWO CASE STUDIES. Thisfinalchapter will discuss how two of the accountability models (democratic and legalistic) have worked on the ground in Northern Ireland. An example of each model will be examined and assessed against the three criteria which were established in the introductory chapter: public involvement, insulation from the conflict, and policy competence.  The two selected are in many ways the antithesis of one another - the Policing Board's main task is to hold the police to account for general performance and to review their plans and policy, whereas the Police Ombudsman's role is to investigate individual complaints. The Board is essentially a political institution, with great care taken to make it representative of the general population. The majority of the Policing Board were originally elected (via the Northern Ireland Assembly); since the NIA was suspended they have been appointed by the Secretary of State on the basis of nominations from political parties. Furthermore, the members are all local to Northern Ireland. The Ombudsman and her staff are appointed on the basis of their expertise (in law and law enforcement) rather than their background; many are notfromNorthern Ireland. The body stresses that it is impartial and independent ; in other words, it seeks to perform 101  a technocratic function which is wholly separate from political debate and conflict.  A statement to that effect is the first thing visitors to the Ombudsman website see: http://www.policeombudsman.org  71  5.1 Legalistic accountability - The Police Ombudsman of Northern Ireland. At the time of the Patten Report, the British Government had already agreed to establish an Ombudsman to investigate public complaints about the police. This was a response to the earlier Hayes Report (Wright and Bryett, 2000,101). The author of the Hayes Report, Dr Maurice Hayes, was also a member of the Patten Commission which endorsed the proposal (1999, paragraphs 6.41 and 6.42). The Office of the Police Ombudsman for Northern Ireland (OPONI) was established in 2000.  5.1.1 Public involvement The CAS data examined in Chapter 4 showed that there had been a high level of demand for an independent Ombudsman throughout die 1990s, so it is not surprising that many took advantage of its services once it was established - in the first three years of operation, ten thousand complaints were received . 102  Despite this, Figure 7 shows that in 2002, when the Ombudsman had been operating for over a year, there was still a significant minority (32%) who had not heard that the institution now existed. However, there is no statistically significant relationship between age, sex, religion or employment status and awareness of the Ombudsman . 103  This suggests that the groups who might be more likely to have problems with the  Police complaints made public, B B C Online, 15 January 2004. A regression was run on ST ATA, calculating the relationship between age, sex, religion and employment status (the independent variables) and awareness of the Ombudsman (the dependent variable). A l l of the results had a p value of 0.062 or higher, meaning that there is more than a 6% possibility that relationships between the variables occurred in this sample by chance, and would not be replicated in the whole Northern Irish population. They are therefore not significant at the 5% level. N = 676, R = 0.0192. m i  103  2  72  Figure 7: Have you heard of the Police Ombudsman for Northern Ireland? (2002 only) Catholics (%)  Non-Catholics (%)  Total population  Yes  68  63  (%) 66  No  29  36  32  Don't know / refusal  2  4  2  Source: Analysis of Community Attitudes Survey 2002 Dataset. 1. The total population figures include those who refused to disclose their religion.  police (Catholics, men, the young or the socially and economically disadvantaged) are as aware of the Ombudsman - and therefore have the opportunity to use her services as the general population.  The public also has the opportunity to use the Ombudsman to gain information about policing, even if they do not have a personal complaint. In January 2004, OPONI began publishing detailed (but anonymous) information about the complaints that they were dealing with . Datasets are available free of charge on their website, showing the 104  religion, age, and gender of those making complaints, the issues complained about, the location and date of complaints and an analysis of common factors in complaints (such as arrest, domestic violence, inebriation etc) . The Ombudsman also publishes reports 105  on the systemic issues arising from complaints, such as the use of police weapons in public order situations; these are also available to the public . The Ombudsman 106  Police complaints made public, B B C Online, 15 January 2004. Office of the Police Ombudsman of Northern Ireland website, http://www.policeombudsman.org/Statistics.cfm Ibid, 2003. These reports are available at: http://www.po lireombudsman.org/Publication.cfm?catTO^ ar=2003 1 0 5  1 0 6  73  therefore provides resources to the public enabling them to assess the behaviour of the police, and identify particular problems in their area.  5.1.2 Insulation from the conflict The structure and personnel of the Ombudsman suggest an attempt to insulate the work of the Office from the political conflicts over policing. The Office is not only independent of the police, but also of the Northern Irish political system. Even when the NIA was sitting, the Ombudsman reported to (and was funded by) the Northern Ireland Office in London. The Ombudsman herself is a solicitor and many of the investigators have either legal training or police experience (so can claim technocratic expertise). 40% of these investigators are not local, some drawn from other U K forces, but many from Commonwealth countries. This echoes the Patten recommendation that the Oversight Commissioner be hiredfromoutside the UK and Ireland to demonstrate his impartiality.  However, this has not protected the Ombudsman from a political reaction to her work. She has been championed by the SDLP, who campaigned in the 2003 NIA election on a promise to seek more resources for the Ombudsman  107  (despite the fact that the NIA has  no power to fund the Ombudsman). Conversely, the Unionist parties, while supporting the concept of an Ombudsman in general (see Chapter 3), have attacked the incumbent, Nuala O'Loan, arguing that she has been over-zealous in her investigations and is undermining the authority and reputation of the PSNI.  A Manifesto for Equality, Justice and Prosperity, SDLP, 2003, p4.  74  An example is the different reaction to the release of summary CAS 2003 data, which showed an increase in public support for the Ombudsman. Eddie McGrady, an SDLP Member of the Policing Board, said that the survey demonstrated "the unique openness" of the institution, whereas Ian Paisley Jnr, a DUP Member of the Police Board, expressed skepticism of the results:  "no manipulation (of survey data) will convince me that the Protestant public or the police believe the Ombudsman is impartial and does a fair job. " B B C Online, Support for Ombudsman "rising", 28 March 2003.  