Printer Friendly

Inter-agency co-operation and community-based crime prevention: some reflections on the work of Pearson and colleagues.

The role of `inter-agency' co-operation in the sphere of crime prevention has been promoted increasingly by central and local government policy initiatives in recent years. In this paper we consider a number of theoretical issues raised by the examination of power relations in inter-organizational contexts and the definitional processes through which local crime `problems' are identified and translated into policies and practice. The work of Geoffrey Pearson and colleagues represents the pre-eminent contribution to criminological understanding in the field. In this paper we develop a sympathetic critique of their work. In doing so we draw on our own two-year research study of the social dynamics of inter-agency co-operation in a number of metropolitan and shire county community-based crime prevention initiatives.

In the past decade a growing body of opinion has developed in academic and government circles that the most effective delivery of crime prevention emphasizes the co-operation of state agencies together with the public at the local level.(1) The importance of `multi-agency' or `partnership' work has been repeatedly stressed in recent years. The Home Office circulars 8/84 and 44/90, and the `good practice' booklet which accompanied the latter circular (Home Office 1990), have acted as important catalysts in this process. In addition to repeated policy statements, multi-agency work has been promoted through financial incentives at both the national and local levels, to the extent that some forms of funding are now only available through a commitment to partnership work.(2) The `multi-agency' approach has increasingly come to be seen as a panacea for recurring crises within criminal justice. Yet, aside from the plethora of `outcome' evaluation studies (see for example Forrester et al. 1988, 1990) there remains little serious criminological attention given to the broader implications of the multi-agency approach for the processes of policy formation and implementation and the emergence of new corporatist administrative arrangements for local governance. While wary of the dangers of reifying `multi-agency' to the level of a pseudo-science,(3) we wish to argue that there is a need to address and understand the complex social processes involved in inter-agency collaboration and their relation to the ongoing restructuring of the system of crime control and prevention.

The work of Geoffrey Pearson and colleagues stands as the pre-eminent reference point for academic debate concerning the `multi-agency approach' (Sampson et al. 1988, 1991; Blagg et al. 1988; Pearson et al. 1989, 1992). In this paper we intend to subject their research findings to a sympathetic critique. In doing so we draw upon our own experiences of a two-year research study of inter-agency relations in the sphere of community-based crime prevention. The research was situated in a number of localities in and around the Greater London area, stretching from inner-city neighbourhoods to semi-rural county locales.(4)

Pearson and colleagues concede that their research is, only a starting point in the evaluation of what is an unfolding phenomenon. They suggest that:

our work should be regarded only as a first step in attempting to understand in the implementation of multi-agency initiatives, leaving a number of blind-spots and areas of uncertainty. (Pearson et al. 1992: 68)

Whilst our research supports the general thrust of much of their work and confirms many of their fieldwork findings, our own research departs from some of their findings, in a number of important areas. Most importantly, while we share their stated objective, to construct a `more socially nuanced understanding of the multi-agency approach' (Sampson et al. 1988: 480), it is our contention that there are important conceptual problems with their work. In this paper we will attempt to address some of their `blind-spots and areas of uncertainty' within their work and consider the manner in which we depart from their conceptual framework. Finally, we will consider potential explanations for the differing findings and emphases.

Organising Framework

The false and misleading dichotomy of `benevolent' and `conspiracy' approaches to represent existing perspectives

Geoffrey Pearson and colleagues, in reporting their research findings, suggest the existence of two dominant perspectives within the literature. These they term the `benevolent' and `conspiracy' approaches (Sampson et al. 1988: 479-80). The former, they argue, assumes an unproblematic consensus within and between both the local state and communities. It is promoted by those `who advocate', and is essentially `a justification for', the multi-agency approach (pp. 480 1). Thus from within the `benevolent' approach, multi-agency is perceived as a `good thing'. By contrast the `conspiracy' perspective emphasizes the coercive nature of the local state and highlights a sinister interpretation of the multi-agency approach. From within this perspective, multi-agency is a `bad thing'. What one viewpoint sees through rose-tinted glasses the other sees through jaundiced eyes.

We suggest that this framework represents a false and misleading image of the current state of debate, not only because it presents an over-characterization of competing views - which is largely inherent in any such literature review - but because it focuses upon a single interpretive theme which is not the core aspect of contention. Pearson et al.'s dichotomous positions are concerned with whether multi-agency is a `good thing', or not. Their analysis focuses on supposed `motivations', be they potential, hypothetical, unintended, or otherwise. These issues are not unimportant but this approach simplifies the debate. It fails to give due significance to the fact that competing perspectives are premised upon very different understandings of the complex structures, relationships, and interactive exchanges within and between state agencies, communities, and the market, generated and embedded in multi-agency approaches. For example, they conflate Left Realist and `managerialist' perspectives under the `benevolent' heading (Sampson et al. 1988: 481). Yet these two dominant strands within contemporary British criminology embody very different understandings of the nature of structural power relations and interactive exchanges which take place across agency boundaries (Young 1986, 1992; Crawford et al. forthcoming). Rather, we need an organizing framework which highlights those structural and interactive relations. This we have attempted to do elsewhere using recent developments in the conceptualization of `corporatism' as the basis for an organizing framework (Crawford 1993a).