Such comments, which cast an institution in the role of protector of one community and scourge of another, make it difficult for an institution to remain unaffected by a conflict. Even if they do not affect the mindset of the Ombudsman staff, they may well discourage Protestants from using the service. They may also make it easier to dismiss the Ombudsman's findings as the product of bias, making it politically easier for the police not to implement the recommendations she makes.  This danger is demonstrated by the reaction to the Ombudsman's report on the PSNI investigation of the Omagh bombing . Several members of the UUP called for her 108  resignation, both when she decided to launch an investigation issued.  110  109  and after the report was  This was on the grounds that she was questioning the integrity of the PSNI  and therefore hampering its work, and that it was not her role to comment on  In August 1998 the Real IRA (a splinter group of the IRA) bombed Omagh (a mixed town near the border with the Republic) and killed 28 people, as part of their campaign against the Good Friday Agreement. The Ombudsman's report, published in November 2001, criticized the dissemination of intelligence within the (then) R U C ; a tip-off received several days before the bombing had not been passed on to local police officers. Statement by the Police Ombudsman for Northern Ireland on her investigations into matters relating to the Omagh bombing of 15 August 1998. Available at: htto://www.policeombudsman.org//Publicationsuploads/omaghreportpdf NI Police Ombudsman "s Role, B B C Online, 7 December 2001. Board aims to end Omagh dispute, B B C Online, 7 February 2002. 1U8  th  109  110  75  "inefficiencies in the police service" (Lord Kilclooney, deputy leader of the U U P )  111  . In  fact, she does have the power to review major police investigations (section 62, Police (Northern Ireland) Act 1998), and one of the functions of the Office of the Ombudsman is to "carry out research to inform and improve policing policy and practice" . These 112  comments gave political cover for the then Chief Constable to refuse to implement the Ombudsman's recommendation that a non-PSNI officer take over the Omagh investigation. He stated that the report was "neither a fair, thorough or rigorous investigation"; thus attacking her technocratic credentials and echoing Unionist arguments that she was biased against the police . 113  5.1.3 Policy competence Figure 8 shows, unsurprisingly, that the major focus of the Ombudsman's work is regulatory, covering complaints about racial discrimination (contrary to the Human Rights Act and anti-discrimination legislation), malpractice (contrary to the internal Police Code of Ethics) and oppressive behaviour (contrary to the Human Rights Act or criminal law). The complaints she receives do not concern fiscal competence, as they are made by members of the public who do not have access to detailed information about the police's use of resources.  Complaints about issues of responsiveness are also relatively low. Those that were received were mainly issues of incivility which were dealt with through informal  Marathon talks on Omagh investigation, B B C Online, 5 February 2002. Annual Report of the Office of the Police Ombudsman of Northern Ireland. 2002-2003. Available at: http://vww.polireombudsman.org/Publication.cfo Ombudsman and police chief in talks, B B C Online, 18 December 2001. 111  1 1 2  113  76  Figure 8: Complaints received by the Police Ombudsman for Northern Ireland, 20002004 Strategic Responsive Regulatory Other Total 3,246 1,950 5,724 1,298 12,218 27% 16% 47% 11% Source: Analysis of Complaints statistics, available on the Police Ombudsman web-site, available at: http://www.policeombudsman.org/Statistics.cfrn  mediation between the complainant and the police officer concerned. They did not result in formal investigation reports, which are sent to the Chief Constable, Policing Board and Secretary of State for further action. The Ombudman's work in this area therefore does not usually result in corporate action by the PSNI, although it may moderate the behaviour of individual officers . As the Ombudsman has an explicitly 114  non-politicised role this fits with Campbell's observation that responsiveness is usually a competence associated with politicians (2001,261).  The Ombudsman has recently sought to improve relations between her Office and the police.  115  A study was undertaken of police attitudes to the Ombudsman and her work,  and a joint working committee of the Office of the Ombudsman, the Police Federation (the trade union of lower rank police officers) and the Superintendents' Association of Northern Ireland (representing senior officers) was established This could be interpreted  An exception was a report published in 2003 concerning the treatment of solicitors, where the Ombudsman was concerned about the quantity of minor complaints received from members of the legal commission (plus two much more serious allegations of R U C involvement in the murder of solicitors by paramilitary groups) and decided to investigate . whether there were systemic causes for this trend. Report 1,2003, A study of the treatment of solicitors and barristers by the police in Northern Ireland. OPONL 2003. Available at: 1 1 4  htto://www.policeombudsman.org/Publication.cfm^ ar=2003 The Ombudsman's research found that 70% of officers questioned did not believe the Ombudsman approached investigations with an open mind. A study of attitudes of members of the PSNI to the OPONI and the new complaints system: Main findings. OPONL 2004, p2. Available at: htto://www.policeombudsman.org//Publicationsuploads/solresearch.pdf 1 1 5  77  as responsiveness on the part of the Ombudsman herself, given the Unionist critique that she has been over-zealous in her investigation of police officers (the initiative was praised by both the SDLP and D U P ) . 116  Figure 8 shows that a significant proportion of the Ombudsman's work concerns strategic issues, through investigations into complaints about a failure of duty. The subsequent investigations can therefore provide useful information to the police on the factors which prevent them fulfilling their functions properly. The Ombudsman shares trend analysis of complaints with the PSNI. This is used as an early warning system to identify areas where further training is needed or where current police procedures, or a lack of resources, preventing them from fulfilling their mandate. The Oversight Commissioner has praised this side of the Ombudsman's work as a "major step forward in accountability and the management of risk" , and it is an example of how 117  accountability for individual incidents can be linked to accountability for general policy. The Ombudsman issues reports on thematic areas arising from this trend analysis. These support the functions of other accountability holders (such as the Policing Board or DPPs) whose mandate focuses on policy but who do not have the resources or provincewide coverage to carry out such analysis themselves.  