However, the essential problem with Pearson et al.'s framework is that - in attempting to mediate between the `benevolent' and `conspiracy' approaches - it introduces a sharp polarity between conflict and consensus. This polarity is to be found elsewhere within the social sciences (Hunt 1982). The danger, however, is that the analysis remains trapped within the dichotomy and fails to break out of its own terms of reference. The general deficiency inherent in such a dichotomous conception of multi-agency relations is that it has a tendency to result in an unstable framework that oscillates between the polarities set up. Thus it produces an `either or' effect in which analysis is reduced to emphasizing `either' the existence of consensus `or' the existence of conflict in inter-agency relations. While both are clearly important characteristics of, and elements embodied in, relations between agencies, this single interpretive theme creates a framework which is always in danger of reducing consensus/conflict to whether its existence is a `good' or `bad' thing. Despite the important findings and the complexities of the social processes involved in multi-agency work that their research reveals, Pearson et al. fail to advance a coherent alternative perspective which manages to combine, and go beyond, the opposing characteristics of inter-agency relations in the `benevolent' and `conspiracy' approaches, without fluctuating irresistibly between the polarities they construct.

Power Relations in Multi-Agency Work

Pearson et al. correctly identify power as the central aspect in the study of inter-agency relations. Our own research confirms the primacy of power relations in inter-agency work (Crawford and Jones 1993). These power relations, they suggest, exist at `a deep structural level' and are represented through differential organizational power (Sampson et al. 1988). They go on to add:

symptomatic forms of inter-agency conflict, such as struggles over confidentiality and privileged access information. (Sampson et al. 1991: 132)

The existence of deep structural conflicts and structural power differentials between agencies is unquestioned by our findings (Crawford and Jones 1993). As a consequence of their different histories, cultures, and traditions, organizations engaging in multi-agency community crime prevention work pursue conflicting ideologies, strategies, and practices. Organizations bring to `crime problems' competing claims to specialist knowledge and expertise, as well as differential access to both human and material resources. However, while these oppositions are, `always latent' (Sampson et al. 1988: 482) the important question, for our purposes, is how these oppositions relate to, and are embedded in, routinized social action. This, Pearson et al. identify as a second or `surface level' issue (Pearson et al. 1992). While the distinction between these two levels may constitute a useful analytic tool, the important issue concerns the nature of the interactive relationship between these two levels. This analysis would appear to represent an important gap within Pearson et al.'s work. They focus almost exclusively upon instances in which structural conflicts are realized and played out in, and through, human interactions: where structural oppositions appear at the `surface level' in the form of overt conflict. However, the issue of how deeper level oppositions are negotiated in non-overtly conflictual interactions seems to be largely ignored. Most notably, they fall to address sufficiently the management of conflict `off stage' in discrete settings which control their impact upon broader inter-organizational relations and community representation. We suggest that there are a number of inter-related conceptual reasons for this omission. These include first, their tendency to conceptualize power as largely constraining, and secondly, their over-emphasis on the substance of conflict where it is realized, rather than the processes through which conflict is often hidden.

Conceptualization of power

Pearson and colleagues appear to treat power relations as principally divisive and restraining. Their conceptualization of power tends to emphasize power embedded in structures which appear as external constraints upon action. They underestimate the creative and productive nature of action within those structural constraints. Such an approach leaves us only a partial understanding of the operation of power in interorganizational contexts. Our own conceptualization of power draws upon an established line of social theorizing which emphasizes power as both enablement and constraint: as productive as well as divisive (see, for example, albeit from somewhat different standpoints: Lukes 1974, 1986; Foucault 1977, 1978; Giddens 1984). For as Lukes points out:

... Social life can only properly be understood as a dialectic of power and structure, a web of possibilities for agents, whose nature is both active and structured, to make choices and pursue strategies within given limits, which in consequence expand and contract over time. (Lukes 1977: 29)

For us this dialectic highlights the relational nature of power. Differential power relations encompass the relative capacity of organizations and actors, drawing upon material and human resources, to achieve desired outcomes. In a multi-agency crime prevention context this often involves the ability of different organizations and actors to impose their definition of a situation upon others and to realize their strategic interests. Giddens notes that, `power is the means of getting things done and, as such, directly implied in human action' (Giddens 1984: 283).

The multi-agency approach to policy negotiations at the local level is intrinsically concerned with `getting things done'. While structural constraints limit and demarcate the range of possibilities, specific outcomes emerge through the often creative management of struggles around strategies, practices, and definitional conflicts. Thus power in multi-agency relations is not purely realized in terms of constraints but is also creative. This creativity is an important aspect of working within the tensions and oppositions that exist between agencies and that are the product of differing organizational practices, cultures, priorities, and management structures. By creativity we are referring to the ability of senior managers and front-line inter-agency workers to manage the tensions and challenges posed by working across agency boundaries while securing, albeit limited, goals and strategies. As Rutherford suggests, in his recent study of criminal justice practitioners, one perennial challenge faced by such criminal justice personnel is,

how to work creatively with the inherent tensions, affording legitimacy to both independence and interdependence. (Rutherford 1993: 126; emphasis added)

Such tensions do not only constrain the deployment of human and material resources but provide opportunities for new modes of working. Rather than differing organizational perceptions resulting in a situation in which `different interest groups pass each other like ships in the night' (Sampson et al. 1988: 488), our research findings suggest that there is great creativity among inter-agency workers in negotiating the deep structural conflicts and oppositions that exist.

At this point it is important to stress that the notion of creativity, as we use it, carries no normative judgement of its moral value. It is not to be treated as a necessarily positive evaluation of innovative working practices. On the contrary, as we hope to show, many forms of creativity evidenced in our research resulted in working practices which were frequently unaccountable, often took place behind the backs of agency and community representatives and posed complex questions for procedures aimed to ensure confidentiality. It is the nature and consequences of these creative practices, and in particular their implications for organizations' own strategies and interests, which are of paramount importance.

An example from our research illustrates this point. The Tenmouth Burglary Project, located in a satellite town outside the Greater London area, was initially conceived as a replication of the successful Home Office sponsored Kirkholt Project (Forrester et al. 1988, 1990). Over a two-year period two full-time officers were seconded from the probation service and the police to collect information on the nature, extent, and impact of burglary within the locality. This included surveys of apprehended offenders, burglary victims, and their neighbours. The seconded officers were overseen by a multi-agency steering committee with representatives from all the major voluntary and statutory agencies, as well as representatives of local business and community organizations. The initial aim had been to agree a corporate plan of crime prevention measures for implementation during the life-time of the project. As the data piled up, they reached a series of impasses over the extent of the recommendations to be jointly agreed and the nature of their implementation. Issues such as how to address the lack of support for victims (despite the efforts of the local Victim Support volunteers), the perceived need for improved security measures in private rented as well as local authority housing, and the form that 'community building' in the area should take, remained unresolved, as did questions concerning who was to fund improvements and resource any new services. The senior police and probation officers involved in the management of the project, when confronted by these problems as well as the pending withdrawal of commitment, by their organizations (as the original timetable for the project was drawing closer), jointly renegotiated the initiative's objectives. Over a drink it was transformed from a practical crime prevention initiative involving the implementation of locally researched recommendations into a research project alone, with no formal implementation stage.

The redefinition of the scope and nature of the Tenmouth project enabled the police and probation service not only to `save face' but to pursue with renewed vigour the notion that what was being done was genuinely about `the community' and in their interest. Thus the redefinition of the project was later `sold' to other members of the inter-agency committee in terms of the recommendations being `handed over to the community' (both phrases were frequently used by senior and front-line officers on the project). Thus the implementation stage was to be the community's responsibility and no longer that of the key agencies involved in the project, for whom the project was to be `put to bed' (Police Chief Inspector, Tenmouth constabulary, informal conversation). Later, at a public launch and to great media acclaim, the findings of the two-year project were duly `handed over to the community'.

In addition to the ability of actors to manage creatively without necessarily resolving - the structural conflicts which present themselves, this example illustrates a further point. Multi-agency community interventions are made up of two layers of fundamental relationships: those between the agencies and those between the agencies and the local communities. It was in relation to the latter that the police and probation officers were able to find `common ground' and to agree on a `solution'. An important objective driving both the police and probation service's involvement in a multi-agency community crime prevention project, in this instance (and in many others besides), was to be seen to be `doing something' and to address any perceived lack of legitimacy in the eyes of the local community and, through publicity, the public generally. Not only are legitimacy and power closely connected properties of social organizations (Sparks and Bottoms 1992: 4), but the extent of legitimacy is a vital component in inter-organizational relations, particularly where they take on the involvement of `community representation'.

As in the above example, power differentials between statutory agencies and channels of community representation constitute an essential locus in the exercise of power in community-based crime prevention projects. The Chair of the Tenmouth Burglary Project (a Victim Support co-ordinator) graphically summarized this power imbalance by describing the gulf between agency professionals and community representatives in the following terms:

professionals are people with the capacity to do things and the community is there to have things done for them.

Yet the agencies involved have different conceptions of what constitutes `community' (Currie 1988) and competing understandings of the nature and role of community representation and participation. `Community' fits into their specific crime prevention and more general organizational strategies in separate but overlapping ways. For example, police and probation officers bring very different notions of community to their work. For the former the traditional individual and client-centred approach to their work renders `community' immediately problematic. Community is something that their clients often feel estranged- and disconnected from. For probation officers `community' may be something that their clients `lack', but for them to say exactly what that constitutes is a more complex question altogether. Probation officers are consequently often less comfortable not only with the notion of community but also with community representatives themselves. For police officers, by contrast, community resents a taken for granted, although in reality a highly specific, reference group. The `community' is often referred to as symbolic legitimacy for aspects of police work (Morgan 1989; Fielding and Fielding 1991). But as Pearson and colleagues note, this concept of community is bound up with notions of `the respected and the respectable' (Sampson et al. 1988: 489). Despite these competing conceptions of community the police and probation service in addition to other agencies are bound together in a much wider policy agenda in which `community', in all its guises, plays a prominent role. Community has become an amorphous site between public and private spheres in which the politics of privatization, voluntarization, and local governance coalesce.(5) While notions of representation, participation, and community are crucial to a broader understanding of the work of multi-agency community-based projects, they lie beyond the immediate scope of this article. Nevertheless, what is clear from our research is that the failure to confront the problems of correspondence between `community representatives' and the communities they represent, remains a fundamental problematic in the democratization of multi-agency initiatives.

Returning to the creativity of practice, our findings suggest that this more often takes a pragmatic and managerial form. It tends to be about `getting through the day's business' and meeting managerially defined objectives, rather than pursuing any moral or political mission. For example, at the Arlington Racial Harassment Monitoring Scheme, one of the officers from the local authority was prevented by her own organization from revealing any information about the number of incidents of racial harassment between council officers within her department. The way in which she managed the tensions between the demands for confidentiality from her own organization and the wishes of the multi-agency forum, allowed her to maintain her active involvement in the forum whilst appearing to follow her organization's guidelines. She achieved this by providing `general information' rather than `specific detail' about the number of racial incidents reported and by asking for this information not to be officially recorded in the minutes. This example illustrates the way such tensions are managed through a negotiated process of pragmatic compromise. Compromise often takes place informally and in this instance unobserved by the host organization. This presents problems for accountability as well as for formal safeguards of due process concerning confidentiality.