5.2 Democratic accountability - The Northern Ireland Policing Board. The Policing Board was established in 2001, replacing the Police Authority of Northern Ireland. It was designed to be a local and consociational model of accountability. All the Body to promote trust in office, BBC Online, 19 March 2004. Report 4, May 2002, p 20. Available at: http://vAvw.oversightcommissioner.org/reports/archiveculeports.asp7page 116  1 1 7  78  members live and work in Northern Ireland and ten of the seats were linked to the Northern Ireland Assembly, to ensure that both the Nationalist and the Unionist communities were represented. The four major parties are entitled to nominate members, in proportion to their strength in the Assembly. The nine independent members were appointed on the basis of their work in the community and their expertise (in areas such as human resources and IT). The Chairman is a professor at the University of Ulster . 118  5.2.1 Public involvement The Policing Board can therefore be viewed as a grand coalition of elites , with the 119  addition of technocratic experts who are there to balance the political struggle which may occur between the political parties when making sensitive decisions. As noted in Chapter 1, consociational arrangements are by their very nature elitist. They exclude the public from decision-making processes on the basis that sensitive and intricate bargaining can only occur within a small group of people who have learned to trust each other.  It is therefore not surprising that public input into the Board itself is limited. The Board meets monthly, and only part of the meeting is open to members of the public. In this section, senior members of the PSNI provide a briefing on a particular facet of policing  Northern Ireland Policing Board website; http://www.nipolicingboard.org.uk Lijphart noted that grand coalitions of elites are not restricted to cabinet government, citing the Dutch Social and Economic Council as an example of a coalition focused on a particular government activity (1977, 32). 1 1 9  79  (two topics in 2004 have been rural policing, and hate crime ). Board Members then 120  question the Chief Constable on any matter they choose. However, members of the public may not ask questions themselves.  These meetings do provide information to the public, both in terms of the responses from the police and the insight gained into the Board's own priorities. However, a member of the public who does not agree with those priorities cannot change them or intervene with their own. Neither can they easily penalize Board members electorally for following an agenda they dislike. Whilst the NIA is suspended, the independent and political members are all appointed by the Secretary of State. However, even when the NIA was in session, Police Board members were not directly elected - the parties are free to nominate any of their MLAs to the Board, as long as they are not Ministers. Members of the public with a particular interest in policing cannot vote for whom they want on the Board; instead they have to cast their vote for a party in the NIA election and hope that this party nominates someone who shares their interests and priorities. Furthermore, it is difficult to assess even this as most Police Board business is dealt with in private session. Although the minutes and agendas are published on the website, points made in discussion are not attributed to individual Members, so it is difficult to ascertain who is advocating what. Furthermore, the agendas are only posted after the meeting takes place, so there is no opportunity for members of the public to ask that a particular issue be put before the Board.  Minutes of Northern Ireland Policing Board Public Meetings, January-May 2004. Available at: http://wrww.nipolicingboard.org.uk/word_docs/minutes_agendas  80  5.2.2 Insulation  from  the  conflict  The operation of the Police Board has been affected by the continued debate between Nationalists and Protestants over police reforms. All four parties delayed joining the Board because there were ongoing disputes over the implementation of the Good Friday Agreement  121  and opposition from Unionist parties to the Patten reforms. SDLP agreed  to sit on the Board in June 2001, and the DUP and UUP joined them in September 2001 . The Oversight Commissioner described this delay as "a critical shortcoming in 122  achieving accountability"  123  as the lack of a Policing Board meant that DPPs could not  be established, a new code of ethics for police officers could not be agreed, nor could procedures for strategic and operational planning be developed, as policing plans have to be agreed by the Board.  To date, Sinn Fein has not joined, as discussed in Chapter 3. This has distorted the political balance on the Board as its seats were re-distributed between the two Unionist parties. As a result, there are now 4 members of the UUP, 3 members from the DUP and 3 members of the S D L P  124  , instead of the 50:50 split between Nationalist and  Unionist political members which was envisaged by Patten. This clearly hinders the Board's ability to work as a consociational institution, as the two segments do not have equal representation (and there is no power of minority veto).  These concerned the lack of decommissioning by paramilitary parties. The two parties made the announcement on the same day, and their leaders had met several times in the preceding months to discuss policing. This suggests a desire to present a united front, in order to further the common Unionist agenda. Unionist leaders meet over policing plan, B B C Online, 20 September 2001. Office of the Oversight Commissioner, 2 Report, December 2001, p 21. Policing Board web-site, http://www.nipolicingboard.org.uk 121  122  123  nd  124  81  There have been further examples of the conflict affecting the work of the Board. Denis Bradley, the Vice-Chair of the Board, and an independent member, has twice been targeted in a wider campaign of intimidation by ERA splinter groups against Catholic members of DPPs . DUP members of the Board have also used the public sessions of 125  meetings to ask the Chief Constable about the backgrounds of various SDLP and Sinn Fein politicians, which suggests that purposive depoliticisation of the policing issue has not yet been achieved . 