A further consequence of the notion of creativity, as we have used it, is that it implies a certain degree of reflexivity on the part of the actors involved. It implies that management and front-line workers are (to an extent) aware of, and monitor, the latent sites of inter-agency tension and conflict. Thus Pearson et al.'s repeated implication that multi-agency workers have a `naive view of consensus', whereby cooperation is easily attained and consent is non-problematic (Pearson et al. 1992: 7 1), is not a finding that our research unconditionally supports. We should not confuse an `ideology of unity' (Crawford 1993a: 20) among inter-agency personnel, which to some degree is rooted in the very idea and existence of multi-agency forums, with a `naive view of consensus'. The former represents an ideology in which conflict is viewed as time consuming and wasteful of effort and in which appeals to consensus are the established credo. It does not presuppose a consensus but rather that `unity' is the desired goal, to be achieved with little regard to procedure or the voicing of conflict. It is thus an ends-oriented rather than a means-oriented ideology. It is often the `quest for unity', even if that remains at the level of appearance only, which informs many of the strategies revealed in our research. The latter, by contrast, suggests a degree of ignorance among inter-agency workers which we find difficult to sustain.

In contrast to Pearson et al., our research suggests that workers are often well aware of potential conflicts. For example on the Westbridge drugs-related project, a council officer devised a complex series of strategies in order to recruit the local health authority to support his own objective of the limited availability on prescription of certain street drugs for local addicts. Far from assuming that their support would be immediately forthcoming he explicitly worked from the premise of conflicting priorities. He devised a high risk strategy in which he spent some time building up what he referred to as an `arsenal', to use against them in order to ensure their supportive involvement. This `arsenal' largely comprised activities which took place unbeknown to, and outside of, the formal multi-agency committee and included a local press campaign criticizing aspects of the health authority's policies. While this strategy yielded a degree of success, the council officer was well aware of the fact that it could have resulted in more entrenched conflict.

While our research revealed only a limited number of examples in which conflict was managed in such a deliberate and utilitarian manner, nevertheless agency workers frequently adopted similar but more everyday routines. This frequently resulted in strategies of conflict avoidance, which we shall now consider.

Conflict avoidance

Largely as a consequence of their conceptualization of power, Pearson et al. overemphasize instances in which overt conflict is realized in inter-agency relations. This is not to suggest that conflict is not a central issue in multi-agency relations. It is rather to suggest the need to interrogate the nature of the relationship between structural conflicts and the routinized and taken-for-granted character of day-to-day activities of inter-agency social relations. Given the structural conflicts identified, what often needs explaining is the absence of overt conflict. The exercise of power in inter-organizational contexts does not necessarily manifest itself in and through conflict. In inter-agency forums conflict is often defined away, avoided, or circumvented. What is often most striking, given the deep structural oppositions that exist, is the absence of overt conflict at

An example drawn from our fieldwork should suffice to illustrate the way in which conflict can be avoided by being defined away. At a meeting of the Arlington Racial Harassment group, the various agency representatives were unable to devise an agreed set of categories for the variety of ethnic minority groups resident in the borough. The agencies present at the meeting had different practices for recording the ethnicity of the victim and offender and used different terms to describe social groups. The police, for instance, had six categories of ethnicity as against the local authority's nine. It was suggested by a representative of a particular ethnic minority group that victims should be allowed to self-identify the ethnic group to which they, felt they belonged. This was dismissed as `impossible' by the police, as for them to do so would contravene their own policy, might `offend the victims' and could `create a perception of police racism'. These organizational differences were side-stepped by the `agreement' that it should be left to individual agencies to code the ethnic origin of victims and offenders in accordance with their own criteria. During the period of our fieldwork, the issue was never again raised, given its likelihood to give rise to `unnecessary conflict'.

The consequence of this `non-decision' was that the dominant power of police recording practices remained unchallenged and unchanged. In other instances from our fieldwork, items were omitted from agendas or not recorded in the minutes. Further, certain problematic issues were managed `off stage' in exclusive and discrete settings which control the impact upon the rest of the group (this we return to below). As Lukes notes, the power of non-decision-making is an important element in the exercise of power itself (Lukes 1974). Thus `doing nothing', while appearing to be nonconflictual, may also be an important aspect of power relations between agencies.

Conflict as pathological

There is an ever present danger, when prioritizing overt conflict in the way that Pearson et al. do, that conflict will be understood as essentially pathological. While they draw back from this view, with a disclaimer to the contrary (Sampson et al. 1988: 482), there is a sense in which whilst reading their work, one is left with the unwritten conclusion that conflict hinders effective multi-agency relations and, therefore, for advocates of the multi-agency approach, conflict is considered to be `a bad thing'. It is precisely this perspective which we refer to earlier as the `ideology of unity'. However, conflict is not only socially constructive (Christie 1977), but is a necessary and important part of inter-organizational relations. For as Rock notes, it is `independent interdependence' which, while giving rise to much inter-agency conflict, is at the same time `the weak force which binds the criminal justice system together' (Rock 1990: 39).

Conflict does need to be recognized, as Pearson et al. state, but more than that, it needs to be constructively managed, through the open exchange of competing perspectives. The central problematic in multi-agency relations is not merely the existence and recognition of conflict, but the manner in which conflict is subsequently managed.