126  However, despite these problems, there are also many instances of the Board defusing tense situations or coming to agreement on sensitive issues. They reached consensus on a new name, uniform and symbols for the PSNI , despite the fact that some of the 127  harshest (Unionist) criticism of the Patten report was over its recommendations in this area . They also agreed on a new Chief Constable, 128  129  implemented the Patten  recommendations for 50:50 recruitment and pressed the PSNI to reform the Special Branch, which conducts covert operations (and has been accused by Nationalists of collusion with paramilitaries). The Oversight Commissioner has praised the Board for  Attack on home condemned, B B C Online, 13 February 2004 and Hoax device at SDLP man's home, B B C Online, 14 September 2003. ("SDLP man" is a reference to a SDLP councillor who chairs a DPP, who was also targeted). Examples can be found in the minutes of the 29 Meeting (available at http://www.nipolicingboard.org.uk/word docs/minutes agendas/NIPB min_apr04.doc) and the 28 Meeting (available at http://www.nipolicingboard.org.uk/word_docs/public_session/minutes_April%2004.doc). The Chief Constable refused to answer the questions. Board marks NI policing milestone, B B C Online, 5 November 2002. Unionists: Patten Report "flawed", B B C Online, 13 September 1999. This was Hugh Orde, formerly of the Metropolitan police, appointed in mid 2002. This was a controversial appointment because Orde had previously investigated allegations of collusion between the R U C and loyalist paramilitaries in the 1989 murder of Pat Finucaine, a Catholic solicitor. th  1 2 6  th  th  127 128  1 2 9  8 2  helping to resolve contentious issues, and "addressing its responsibilities w i t h professionalism and d e d i c a t i o n . "  130  A notable example o f the B o a r d using its cross-sectional representation to defuse a situation was the stand-off between the Ombudsman and former C h i e f Constable over the O m a g h b o m b i n g report (as discussed i n Section 5.1.2). The B o a r d listened to representations f r o m both sides and appointed a w o r k i n g group ( w i t h a representative f r o m all three parties and an independent member) to consider the actions the B o a r d could take. I n response to the O m b u d s m a n ' s criticisms o f the investigation, it recommended appointing an officer f r o m another police force to oversee the inquiry. H o w e v e r , they stipulated that the P S N I w o u l d remain i n operational control, w h i c h met U n i o n i s t objections that oversight w o u l d threaten the autonomy o f the P S N I  1 3 1  . The  B o a r d approved this suggestion unanimously and the C h i e f Constable and Ombudsman agreed to the compromise. A h i g h l y contentious issue was therefore dealt w i t h i n a w a y w h i c h respected the concerns o f both sides, and w h i c h was also sensitive to the w i d e r political ramifications o f the dispute ( w h i c h was w h y the B o a r d was anxious to get agreement from all members); i n fact a perfect example o f consociational decisionmaking.  5.2.3 Policy competence. Figure 9 shows the issues raised i n recent Police B o a r d meetings, categorised b y competence. I t suggests that the B o a r d has been f a i r l y successful i n p r o v i d i n g a  1 3 0 131  Office of the Oversight Commissioner, 6 Report, December 2002, p26-27. B B C Online, Marathon talks on Omagh investigation, 5 Feb 2002. th  83  balanced message on priorities. All four form more than 10% of the total of queries or discussions, although there is a marked emphasis on fiscal and strategic competence. Interestingly, given that a majority of the Board are politicians, and that the literature states politicians have incentives to focus on medium term outputs, issues of responsive competence were raised the least. This may be because members of the Police Board do not have the same accountability role as a Minister would. They do not have to answer to Parliament for the actions of the police (policing is not a devolved power, so is not controlled by the NIA, and at Westminster the Secretary of State for Northern Ireland answer questions about policing). Furthermore, they are a collective body and do not have operational power over the day to day running of the police in the way a Minister does in a central department. It is therefore unlikely that any political member will be blamed individually if the police do not provide valuable outputs. Unionist politicians in particular are safeguarded from being criticized over policing failures in a way that would damage them politically, as their opposition (i.e. the other Unionist party) sits on the Board with them.  The SDLP is in a more difficult position, as it is under pressure from Sinn Fein over its decision to join the Board. As noted in Chapter 3, its public pronouncements and campaign literature shows a clear desire to be responsive - listing all the changes to the police which should benefit (or please) Nationalists, and linking them to the SDLP's position on the Board. However, the SDLP is in a minority position on the Board (3 out of 19 members), and has in the past shown  84  Figure 9: Issues raised in Northern Ireland Policing Board Meetings. January-June 2004. Private Session 9 5 7  Public Session 2 0 0  Total for each Competence  Regulatory Compliance w/ misc leg or court decisions 6% 2% Tribunals re PSNI officers/former officers 3% 0% Human Rights 5% 0% Individual complaints raised w/ or referred to Board 10 7% 7% 6 17% Sub-total 31 21% 8 10% Fiscal Expenditure 15 10% 5 6% Resource management 41 27% 5 6% Best Value reviews 5 3% 0 0% Partnerships/Service Level Agreements 1 1% 2 2% Training (a human resource) 5 3% 0 0% 34% Sub-total 67 45% 12 15% Strategic Operation of DPPs 9 6% 4 5% General crime levels 0 0% 9 11% Strategic planning processes 3 2% 0 0% Performance management techniques 0 0% 1 1% Anti-social behaviour 4 3% 0 0% Housing Security 1% 2 0 0% Organised/sectarian crime/public order problems 13 9% 12 15% Alternative policing & restorative justice schemes 2 1% 0 0% Evidence preparation (for prosecutions) 1 1% 2 2% Hate crimes (racist/homophobic) 0 0% 9 11% Other types of crime 0 0% 8 10% Rural policing 0% 0 4 5% 36% Sub-total 34 23% 49 60% Responsive to Police Board 6 4% 1 1% to public as a whole 3 2% 1 1% to victims/families 3 2% 8 10% to Police Federation/police officers 1 1% 0 0% to complainants 1% 1 0 0% to young people 2 1% 0 0% 11% Sub-total 16 11% 10 12% Other Specific investigations/individuals (Stormont; Membership of IRA) 2 1% 2 2% Defending police against third party criticism 1 1% Sub-total 2% 2 1% 3 4% TOTAL 150 82 Source: Analysis of Minutes and Agendas of meetings of the Northern Ireland Policing Board, JanuaryJune 2004.  85  an accommodating political style  132  which might explain why its situation does not  seem to have had a marked effect on the Board meetings themselves.  It should be noted that the balance between the competencies shifts significantly in sessions that are public. Issues of strategic competence are mentioned far more in the public meetings, whilst fiscal and regulatory competencies become less popular with Board members in open session. This is not surprising - regulatory issues often concern individual people's complaints so considerations of privacy (and a desire not to jeopardize ongoing police investigations) would encourage Members not to discuss them in public. Presumably they also feel that members of the public who come to a meeting of the Policing Board want to hear about policing (i.e. about crime levels and the action being taken to curb anti-social behaviour or to improve security on housing estates), as opposed to a line by line analysis of a budget proposal or the costeffectiveness of transport workshop sites . A combination of public and private 133  sessions therefore appears to encourage the Board to focus on a balance of competencies, although this comes at the expense of full public participation.  5.3 Summary These case studies show that two accountability holders studied operate very differently, and often complement each other. The Ombudsman involves the public in It was John Hume, the former SDLP leader, who was largely responsible for the beginnings of the peace process in 1993, when he negotiated between Sinn Fein and the British government. (Farren and Mulvihill, 2000,157-159). 1 3 2  Minutes of the 27 Meeting of NIPB, (private session) 5 Feb 04, pi 5 Presentation on Best Value Review Programe.Year 1: 2003-2004. Available at: http://www.nipolicingboard.org.uk/word docs/minutesagendas/NIPB min apr04.doc 1 3 3  th  th  86  her work in a way in which the Board (as a structure based on elites) is unable to do. The work of OPON1 also monitors the regulatory competence of the police, which the Board can only do in private meetings, and without the resources to investigate specific allegations. Conversely, the Board focuses on fiscal and strategic competencies, which are either not in the remit of the Ombudsman, or only a secondary function. Whilst neither organisation has been insulated from the conflict, the Board, as an overtly political and consociational institution, has been more successful in resolving UnionistNationalist disputes.  87  CONCLUSION This thesis has examined four models of accountability, assessing them against theory, comparator cases, elite preferences, and public opinion. When assessed against the criteria of public involvement, insulation from the conflict (or helping to resolve a conflict) and policy competence, it becomes apparent that each model has flaws.  The writings of Lijphart, Nordlinger and Lustick predicted that consociational democratic accountability would exclude the public. This is borne out in the experiences of the Police Board and in the public's reluctance to contact the democratic institutions. The consociational theories also described how majoritarian structures particularly exclude minorities from decisions, which was seen in the lack of police accountability under the 1922-1972 Stormont government. Experience in England and Wales also suggests that central government will try to retain control over police forces, thus undermining local democratic accountability. During direct Westminster rule over Northern Ireland, a remote democratic model effectively excluded the whole of Northern Ireland from making decisions about their own police force. This shows how, in extreme conditions, an agency can develop bureaucratic autonomy without community wide support, because political power resides outside that community.  The rejection of the administrative-political divide by various public management scholars suggested that even technocratic institutions wouldfindit difficult to insulate themselves from conflict in a severely divided society. This could be seen in the Unionist attacks on the work of the Oversight Commissioner. They did not separate matters of policy (whether the PSNI should recruit equal numbers of Catholics and  88  Protestants, for example) from matters of implementation (how efficiently the recruitment agency fulfilled this task, and whether they provided good value for money).  Bozeman and Rosen both predicted that the legalistic model would not be effective in dealing with matters of strategic, responsive or fiscal competence, as it only focuses on two-party disputes. Analysis of the issues dealt with by the Police Ombudsman of Northern Ireland support this as far as fiscal and responsive competence are concerned. However, on strategic competence, the evidence examined does not support Rosen's argument. The Ombudsman's work on trend analysis, which is shared with the PSNI, and her thematic reports suggest that there is scope for legalistic models to move beyond questions of regulatory competence. The Dutch experience with complaints commissioners supports this, and the public also appear convinced - with a clear majority thinking a legalistic institution can "help the police do a good job".  In the case of community based institutions, accountability can again be hampered by conflict. The Unionist parties are concerned that DPPs will be influenced by paramilitaries. Although ex-prisoners have not yet been allowed to join DPPs, they are still affecting their operation through an intimidation campaign against Catholic members. This may harm the work of the Partnerships in two ways. Firstly, it will probably deter some Catholicsfromjoining the DPPs, which could threaten their ability to represent the whole community. Secondly, the individuals who have been targeted have received extra police protection, which might affect their impartiality - presumably  89  it is d i f f i c u l t to p u b l i c l y criticize officers w h o have been protecting your home and family.  A p p l y i n g j u s t one o f these models to N o r t h e r n Ireland w o u l d therefore leave gaps i n the accountability framework. Figure 10, over the page, summarises the evidence in this thesis, s h o w i n g the strengths o f each model. I t shows that although no model is perfect, w h e n assessed collectively the four models do meet all o f the criteria.  C o m m u n i t y and legalistic models meet the first criterion o f i n v o l v i n g the public. I n the case o f c o m m u n i t y based models, this i n v o l v e m e n t is f o r m a l - the public have the opportunity to j o i n institutions w h i c h meet regularly w i t h the local police. The degree o f control w h i c h these groups have varies - i n E n g l a n d and Wales c o m m u n i t y based mechanisms have been advisory, although this m a y change over the next f e w years. The newly-established D i s t r i c t P o l i c i n g Partnerships (DPPs) give the N o r t h e r n I r i s h public a m u c h more concrete role - the local p o l i c i n g p l a n has to be agreed w i t h the D P P , meaning that the public can actually change p o l i c i n g priorities and practices i f they w i s h to. Furthermore, these local plans can diverge f r o m the overall p o l i c i n g plan f o r N o r t h e r n Ireland ( w h i c h is agreed w i t h the P o l i c i n g Board). This allows resources and techniques to vary according to the needs and wishes o f local areas - particularly important i n a d i v i d e d society.  C o m m u n i t y based institutions set their o w n agendas so may focus on both general p o l i c y and i n d i v i d u a l incidents. I n contrast, the legalistic model is more restrictive, being designed to address i n d i v i d u a l complaints about the police. H o w e v e r , this can be  90  Figure 10: Evidence for the strengths of the accountability models Page references are to this thesis; CAS - Community Attitudes Survey.  Democratic Public involvement  Conununity  Technocratic  Legalistic  Feldman and Khademian, p26.  