The Problem of Informality

Pearson et al. arrive at an acknowledged contradictory conclusion regarding what they call `the question of hierarchy: formality and informality' (Pearson et al. 1992: 63). They state that,

It must be frankly acknowledged, however, that there is a significant contradiction within our research evidence. On the one hand, informal systems of inter-agency working and information exchange are risky encounters which can endanger important confidentialities and might even sometimes constitute a threat to civil liberties. On the other hand, more informal and fluid systems of inter-agency relations seem to offer a more workable basis for communication and negotiation. (pp. 64-5).

This contradiction is particularly evident in their discussion of gender relations (Sampson et al. 1991). Their research revealed gender to be an important site of' informal inter-agency networks. This finding is not in question, particularly given the way in which gender is constitutive of organizations such as the police, probation, and social services. In our own research we were also confronted by evidence as to the importance of gender and other forms of social identity to inter-agency relations. However, Pearson et al.'s desire to celebrate such bonds of `mutually supportive inter-agency relations among women workers' leads them to an implicit celebration of the `informal' (Sampson et al. 1991: 132). Given what we have already said about the importance of the constructive management of conflict, our research suggests the need to be more unequivocal. As Atkinson (1982) has noted, there is a persistent danger among critical writings of celebrating the `informal', while to describe social arrangements as formal is seen as synonymous with criticizing them. This is a trend which critical writers need to, and are beginning to, address (Matthews 1988). While we do not dispute that informal settings may be more conducive to `getting things done' at the level of practice, the problematic issues remain the values which inform that practice. Informal contexts leave differential power relations unchecked, hide decision-making processes from any review and remove them from any democratic input or control. We are not seeking to replace a celebration of the informal with a celebration of the formal. Rather we are suggesting that in community-based inter-agency relations where organizations may be pursuing very different priorities and objects, where they may have competing notions of `crime', `locality', and `appropriate interventions' and where high levels of mutual mistrust exist (Thomas 1986; Holdaway 1986; Crawford 1993b), there is a need for the management of conflicts to be delineated by local community demands within a framework which accords to notions of social justice. Our research experiences suggest that this is less likely to occur in informal settings.

A few examples from our fieldwork illustrate this point. On a number of occasions, on the Illsworth estate project, we witnessed officers from the local authority and the police discussing `problem families', by identifying youths and homeless people on estates. When this was raised at meetings, officers described the people in neutral terms and did not identify people by name. However, both council and police officers were observed taking time before and after official meetings to share information more casually. In one instance, a housing officer discussed with the local Home Beat Officer a `need assessment' interview with a homeless couple who were squatting on a nearby estate. In breach of the housing officer's own guidelines on confidentiality, hearsay information was exchanged which may have mitigated against the couple receiving any housing priority.

As our research was focused on proactive crime prevention work, we found informal relations tended to relate to the process of policy formation, rather than the particular interests of `clients' or individuals. Interestingly, where workers had been seconded to a project from their organization, a second line of accountability developed where senior officers would meet with the seconded officers to discuss the direction of their work. Officially, the workers were accountable to the steering committee, which was the organization to which they were seconded. However, these informal meetings were not mentioned or reported at the official committee meetings. Such unaccountable informal relations between agency representatives sometimes took place at crucial and strategic moments in the life of a crime prevention project. In the Tenmouth Burglary initiative, for example, senior representatives from the police and probation service circumvented the agreed formal procedure of regular meetings. Here, the senior officers arranged an informal meeting in the police bar where, in a `frank discussion', they put their `cards on the table' in an attempt to address a series of problems the project was facing. At this meeting the two senior managers agreed to end involvement in the project several months earlier than expected. This was a discussion to which the other agencies, most notably Victim Support and the local authority, were not privy. Their own particular concerns about the project were given no voice as the change of policy was presented as fait accompli. The contents of the discussion were never subsequently placed on the forum's agenda nor referred to in an open forum meeting. Importantly this meant that the community representatives were denied any say in this major shift in the direction of the project. Thus the two officers involved were able to control the impact of their joint decision and limit any resultant conflict, particularly from the community representatives who were subsequently to be asked to resource much of the future crime prevention work.

In this example, as in others, informal inter-agency relationships tend to increase power differentials between agencies. Access to the relevant informal settings itself becomes a powerful resource. Those excluded from such informal discussions are consequently marginalized. For those included in the discussions, social power is left largely unregulated and unlimited. Further, in proactive community crime prevention initiatives, members who are uninvited to participate in informal negotiations may be unaware that these have even taken place, and of the outcomes and impact of the decisions that have been made. This is because the initiatives are often vague, unlike case-based work, allowing changes to be made to the direction and objectives without all members realizing that this has happened. An extreme example of this took place at the Illsworth Crime Reduction initiative. Due to cut-backs in local authority funding, the council officers slowly began to close down the project, removing the assigned worker to another crime prevention project which had maintained its funding. However, the council officers did not wish officially to close down the project because this, they felt, would not be politically expedient. One council officer commented:

Politically, we can't wind it up, because [the Chair, a senior local councillor] won't let us. He doesn't want to be seen to be not providing something in his area. It's a waste of time if there is no money - and there is no money ... It is just not important any more. We'll let it die ... slowly stop arranging meetings and the like. (Council Officer, informal telephone interview)

A related problem lies in the way in which Pearson et al. appear unnecessarily to conflate organizational hierarchy with (in)formality. We have preferred to maintain an analytic distinction between organizational hierarchy and what we prefer to refer to as the `relational form' (Crawford and Jones 1993). It is useful to maintain these as distinct axes along which sites of inter-organizational linkages can be located. While the two often dovetail and are interrelated, the need for greater formality does not necessarily require that those inter-agency relations occur at a different hierarchical level. Rather, formality means both the extent to which arrangements differ significantly from everyday routines and conversational interactions and the increased requirement of certain procedures and safeguards which regulate the basis upon which networks take place. This requires formal procedures which identify, nurture, and render conflict visible, allowing the constructive management of conflicts. This can and should be located at the level of front-line workers - where more productive relations may exist - as well as at senior management level.