Light p25  Origin of the Dutch complaints system, p42  Dutch complaints system, p42 CAS data, p60-63  Managing the conflict  Encouraging strategic competence  Encouraging responsive competence  Dutch experience, p41. Policing Board experience, p78-80 England and Wales experience, p37  Problemsolving approach, Patten, p31  England and Wales experience, p37  Ombudsman experience, p78  Public attitudes re councils & MPs, p63  SDLP opinion of DPPs, p54  Provide long term view if overseers are career police officers, p29  Dutch complaints system, p42  Policing Board experience, p82-84 Campbell, p30  CAS data, p68-69  CAS data, p62  Dutch and Basque experience, p41 and 44 SDLP view of DPPs, p54 CAS data p63-64  CAS data, p62 (vulnerable groups)  SDLP's view of the Policing Board, p56-  91  57 Encouraging fiscal competence  Encouraging regulatory competence  Campbell, p30 Policing Board experience, p85-86 Spanish experience (Madrid bombings) p 43-44  Patten, p31  If the overseers are auditors p29 Light, p27.  Rosen, p30  Behn and Light. p29  Dutch complaints system, p42 England and Wales public law, p36-37 DUP, SDLP and Sinn Fein, p51-52  an advantage as the model involves members of the public who would not take part in community based models - either because they do not have time, because they are not generally interested in policing issues or because they do not wish to be associated with the police . Such individuals can use the Ombudsman or the courts to address their 134  concerns without having to become involved in policing over the long-term. Less formal models such as an Ombudsman or Complaints Commissioner can be particularly effective in this respect as there are no legal fees, meaning access to the mechanism is not affected by wealth. In addition, both the Ombudsman in Northern Ireland and the Dutch complaints system use mediation to resolve some complaints. This enables the complainant to control the investigation and outcome of the complaint, whereas in court  For example, Sinn Fein members have argued that their opposition to the PSNI would be compromised if they joined DPPs, but they have nevertheless felt able to support the (independent) Ombudsman. l j 4  92  the former is determined by legal representatives and the rules of court procedure, and the latter is decided by the jury and the judge.  As has been discussed, no model appears entirely insulated from severe conflict - not even the "depoliticized" legalistic and technocratic models. This supports the argument of scholars such as Aberbach and Rockman that politics and implementation cannot be completely disentangled. The legalistic and technocratic mechanisms examined are designed to investigate an issue and establish factual conclusions which can be accepted by everyone, regardless of political ideology. However, in Northern Ireland disagreement is so sharp that there is little which is not political. "Facts" are contested for example Unionist politicians disputed the Ombudsman's findings on the Omagh bomb investigation. Alternatively, factual conclusions are interpreted differently - the Oversight Commissioner's reports on the progress in implementing 50:50 recruitment did not resolve disagreement over whether this was desirable in the first place.  However, the democratic model explicitly recognises the political cleavage rather than trying to rise above it and can therefore help to manage this conflict. The experience of the Northern Ireland Policing Board and the Dutch approach to policing show that consociational structures do not necessarily produce consensus. As discussed in Chapter 5, there have been several instances where the Board has divided along sectarian lines. However, as Lijphart and Nordlinger predict, these arrangements allow grievances to be articulated in a "safe" arena (rather than being expressed through violence). Furthermore, they provide a forum for bargaining - the Board's action over the  93  Ombudsman's report on the Omagh bomb is a clear example of a compromise in which sharp divisions of opinion were not reconciled, but were accommodated.  These considerations show that the democratic, legalistic and community based models are all necessary. Turning to the third criterion of policy competence, the four models tend to focus on different aspects of competence. Together they encourage the police to balance fiscal, responsive, strategic and regulatory considerations. The technocratic model is particularly flexible as it can incorporate expertise from several different fields. Institutions such as the Oversight Commissioner utilize experience in law enforcement to assess how efficiently the police use their resources, and their effectiveness in preventing crime - a focus on both fiscal and strategic competence. Auditors obviously focus on fiscal competence, but also assess compliance with accounting standards and procurement rules (regulatory competence) and may benchmark the performance of the police against other agencies (strategic competence). As discussed in Chapter 2, technocratic institutions also provide information to politicians and the general public. This strengthens the capacity of democratic and community based models to assess the competence of the police.  The technocratic model is depoliticized and therefore not designed to assess the responsiveness of the police. This gap is filled by the democratic and community models. The Dutch and Basque experiences show how community based models can ensure that policing meets local priorities, and this is something that the U K Home Office wants English and Welsh policing to emulate. Additionally, politicians (whether as ministers, backbenchers or members of Policing Boards and parliamentary  94  committees) scrutinize whether policing follows their policy preferences and whether police performance will enable the government to keep its manifesto promises.  Finally, the legalistic model scrutinizes individual incidents. This assesses whether the police are complying with the law and with human rights standards - not just at the organizational and policy level but also in their day-to-day activities.  Taken together, the four models provide a comprehensive approach to accountability the public is involved, conflict is managed and a balance of competencies is encouraged. At first sight, the Patten blueprint for police accountability appears contradictory and overly bureaucratic. However, when the models are applied to divided societies, it becomes apparent that this is a reflection of the complexities of accountability itself. The issues explored in this thesis show that accountability, far from being a technical, administrative matter is inherently political, particularly in a society which does not have a shared vision of what it wants from its police.  95  BIBLIOGRAPHY Aberbach, Joel D. and Rockman, Bert A. In the Web of Politics: Three Decades of the US Federal Executive. Washington D.C. Brookings Institute Press, 2000. Allison, Graham T. "Public and Private Management: Are They Fundamentally Alike?" Proceedings of the Public Management Research Conference. November 19-20,1979, Washington D.C. BBC Online, Unionists: Patten Report "flawed". 13 September 1999 B B C Online, Unionist leaders meet over policing plan, 20 September 2001. BBC Online, NI Police Ombudsman's Role. 7 December 2001. BBC Online, Ombudsman and police chief in talks. 18 December 2001. B B C Online, Marathon talks on Omagh investigation. 5 February 2002 BBC Online, Board aims to end Omagh dispute. 7 February 2002. BBC Online, Durkan jabs Sinn Fein on policing, 25 February 2002 BBC Online. Call to join NI policing partnerships., 6 September 2002. BBC Online, Board marks NI policing milestone. 5 November 2002. BBC Online. Unionists "will be driven off boards". 3 December 2002. BBC Online. Chief Constable addresses SDLP. 1 March 2003. BBC Online Policing body "powerful vehicle". 19 March 2003. BBC Online. Ex-prisoner clause "likely". 24 March 2002. BBC Online. SF "close to biting policing bullet". 25 March 2003 BBC Online. Support for Ombudsman "rising", 28 March 2003. BBC Online. SF "hedging bets over policing". 31 March 2003. BBC Online. Policing protest "intimidation". 22 May 2003. BBC Online, MP critical of "wasted money". 27 May 2003 BBC Online, Hoax device at SDLP man's home. 14 September 2003. th  96  BBC Online. DPP Meeting banned. 23 October 2003. BBC Online. Spain to block Basque plan. 31 October 2003. st  BBC Online. Voters can "move process forward". 11 November 2003. BBC Online. Paisley's party tops the poll. 28 November 2003. BBC Online, Continued confidence in NI Policing. 11 December 2003. BBC Online Police complaints made public. 15 January 2004. BBC Online Attack on home condemned. 13 February 2004 BBC Online, Body to promote trust in office. 19 March 2004. BBC Online Policing "needs Sinn Fein support" 27 April 2004. BBC Online. Spain MPs investigate train bombings. 20 May 2004. BBC On-line. Embattled police chief suspended, 2 July, 2004. BBC Online. Autumn talks "very important". 7 July 2004. BBC Online Madrid "rushed to blame Basques". 7 July 2004. Barzelay, Michael. The New Public Management: Improving Research and Policy Dialogue. Berkeley: University of California Press, 2001. Barzelay, Michael and Campbell, Colin. Preparing for the Future. Washington D.C. Brookings Institute Press, 2003. Baxter, Norman S.J. Policing the Line. Ashgate Publishing Ltd, 2001. Behn, Robert D. Rethinking Democratic Accountability. Washington D . C : Brookings Institute Press, 2001. Belfast Agreement (also known as the Good Friday Agreement). Her Majesty's Stationery Office, 1998. Available on-line at: http://www.nio.gov.uk/issues/agreement.htm Bozeman, Barry. Bureaucracy and Red Tape. Upper Saddle River, New Jersey: Prentice Hall, 2000. Brants, Chrisje. "Turmoil and Transition - Problems of Criminal Law Enforcement in The Netherlands." Policing and Society. Vol. 11 (2001) pp 29-66.  97  Brown, Alice; Barnett-Donaghy Tahyna; Mackay, Fiona and Meehan Elizabeth. "Women and Constitutional Change in Scotland and Northern Ireland." Parliamentary Affairs. 55 (2002), pp 71-84. Campbell, Colin. "Juggling Inputs, Outputs and Outcomes in the Search for Policy Competence: Recent Experience in Australia" Governance: An International Journal of Policy and Administration. Vol. 14, No. 2, April 2001, pp 253-282. Campbell, Colin and Wilson, Graham K. The End of Whitehall: Death of a Paradigm? Oxford: Blackwell, 1995. Carpenter, Daniel P. The Forging of Bureaucratic Autonomy. Princeton University Press, 2001. Caygill, H. "Perpetual Police? Kosovo and the Elision of Police and Military Violence." European Journal of Social Theory. 4 (1) (2001), pp 73-80. Democratic Unionist Party. Draft Report on Justice Issues. Belfast: DUP, 1997. Democratic Unionist Party. Democracy - not Dublin rule! Belfast: DUP, 1997. Democratic Unionist Party. Statement to the Independent Commission on Policing in Northern Ireland. Belfast: DUP, 1998. Devenport, Mark. "Battle within nationalism". BBC Online. 11 November 2003. Devenport, Mark. "Unionists neck and neck". BBC Online. 11 November 2003. Dwan, Renata (ed). Executive Policing: Enforcing the Law in Peace Operations. Stockholm: International Peace Research Institute, Oxford University Press, 2002. Easton, Mark. "Control of policing", BBC Online. 2 July 2004. Farrell, M . Arming the Protestants: The Formation of the Ulster Special Constabulary and the Roval Ulster Constabulary 1920-1927. Dingle: Brandon, 1983. Farren, Sean and Mulvihill, Robert F. Paths to a Settlement in Northern Ireland. Gerrards Cross: Colin Smythe, 2000. Feldman, Martha and Khademian, Anne. Inclusive Management: Building Relationships with the Public. Delivered at the 20 Anniversary Conference of the Structure and Organisation of Government Research Committee of the International Political Science Association, June 15-18 2004, Vancouver. th  Her Majesty's Inspectorate of the Constabulary, The Role of Her Majesty's Inspectorate of the Constabulary. HMSO, 2004. Available at: http://www.homeoffice.gov.uk/nmic/hmicrole.pdf  98  Heywood, Paul. The Government and Politics of Spain. Basingstoke: Macmillan Press Ltd, 1995. Home Office, (UK). Policing: Building Safer Communities Together. (Consultation Document) HMSO, 2003. Available at: http://www.policereform.gov.uk/docs/policing_bsct.pdf Independent Commission on Policing in Northern Ireland. A New Beginning: Policing in Northern Ireland (also known as the Patten Report). Her Majesty's Stationery Office, 1999. Available on-line at: http://belfast.org.uk/report.htm. Jimenez, Fernando. "Political Scandals and Political Responsibility in Democratic Spain", pp 80-93 in Heywood, Paul (ed) Politics and Policy in Democratic Spain. Frank Cass publishers, 1999. Kohler, Beate. Political Forces in Spain. Greece and Portugal. European Centre for Political Studies, Policy Studies Institute, 1982. Light, Paul C. Monitoring Government: Inspectors General and the Search for Accountability. Washington D . C : Brookings Institute Press, 1993. Light, Paul C. The Tides of Reform: Making Government Work 1945-1995. New Haven: Yale University Press, 1997. Lijphart, Arend (ed). Conflict and Coexistence in Belgium. Berkeley: Institute of International Studies, University of California, 1981. Lijphart, Arend. Democracy in Plural Societies. New Haven: Yale University Press, 1977. Lijphart, Arend. Patterns of Democracy: Government Forms and Performance in ThirtySix Countries. New Haven: Yale University Press, 1999. ' Loveland, Ian. "The War Against the Judges". Political Quarterly. 68:2 (March-June 1997), pp 162- 170. Lustick, Ian. "Stability in Deeply Divided Societies: Consociationalism versus Control." World Politics. No. 31, April 1979, pp325-344. McGarry, John and O'Leary, Brendan. Policing Northern Ireland: Proposals for a new start. Belfast: The Blackstaff Press, 1999. Nordlinger, Eric A. Conflict Regulation in Divided Societies. Centre for International Affairs, Harvard University, 1972. Northern Ireland Assembly. Official Report (Hansardl. Available at:  99  htfo://www.ni-assemblv.gov.uk/recora^ Northern Ireland Office. Policing in Northern Ireland: Current Arrangements (briefing note). Available on-line at: http://wvvvy.nio.gov.uk/pdf/nipbcurrent.pdf Northern Ireland Policing Board. Minutes of the 27 Meeting of the Northern Ireland Policing Board. 