Intra-Agency Relations

The nature of inter-organizational relations is in part determined by the relations within specific organizations. For, as Pearson et al. suggest correctly:

what often appear as inter-organisational conflicts are sometimes more appropriately understood as intra-organisational conflicts (Pearson et al. 1992: 65).

Importantly they identify the role of gender (Sampson et al. 1991) and organizational hierarchy (Pearson et al. 1992: 60). However, they fail to extend this insight further to include the diverse social identities that people bring to an organization, which affect organizational behaviour and action. Similarly, horizontal divisions of tasks, resources, and the status perceptions to which they give rise within organizations can also generate conflicts which seriously affect inter-agency relations. For example, the Tenmouth Burglary Project was caught up in conflicts which had their origins in the internal police perception of inter-agency crime prevention work. In order to invert the prevailing caricature of crime prevention work within the police - so aptly described by Harvey et al. as `a sort of pre-retirement course for experienced but tired detectives' (1989: 88) - the senior police officer involved in the project recruited, as the seconded officer, someone recognized throughout the local force as destined for higher office. He had hoped that the association of such a `high-flyer' would raise the profile of inter-agency crime prevention within the force. However, over time it became clear that in order for the seconded officer to proceed upwards within the organization, she needed to spend some time in the CID. Thus the senior police officer reluctantly let her go from the project in order not to stand in her promotion path. The subsequent withdrawal of a central and prominent figure in the project precipitated a immediate crisis at the inter-agency forum in which the probation service threatened to withdraw their seconded officer. Future relations became strained as the change of personnel was felt to be a down-grading of the general police commitment to the project.

This example highlights that the relationship between front-line inter-agency workers at the periphery of an organization and core agency officers should not be taken for granted. Our research suggests that as a result of intra-organizational tensions, front-line inter-agency workers are often unrepresentative of the membership of a given organization. As Sampson et al. suggest, gender is an important constitutive factor in this regard. For example, within the police forces in our research sites, there was high over-representation of women police officers working on multi-agency projects. This was particularly evident where more `innovative' projects required the police to operate outside of - as opposed to projects which were more `accommodating' of - traditional core organizational structures, roles, and practices (Crawford and Jones 1993). This is partly explained by the fact that this work is perceived by many ordinary police officers as `social work' and `not real police work' and therefore `women's work' (Crawford 1993h: 87-90).

In addition, our research suggests that the ability of front-line workers to go beyond their professional roles and to develop inter-personal trust relations with `outsiders' may be an important characteristic of certain more effective inter-agency personnel. For example, in the Tenmouth initiative, the ability of the probation officer to `get on' with police officers in the canteen was identified as an important aspect in the early success of the project. This was certainly not something that the majority of her probation colleagues either would have been able or willing to do. A senior police officer involved in the initiative stressed this point:

I think the way [the project] was accepted at Tenmouth Police Constabulary was due not just to [the police officer] but also the way the probation officer had got on well with the police. When you are working with other agencies you do need the ability to switch into their culture, know what makes them tick and understand it. [The seconded probation officer] would often come in for coffee or lunch and would sit in the [police] canteen and wouldn't sit with me, but would go and sit over on that table with two or three constables, and that had quite a beneficial effect on, not just publicising the project, but here's a probation officer who hasn't got horns coming out of the side of their head or something like that ... This actually helps. (Chief Inspector, Tenmouth Police, interview)

This highlights the importance of inter-personal trust relations which go beyond mutual prejudices and misperceptions (Crawford 1993b: 82-4). In this instance, the ability of the probation officer not to appear hostile to police officers as a member of the `opposite side of the camp' (police constable, interview), was an important aspect of effective inter-agency work.

Explaining the Differences

Let us now turn to consider some of the potential explanations for the different findings and emphases between the two research projects. First, our research was focused solely on multi-agency relations in the context of proactive community-based crime prevention. We were not concerned with inter-agency contacts which revolve around reactive `case-based' or `client-driven' work. This stands in contrast to Pearson et al., whose research was much broader and included reactive forms of intervention, i.e. courtbased initiatives and case conferences. The agencies working in community crime prevention are unfettered by formal legal regulation or constraint. Thus administrative discretion and the `administration of policy' are the order of the day. Consequently, inter-agency relations are more open-ended. This meant that the initiatives in our research involved less certainty and were less directed. The lack of focus potentially creates greater space in which actors can work creatively, avoiding overt conflicts, sidestepping competing objectives, or even subsuming them within the open-textured nature of the initiatives. Often this involved not actually doing or achieving very much at all. This allows unrestrained power relations a dominant position. It leaves more powerful agencies with little constraint in determining the substance and form of interventions. The open-ended nature of many multi-agency initiatives allows agencies potentially greater space in which to pursue their own objectives and agendas in the formation and implementation of policy, and in which to exert their own organizational ideologies. Secondly, there were differences in research methodologies used. In order to take seriously the relationship between intra- and inter-organizational conflicts, we felt that appropriate research methods were needed to address the internal dynamics of the relevant organizations, and the nature and extent to which they reference inter-organizational relations. For the purpose of this kind of institutional analysis, we were principally interested in larger aggregates than could easily be accommodated within qualitative ethnographic terms. In this regard we used survey techniques to examine the nature of the predominant perceptions and experiences of crime prevention work within different organizations, as well as the level of informal inter-agency relations among non-designated front-line officers (Crawford 1993b)'. Given the size of organizations like the police and probation services, a survey of experiences and attitudes of all officers contributed significantly to the research in a number of ways. First, quantitative methods enabled us to draw on a broader base of data, which helped check on the accuracy of interviews with key workers as well as our own impressions and anecdotal evidence. In this way methods of `counting' acted as a supplement to qualitative data, enhancing their authority. This is a valuable goal even if it does come at the expense of imposing a `fixity on social life that it does not in fact have' (Giddens 1984: 330). A survey also allowed us to address specific issues concerning intra-agency relations. This includes the nature of internal hierarchical relations and predominant perceptions within the different organizations. It enabled us, for example, to identify and explain the marginality of many front-line inter-agency workers from their agency colleagues.