5 February 2004 and 19 February 2004. Available on-line at: http://www.nipolicingboard.org.uk/word_docs/minutes agendas/NIPB_min_feb04.doc (private session) and http^/wvvvy.nipolicingbo (public session). th  th  th  Northern Ireland Policing Board. Minutes of the 28 Meeting of the Northern Ireland Policing Board. 3rd March 2004. Available on-line at: http://www.nipolicingboard.org.uk/word docs/minutes agendas/NIPB_min_Mar04.doc (private session) and http://www.nipolicingboard.org.uk/word_docs/public session/minutes_mar04.doc (public session). th  Northern Ireland Policing Board. Minutes of the 29 Meeting of the Northern Ireland Policing Board. 1st April 2004. Available on-line at: http://www.nipolicingboard.org.u (private session) and http://www.nipolicingboard.org.uk/word_docs/public_session/minutes_April%2004.do c th  Northern Ireland Policing Board. Minutes of the 30 Meeting of the Northern Ireland Policing Board. 12th May 2004. Available on-line at: http://www.nipolicingboard.org.uk/word_docs/minutes agendas/NIPB min mav04.doc (private session) and http://www.nipolicingboard.org.uk/word_docs/public session/minutes May2004.doc th  Northern Ireland Policing Board. Minutes of the 31 Meeting of the Northern Ireland Policing Board. 20th May 2004. Available on-line at: http://vvVvw.nipolicingboard.org.uk/word_docs/minutes_agendas/MPB_mm_m oc (private session only). st  Northern Ireland Policing Board. Agenda of the 32 Meeting of the Northern Ireland Policing Board. 3rd June 2004. Available on-line at: http://www.nipolicingboard.org.uk/word_docs/minutes agendas/NIPB age_iune04.doc nd  Northern Ireland Policing Board. Annual Report, 2002-2003. Available on-line at: http://www.nipolicingboard.org.uk/publications Northern Ireland Statistics and Research Agency. Community Attitudes Survey 19921993 to 2002, inclusive. Data sets. Study description and on-line documentation  100  available at: http://www.data-archive.ac,uk/fmdingData/snDescription.asp?sn=4790 (log-in account needed to access the data set).  O'Rawe, Mary and Moore, Linda. Human Rights on Duty: Principles for Better Policing - International Lessons for Northern Ireland. Belfast: Committee for the Administration of Justice, 1997. Office of the Oversight Commissioner, Overseeing the Proposed Revisions for the Policing Services of Northern Ireland. Report 1. Belfast: HMSO, 2000. Available online: http://www.oversightcommissioner.org/reports/archivedReports.asp?page=reports Office of the Oversight Commissioner, Overseeing the Proposed Revisions for the Policing Services of Northern Ireland, Report 2. Belfast: HMSO, 2001. Available online: http://www.oversightcommissioner.org/reports/archivedReports.asp?page=Teports Office of the Oversight Commissioner, Overseeing the Proposed Revisions for the Policing Services of Northern Ireland, Report 4. Belfast:HMSO, 2002. Available online: http://vv^ww.oversightcommissioner.org/reports/archivedReports.asp?page=reports Office of the Oversight Commissioner, Overseeing the Proposed Revisions for the Policing Services of Northern Ireland, Report 6. Belfast:HMSO, 2002. Available online: http://www.oversightcommissioner.org/reports/default.asp ?page=reports Office of the Oversight Commissioner, Overseeing the Proposed Revisions for the Policing Services of Northern Ireland. Report 9. Belfast:HMSO, 2003. Available online at: http://www.oversightcommissioner.org/reports/default.asp?page=reports Office of the Police Ombudsman for Northern Ireland. Statement by the Police Ombudsman for Northern Ireland on her investigations into matters relating to the Omagh bombing of 15 August 1998. Her Majesty's Stationery Office: Belfast: 2001. Available at: http://www.policeombudsman.0rg//Publicationsuploads/omaghreport.pdf th  Office of the Police Ombudsman for Northern Ireland. Annual Report of the Office of the Police Ombudsman of Northern Ireland. 2002-2003. Her Majesty's Stationery Office: Belfast, 2003. Available at: http://www.policeombudsman.org/Publication.cfm?CatID=3&action==list Office of the Police Ombudsman for Northern Ireland. A study of the treatment of solicitors and barristers by the police in Northern Ireland. Her Majesty's Stationery Office: Belfast, 2003. Available at: http://www.policeombudsman.org/Publication.cfo ge&year=2003  101  Office of the Police Ombudsman for Northern Ireland. A study of attitudes of members of the PSNI to the OPONI and the new complaints system. Her Majesty's Stationery Office: Belfast, 2004. Available at: http://www.policeombudsman.Org//PubIicationsuploads/solresearch.pdf Office of the Police Ombudsman for Northern Ireland. Complaints Statistics. Available on-line at: http://vvvvw.policeombudsman.org/Publication.cfm7Catro Progressive Unionist Party. Submission to the Northern Ireland Office by the Progressive Unionist Party on Policing and Related Matters. Belfast: PUP, 1995. Radin, Beryl. The Accountable Juggler: The Art of Leadership in a Federal Agency. Washington D . C : CQ Press, 2002. Roberts, Alasdair. So Called Experts: How American Consultants Remade the Canadian Civil Service, 1918-21. Toronto: The Institute of Public Administration of Canada, 1996. Rosen, Bernard. Holding Government Bureaucracies Accountable. Westport: Praeger, 1998. Scotsman, The. "Transfer of Police Powers Critical Issue for Republicans", PA News, 29 February 2004. th  Sinn Fein. Policing in transition: a legacy of repression, an opportunity for justice. Belfast: Sinn Fein, 1996. Sinn Fein. A Police Service for a New Future. Belfast: Sinn Fein, 1998. Social Democratic and Labour Party. Policy Paper: Independent Commission on Policing for Northern Ireland. 1999. Available at: http.7/www.sdlp.ie/policy/documents/pdpolicing.shtm Social Democratic and Labour Party. A Manifesto for Equality. Justice and Prosperity. Belfast: SDLP. 2003. Smythe, J. "Symbolic power and police legitimacy: The Royal Ulster Constabulary." Crime. Law and Social Change 38 (3), Oct 2002, pp 295-310. Treasury Solicitor, U.K. The Judge Over Your Shoulder: A Guide to Judicial Review for Government Administrators. 3 ed. Her Majesty's Stationery Office, 2000. Available on-line at: http://www.freasury-solicitor.gov.uk/Publications/judge.pdf rd  Ulster Unionist Party. An Ulster Unionist Party Perspective on the Royal Ulster Constabulary: Policing in Northern Ireland. Belfast: UUP, 1998.  102  United Kingdom Unionist Party. Submission of the United Kingdom Unionist Party to the Independent Commission on Policing in Northern Ireland. Belfast: UKUP, 1998. United Kingdom Unionist Party. Independent Commission on Policing in Northern Ireland: Submission of the United Kingdom Unionist Party. Belfast: UKUP, 1998. Weitzer, Ronald. Policing Under Fire: Ethnic Conflict and Police-Community Relations in Northern Ireland. Albany: State University of New York Press, 1995. Wright, Joanne and Bryett, Keith. Policing and Conflict in Northern Ireland. Basingstoke: Macmillan Press, 2000.  103  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items