Finally, Pearson et al. correctly recognize that multi-agency work is a `rapidly shifting terrain' (1992: 68). In the time since the mid-to-late 1980s, when Pearson and colleagues conducted their research, a great many more multi-agency initiatives have emerged. Similarly, many lessons have been learnt and disseminated. In many senses the pertinent political questions today are different from those of a decade ago. It seems less relevant to be asking whether or not the multi-agency approach should be pursued, than to question its implications and implementation, and to highlight problems which derive from the inherent tensions between independence and interdependence between agencies and the manner in which conflict is managed. The changes that have occurred over time have meant that the early naivety (which undoubtedly existed) has been, in part, superseded. This may be evidence of the impact and effects of Pearson and colleagues' own research (as well as that of others) on practitioners. As such this illustrates the reflexivity of criminological knowledge, whereby academic research has re-entered the universe of actions and helped constitute those actions.


It has been our intention to highlight a number of conceptual and empirical differences between our own research and that of Pearson and colleagues. In doing so we have not sought to undermine the insights they provide but rather to extend in a particular direction the kind of socially nuanced approach that they articulate. In addressing some of the `blind spots' and areas of uncertainty within their work, we hope to provide.

Our research suggests that when analysing inter-agency relations there is a need to give due emphasis to the creativity of human actions within structural constraints, and for greater specificity of analysis of power relations. Our research findings highlight a tendency within inter-agency arrangements whereby conflict is not overtly confronted in public through constructive negotiation. Rather, conflict often appears to be `defined out' of existence or circumvented. The avoidance of conflict in multi-agency forums merely leaves deep structural conflicts and power relations unaddressed. Thus the power of non decision-making remains an important strategy in multi-agency relations. Similarly, the informal management of disputes in multi-agency settings presents real problems regarding the open and democratically accountable processing of inter-agency and inter-group interests.

The central problematic within multi-agency relations is, therefore, not only the existence and recognition of conflict, but the manner in which conflict is subsequently managed and regulated. There is a need for room to be made in inter-agency relations for constructive debate concerning the competing contributions, priorities, and aims of the agencies involved, as well as for the expression of intra-communal conflicts. Conflict may be the healthy and desirable expression of different interests. Mutual recognition of difference represents a more preferable premise for inter-agency relations than either an assumed consensus or an ends-oriented `quest for unity'.

Our research leads us to the conclusion that the manner in which inter-agency and inter-communal conflicts are presently managed in multi-agency forums increasingly resembles neo-corporatist arrangements for the administration of crime control, in which invisible and unaccountable discretion is the order of the day. In this context power differentials remain largely unregulated between agencies. This impacts upon local policy definitions of `locality' and `community' as well as crime `problems' and their `appropriate' social intervention. It results, we would suggest, in the prioritization of certain kinds of crime and forms of intervention at the expense of others which remain all but silenced. By contrast it is important that conflict is negotiated in an open and accountable manner which recognizes and appropriately compensates for power differentials. Such a process needs to be structured by both local democratic demands and adherence to notions of social justice which protect the rights of those against whom crime prevention initiatives are directed.


Atkinson, J.M. (1982), `Understanding Formality: The Categorization and Production of "Formal" Interaction', British Journal of Sociology, 3/1: 86-118. Blagg, H., Pearson, G., Sampson, A., Smith, D., and Stubbs, P. (1988), `Inter-agency Cooperation: Rhetoric and Reality', in T. Hope, and M. Shaw, eds., Communities and Crime Reduction. London: HMSO. Christie, N. (1977), `Conflicts as Property', British Journal of Criminology, 17/1: 1-15. Crawford, A. (1993a), Corporatism and Criminal Justice at the Local Level. Unpublished paper presented to Socio-Legal Studies Conference, University of Exeter. - (1993b), Crime Prevention and the Multi-Agency Approach: A Study of the Attitudes and Experiences of Police and Probation Officers. Hatfield: University of Hertfordshire. Crawford, A., and Jones, M. (1993), Researching Inter-Organisational Contexts: A Study of Community-Based Crime Prevention. Unpublished paper presented to British Sociological Association Conference, University of Essex. Crawford, A., Jones, T., Pearce, F., and Young, J. (forthcoming), Crime in the Inner City. London: Routledge. Currie, E. (1988), `Two Visions of Community Crime Prevention', in T. Hope and M. Shaw, eds., Communities and Crime Reduction. London: HMSO. Fielding, N., and Fielding, J. (1991), `Police Attitudes to Crime and Punishment', British Journal of Criminology, 31:1, 39-53. Forrester, D., Chatterton, M., and Pease, K. (1988), The Kirkholt Burglary Prevention Project, Rochdale. Crime Prevention Unit Paper 13. London: Home Office. Forrester, D., Frenz, S., O'Connell, M., and Pease, K. (1990), The Kirkholt Burglary, Prevention Project, Phase II. Crime Prevention Unit Paper 23. London: Home Office. Foucault, M. (1977), Discipline and Punish. Harmondsworth: Penguin. - (1978), The History of Sexuality. Harmondsworth: Penguin. Giddens, A. (1984), The Constitution of Society. Cambridge: Polity Press. - (1987), Social Theory and Modern Sociology. Cambridge: Polity Press. Harvey, L., Grimshaw, P., and Pease, K. (1989), `Crime Prevention Delivery: The Work of Crime Prevention Officers', in R. Morgan and D. J. Smith, eds., Coming to Terms with Policing. London: Routledge. Holdaway, S. (1986), `Police and Social Work Relations: Problems and Possibilities', British Journal of Social Work, 16: 137-60. Home Office (1990), Partnership in Crime Prevention. London: Home Office. - (1992), Partnership in Dealing With Offenders in the Community: A Decision Document. London: Home Office. - (1993), Police Reform: A Police Service for the Twenty-First Century, Cm. 2281. London: HMSO. Hunt, A. (1982), `Dichotomy and Contradiction in the Sociology of Law', in P. Beirne and R. Quinney, eds., Marxism and Law. New York: John Wiley. Lukes, S. (1974), Power: A Radical View. London: Macmillan. - (1977), Essays in Social Theory. London: Macmillan. - ,ed. (1986), Power. Oxford: Basil Blackwell. Matthews, R., ed. (1988), Informal Justice? London: Sage. Morgan, R. (1989), `Policing by Consent: Legitimating the Doctrine', in R. Morgan and D. J. Smith, eds., Coming to Terms with Policing. London: Routledge. Morgan, J. (1991), Safer Communities: The Local Delivery of Crime Prevention Through the Partnership Approach. A Report of the Standing Conference on Crime Prevention. London: Home Office. Pearson, G., Sampson, A., Blagg, H., Stubbs, P., and Smith, D. (1989), `Policing Racism', in R. Morgan and D. J. Smith, eds., Coming to Terms with Policing. London: Routledge. Pearson, G., Blagg, H., Smith, D., Sampson, A., and Stubbs, P. (1992), `Crime, Community and Conflict: The Multi-Agency Approach', in D. Downes, ed., Unravelling Criminal Justice. London: Macmillan. Rock, P. (1990), Helping Victims of Crime: The Home Office and the Rise of Victim Support in England and Wales. Oxford: Oxford University Press. Rutherford, A. (1993), Criminal Justice and the Pursuit of Decency. Oxford: Oxford University Press. Sampson, A., Stubbs, P., Smith, D., Pearson, G., and Blagg, H. (1988), `Crime, Localities, and the Multi-Agency Approach', British Journal of Criminology, 28: 473-93. Sampson, A., Smith, D., Pearson, G., Blagg, H., and Stubbs, P. (1991), `Gender Issues in Inter-Agency Relations: Police, Probation and Social Services', in P. Abbott and C. Wallace, eds., Gender, Power and Sexuality. Houndmills, Basingstoke: Macmillan. Sparks.,J.R., and Bottoms, A.E. (1992), Order and Legitimacy in Prisons. Paper presented to the American Society of Criminology, New Orleans. Thomas, T. (1986), The Police and Social Workers. Aldershot: Gower. Young, J. (1986), `The Failure of Criminology', in R. Matthews and J. Young, eds., Confronting Crime. London: Sage. - (1992), `Ten Points of Realism', in J. Young and R. Matthews, eds., Rethinking Criminology: The Realist Debate. London: Sage. (1) An earlier draft of this paper was presented to the British Criminology Conference, University of Wales, Cardiff, 30 July 1993. We would like to thank Judy Heather and Ruth Goatly for their assistance in the research and Alice Sampson, Ian Brownlee, Clive Walker, and Susan Flint for comments on an earlier draft. (2) The most notable recent central government funding sources have been the Five Towns Initiative and its successor the Safer Cities Projects (which has recently been expanded) as well as the more recent Urban Crime Fund. However, in addition the Urban Programme and City Challenge have acted as stimuli to multi-agency schemes. (3) We would like to thank Ian Taylor for reminding us of the dangers of such a course of action. (4) The focused initiatives included in the research were comprised of two estate-based projects, two neighbourhood anti-burglary projects, an area-based street offences project, a town-wide anti-cycle theft initiative, a borough-wide racial harassment project, and a drug-related crime prevention project. Our research involved the examination of relations between a number of key organizations: the police, probation, and social services, and local authority chief executive and housing/architects departments, supplemented by other agencies involved in specific initiatives in the research; including health service workers, voluntary organizations like Victim Support, and drug counselling groups as well as local community representatives. The names of the places have been changed in order to avoid the identification of specific individuals or teams of professionals, as was agreed in the negotiation of access. The fieldwork was conducted over a two-year period between 1991-3, and the research was funded by the Research Committee of the University of Hertfordshire. (5) For example, examine the combined weight of recent Home Office publications in which this policy agenda is articulated for the different agencies (Home Office 1990, 1992, 1993; Morgan
COPYRIGHT 1995 Oxford University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:United Kingdom; Geoffrey Pearson
Author:Crawford, Adam; Jones, Matthew
Publication:British Journal of Criminology
Date:Jan 1, 1995
Previous Article:The opening stages of criminal justice policy making.
Next Article:Private prison research in Queensland, Australia: a case study of Borallon Correctional Centre, 1991.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |