Copyright Sociological Research Online, 1999


Martin Innes (1999) 'Beyond the Macpherson Report: Managing Murder Inquiries in Context'
Sociological Research Online, vol. 4, no. 1, <>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 24/03/99      Accepted: 25/03/99      Published: 31/3/99


This paper draws upon data collected as part of a three year project examining how the police investigate murder, to criticise and develop some of the conclusions of Sir William Macpherson's report into the Stephen Lawrence investigation. It is argued that the majority of attention in the aftermath of the report has been concerned with issues of police racism, but that this ignores a potentially significant theme connected to the management practices and procedures used by the police in relation to murder inquiries. It is argued that many of the problems identified within the Macpherson report may well be systemic problems that result from the situational context in which murder investigations are performed, rather than difficulties unique to the Lawrence investigation.

Detectives; Macpherson Report; Murder Investigation; Policing; Stephen Lawrence


This discussion draws upon data collected as part of a three year project examining how the police investigate murder, in order to criticise and develop some of the conclusions of Sir William Macpherson's Report into the Stephen Lawrence investigation. The majority of attention in the aftermath of the Report has been concerned with issues of police racism, but that this ignores a potentially significant theme connected to the management practices and procedures used by the police in relation to murder inquiries. Also many of the problems identified within the Macpherson Report may well be systemic problems that result from the situational context in which murder investigations are performed, rather than difficulties unique to the Lawrence investigation.

The Macpherson Report, whilst broadly identifying a number of weak points in terms of the ways in which the police tend to investigate murders, fails to understand the complexities and subtleties intrinsic to, and constitutive of this particular aspect of police work. The Report also displays a tendency to conflate what are in actuality separate problems of racism and systemic management failures, and in doing so a situation may have been created which has the potential to have a long-term negative impact upon the effectiveness of the policing of murder in this country. If, as it is argued in this paper, many of the key problems Macpherson identifies can be more adequately explained as systemic failings within the police investigative procedures and management systems, rather than the products of racial discrimination, then the possibility remains of similar failings being repeated in future investigations.

I am not suggesting that there is no racism in the police, nor that racial discrimination by officers was not a factor in the Lawrence investigation. Rather I draw upon data taken from a three year research project into how the police investigate murders,[1] to analyse what is as yet, a comparatively neglected aspect of this Report. I address and expand on a number of central issues identified by Macpherson and seek to understand them, and their implications, through a sociologically informed perspective on these matters.

The Aetiology of Homicide and Investigative Knowledge

It is an established orthodoxy within the sociological literature that most victims of violent crimes know the person who attacks them and that, in contrast to the predominant mediated representations of the 'stranger murder', there tends to be little relational distance between victim and offender. In England and Wales there are somewhere in the region of 700 murders per annum and the police have a clear up rate of approximately 88%. Rock (1998) has argued that the comparatively low number of these crimes and the consistently high clear-up rate has been a major factor in terms of why, as an area of sociological interest in this country, homicide and its investigation remains under-researched. Building upon such speculation, it is arguably the case that it is precisely this situation that has allowed the police to pay comparatively little attention to how they manage these forms of investigation.

In relation to the crimes that are solved, 86% of female and 63% of male victims had some form of prior relationship with their killer.[2] Males make up about 60% of all victims and in these incidents 31% of the victims were killed by a stranger; 8% by a present or former partner; 12% by another family member; and 34% by a friend or other associate. Furthermore, it is known that both victims and protagonists tend to be young men:

Killing is concentrated among young men, and so to, to a lesser extent is the risk of being killed. (Daly and Wilson, 1988: p. 168)

Both Polk (1994) and Daly and Wilson (1988) identify that one of the most common scenarios in which homicide occurs is as part of a confrontational argument between young men. In my study, 20 of the 75 cases in the sample fitted this classification.

The police as the state agency tasked to provide a public response to such incidents are aware of these types of social facts and they are integrated within the 'recipe knowledge' or 'working rules' that they implement to guide their investigative strategies.

The police officers I studied drew a broad distinction between those murders that they identified as 'self-solvers' and those that were 'whodunits'. This is similar to findings from the United States, where murder squad detectives were found to use the labels of 'dunkers' or 'whodunits' (Simon, 1991). In making such distinctions, the detectives were tacitly acknowledging their awareness that the majority of murders, labelled 'self-solvers', are comparatively easy to solve, in that they tend to involve people who are known to each other and who kill one another in an emotional state. The archetypal dramatic device of the stranger murder, popularised through mediated representations, in actual fact constitutes only about 30% of all murders that occur in England and Wales. In providing a response to this latter type of crime, the police activity tends to be oriented around a search for an unknown assailant rather than simply being engaged in 'constructing the case for the prosecution' (McConville et al, 1991), which is their more usual mode of operation.

In relation to the provision of an investigative response to murders, the police organisation can be understood as a mechanism for the transmission of knowledge about murders, the ways in which they occur and consequently how to react to such incidents in an efficient and effective manner. In addition then to procedural guidelines and strategies the organisational setting enables the informal transfer of knowledge about murders and how they have been solved previously between police officers. Such information forming the basis of subsequent responses to this particular form of crime.

Faced with 'a body on the floor', the police are aware of the fact that it is more than likely that the person responsible will have been known to the victim. This knowledge is factored into and informs the police investigative methodology in the early stages of an inquiry. At the outset, the first lines of inquiry undertaken will focus upon securing the crime scene and the various different kinds of evidence that will be contained within it. From this point onwards, if a suspect is not clearly identified, then police inquiries will tend to concentrate upon collating as much information about the victim as possible and eliminating those persons known to the deceased who might have had the physical opportunity or a motive to kill the victim. Generally it is the case that only once the possible suspects from amongst the family, friends and colleagues of the deceased have been eliminated, that the police inquiries tend to shift towards a search for the stranger murderer (Innes, 1999a).

These sorts of findings can be applied to the action taken by the investigating officers in the early stages of the Lawrence investigation and their identification of Duwayne Brooks as a suspect, which is the implication made in ¶5.11 and ¶5.12 of the Report:

Inspector Steven Groves ... and others appear to have assumed that there had been a fight. ... We believe that Mr Brooks' colour and such sterotyping played their part in the collective failure of those involved to treat him properly and according to his needs.

This does not necessarily have to be connected to Brooks' race, for identifying as a suspect an individual who is known to the victim and had the physical opportunity to commit the crime as a suspect would be standard police practice, in the absence of any available substantive evidence to the contrary. Paul Rock's work (1998) on the families of homicide victims provides further evidence to support such a contention, for the police often view family members and those known to the victim as prime suspects, an experience that can be extremely distressing for those concerned.

Preliminary Inquiries

The actions taken by police upon arrival at the scene of the Lawrence murder were subject to particularly strong criticism by the Macpherson team; thus at ¶11.1 it is stated that,

Anybody who listened to the evidence of the officers involved in the initial police action after the murder would, so all the members of the Inquiry feel, be astonished at the lack of command and lack of organisation of what took place. Police officers who gave evidence before us believed that everything they had done had been properly organised and professionally carried out. That does not appear to us to be the position.

Reinforcing these beliefs at ¶11.16, they say,

It can be seen at once that the whole picture is one of disarray and uncertainty. What is certain is that Mr Groves never established any degree of direction or control, except perhaps in the ultimate dragon light search which did take place around midnight...His prime responsibility it seems to us was to establish as far as possible what had happened and to take control of the scene...

Whilst one would not take issue with these findings, the central issue to be addressed though is to what extent these problems within the preliminary inquiries were extraordinary or unique to this particular incident. Although there has been only a limited amount of detailed research upon how the police investigate serious crimes, much of the available evidence does point to the fact that one of the perennial problems in designing an efficient response to such crimes is how to handle the earliest stages.

This is indicated by the American literature which suggests that establishing control of the scene and minimising any contamination of the available evidence, is identified as a key task for the officers providing an immediate response to a crime (Geberth, 1995; Simon, 1991). In my study senior detectives repeatedly expressed concerns about the difficulties of ensuring that the correct actions were taken in setting up a major investigation. They were concerned to quickly identify what are the most appropriate actions taking into account the particular circumstances, because they were firmly of the opinion that most cases tended to be solved in the first 48 hours. One detective said,

Everyone will tell you that it's the first 24-48 hours that are most important. If you haven't picked up a decent lead by then, you can be fairly sure you're in for the long haul.

Officers were very much aware that it was in the early stages that the most evidence was likely to be available and the suspect would have the least opportunity to destroy any potentially incriminating items. But this orientation to taking quick decisions and decisive actions had to be tempered with the consideration, that selecting an inappropriate course of action at an early stage could be detrimental to the overall success of the investigation.

In particular, the increasing sophistication of forensic science and the discriminatory power that it provides has perhaps contrary to expectations effectively served to exacerbate the problems that the police experience in the early stages of an investigation. Advances in forensic technologies have opened up a whole range of new possibilities for officers in terms of how they identify suspects, but at the same time this has been accompanied by a realisation of just how much care is required in identifying and collected physical evidence, if the results of forensic tests are to be legally valid. One middle-ranking detective summed up these problems,

It's not as big a problem as it used to be, because I think officers are more aware, but one of the problems we still have is protecting the scene until a forensic team can get in to do their thing.

This creates a potential paradox for police actions at the scene of a murder, on the one hand they want to get to the scene and examine the physical distribution of evidence, in order that they may construct an idea of the dynamics of the assault, and consequently how best to proceed. But this is countered by the need to prevent any contamination of evidence and to allow forensic specialists to take the requisite time and care in collecting any materials, in order that it should be available as reliable evidence for any future prosecution. There are then contradictory pressures placed upon investigating officers at the scene of a crime, they have to balance the requirement to take decisive action to assist in the early identification of potential suspects, with the need to proceed in such a way as to prevent any evidential improprieties of oversights.

These sorts of problems are compounded by more practical logistical concerns linked to the way in which Senior Investigating Officers are appointed and the difficulties associated with trying to manage a large number of officers whilst setting up the Major Investigation Room, which will act to control and co-ordinate the diverse range of activities performed by the investigating officers.

Macpherson states,

The standard of command and co-ordination during the first two hours after this murder was in the opinion of the Inquiry abysmal. (¶11.36)

Whilst this may be true, it does not make it unique. On the basis of the research that I have conducted, the earliest stages of a murder inquiry appear to often be a weak point and it can often take several hours for an organisational framework to be established and to bring a clear sense of order to the police activities. This may explain why, despite the failings that Macpherson identifies, the police felt that their actions in the aftermath of the incident 'had been properly organised and professionally carried out'.

Managing Information

The primary objective of a police criminal investigation is the production of information, with the intention of identifying the person responsible for the crime in order to support a successful prosecution. Investigative work is, then, first and foremost information work (Manning, 1980; Hobbs, 1988; Ericson, 1993) - the identification, use and construction of information. The ways in which such activities are undertaken by police are both regulated and enabled by the law and the internal organisational policies and procedures. The law acts to constrain the ways in which the police can establish relevant information, but at the same time it provides them with a range of powers to facilitate their actions.

This notion of information work is particularly important to an understanding of major crime investigations. Such inquiries are distinguished by the large amount of information that is discovered by the investigating officers and the need to manage it in an attempt to ensure an efficient and effective investigative response. Indeed this can be identified as one of the defining problematics associated with this kind of work.

It was a recognition that managing information was central to the activities performed by the police on major crime inquiries that led to the introduction of the Major Inquiry Standard Administrative Procedures (MIRSAP), subsequent to the criticisms made by Lord Byford responding to the failures within police systems in the hunt for the so-called 'Yorkshire Ripper', Peter Sutcliffe. MIRSAP established a protocol for the handling, use and storage of information collected by police as part of a major inquiry. As part of this, a specific division of labour was established within the inquiry team, between the strategic management of the inquiry, the officers tasked to conduct the investigative actions and the Major Incident Room which was responsible for the organisation and direction of investigative activities, as well as researching and analysing all incoming information. The objectives behind this division of roles was to establish a system that worked at optimal efficiency whilst minimising the potential for corruption and malfeasance by officers (Maguire and Norris, 1992).

Undoubtedly, the bureaucratic organisation of investigative effort that this management system facilitates has been an important influence upon police practice in the area of major crime investigations (Maguire and Norris, 1992). Nevertheless, it is important to identify its limitations, as these have been alluded to in the Macpherson Report.

The central problem that can be identified in relation to the management of murder inquiries is that policing in 'the information age' has been operationalised with a limited conception of what information is and the police role in respect of it. Specifically, the complex issues relating to the quality of information that the police are uncovering and how this influences their subsequent investigative actions requires extensive consideration. Whilst the police have acknowledged their role in the production of information through the introduction of the MIRSAP protocols, they have struggled to come to terms with how to cope with issues concerning the quality of information that is produced by their actions and how this should be integrated within their procedures.

In sum, the success of future investigative actions in producing information useful to the inquiry in question is dependent upon the validity and relevance of that which has already been collected. If actions are taken upon irrelevant or false information then any further information generated by this line of inquiry is likely to be of little consequence to the established objectives. As such, great care is required in evaluating the reliability of any information 'flowing' into the inquiry team's 'pool of knowledge'. Ideally, the aim should be to collect information of only direct relevance to the incident under investigation, but this is not a practical undertaking given the discovery-orientated role of the investigation.

Therefore the ability to distinguish between 'good' and 'bad' information requires assessments of the quality of the information to be taken, in order to prevent a 'low - productivity spiral' being entered, whereby bad information leads to more bad information. However, any such decisions involve complex assessments in terms of credibility and trust in the information. It is therefore a fallacy to portray crime investigation work as simply concerned with the collection of information. Central to the investigative role which is enacted within a particular legal, organisational, and social context is the interpretation and construction of this information in order to render it in a form which makes it relevant and useful to the police objectives.

The Home Office Large Major Enquiry System (HOLMES), which is essentially a computerised data storage and retrieval system building upon the foundations of MIRSAP, was introduced to assist in the management of information and these sorts of problems. There are though problems associated with the technological limitations of the system which require discussion, these problems are linked to the issue of information quality introduced above.

Establishing the importance and value of a specific piece of information is a difficult job for a police officer, they have to retain an awareness that witnesses can be mistaken (Gudjonssen, 1992); suspects are motivated to lie and informants sometimes deliberately seek to provide dis-information (Innes, 1999b). Establishing trust in a piece of evidence or information can be accomplished via a subjective evaluation of the reliability of the information source and considering how this information fits in with the pool of investigative knowledge established so far.

Any such problems are compounded in a murder investigation where there is a lot of information becoming available to officers, from various sources, over a short period of time. In the first generation of HOLMES computers that were in use when the Stephen Lawrence murder occurred, it was thought that trying to formally evaluate the quality of information and subsequently to prioritise its value to an investigation, was too difficult and dangerous. Consequently information was dealt with on a 'first come first served basis', in that it would be put onto the system in the order it arrived at the Major Incident Room.

The problem with this approach was that a lot of effort was being put into getting information onto the system that was of no real relevance to the investigative objectives. At the same time, there were delays in getting vitally important facts that the investigating officers needed onto the system. Where all incoming information was treated as having equal value, this meant that investigators were in effect often having to wait for the system to catch up. This was not a unique problem for the Lawrence investigation though. A Deputy SIO in my study suggested that:

HOLMES may blunt the efforts of good investigators...the whole system can easily become bogged down in information and the whole team of investigators are looking at this bit of information, which is probably irrelevant to what they're trying to achieve. It doesn't matter how good the investigators are if they're working in that situation.

These sorts of systemic problems, which are paid only cursory attention in the Macpherson Report, may in help help us to provide explanations for some of the criticisms that were made.

One of the central problems that the Macpherson Report draws attention to in Chapter 13 (¶13.24; ¶13.25) is the fact that there were a number of sources providing information to the police early on in their investigation that with the benefit of hindsight would have appeared to be important. There are evident failings in terms of how this was handled, but it should be recognised that these are at least partially reflections of system failures to do with trust in information and how issues of quality of information were enacted within the police investigative systems. Although current investigative practices have tried to address issues to do with distinguishing the quality of information and prioritising it accordingly, it remains a central problem for the police.

Financial Pressures

Murder investigations are highly resource intensive, a fact that is widely acknowledged throughout the police service. Indeed, as the police have increasingly been subject to constraints over their funding so concerns with efficiency in relation to murder investigations have risen up the agenda of problems. This situation needs to be understood in relation to the broader policing context, in that resources diverted to murder investigations cannot be deployed to carry out other policing services. A number of the Senior Investigating Officers interviewed for my research expressed increasing concern that financial considerations and concerns with efficiency were impacting upon the effectiveness of their investigations. One Detective Chief Superintendent interviewed acknowledged that these sorts of problems were increasingly prevalent in shaping policy decisions in relation to murder inquiries:

There's a school of thought in the police service saying too much money is put into investigating murders...Certain senior officers will say 14 days and bonk... We'll throw money and resources at a case for 14 days and if you haven't made progress then that's it. I'm fighting some real battles there...I think they're more questioning now and will be taking into account a whole range of issues. I don't know what the public's attitude is to whether you 'pull' a murder enquiry after three weeks...

These issues are addressed in particular in Chapter 32 of the Macpherson Report, where lack of staff was identified as having a specific detrimental impact upon the Lawrence investigation,

...this investigation was grossly understaffed both in the incident room and externally. (32.16)

This is indicative though of the routine problems that are facing the police service at this time. There is an almost unquenchable demand for police services in relation to different sorts of problems and consequently decisions have to be made as to the proportion of resources to be made available for the provision of different policing services. The Macpherson team try to circumvent this problem by comparing the Metropolitan Police Service's resources with a number of other forces. They conclude that in comparative terms the Metropolitan Police was not subject to inadequate resources or excessive demands in relation to homicide. This conclusion does not though mean that all police forces in England and Wales have not been finding it increasingly difficult to sustain an intensive resourcing commitment for a small number of serious crimes.

Policing Murder: A Situational Perspective

The sorts of failings which the Macpherson Report draws attention to in relation to the police investigation into the murder of Stephen Lawrence are important, but the explanation as to how and why they occurred is somewhat limited. The problems identified are not unique to this case or racial murders; rather they are related to the social, legal and organisational environments in which this sort of police work is undertaken. Understanding the police response to murder in terms of the situational context in which it takes place, assists us in comprehending the complexities involved in this area of their work.

A situational perspective points us to the ways in which the process of investigation can be identified as dynamic, in that the focus of police activities shifts as progress is made by detectives. This notion of process is central to understanding, because relevant information and evidence is progressively revealed as a result of investigative actions, which forms the basis for further activity. Even when a suspect is easily identified, my research suggests that actually pulling together information from different sources to construct the case for the prosecution in accordance with the legal requirements can be a difficult and complex undertaking.

The situational perspective also seeks to account for the influence manifested as a result of the interaction that occurs between: the specific qualities of the incident under investigation; the organisational practices and methodologies of the police service; and the interpretations and actions of individual officers in relation to these two influences. The particular shape, dynamics and trajectory of an investigation can be understood as the outcome of the interplay of these influences. The adoption of this approach then, is useful in that it illustrates the limitations to the conclusions of the Macpherson Report.

Investigating murder is a complex undertaking that often involves a range of contradictory pressures that have to be accounted for and compensated for in terms of the strategies that are implemented. Therefore it is most useful to understand the police actions as information work. The investigating officers are seeking to produce sufficient information to enable a suspect to be identified and to support a prosecution of them. In doing so though, the police do not investigate all incidents according to a set procedure. Rather the investigative methodologies that they employ are intended to be adaptive, in that different incidents involve different investigative problems that have to be accounted for and resolved if the enquiry is to be successful.

More generally, a situational perspective leads us to consider recent developments in terms of the role of the investigator and concerns with the nature of investigative skills within the police organisation. Detectives have traditionally tended to portray the key skills needed for detective work as being a discrete blend of the rational and intuitive. This has been part of the 'mythology' of detective work, that given sufficient resources, including an individual officer with the appropriate skill and knowledge, any crime is solveable (Innes, 1999a). This mythology has been contested by sociological work on how crimes are solved. Such studies have repeatedly demonstrated that by far the most significant determinant as to whether or not a crime will be solved is the quality of information provided to investigators by members of the public (Greenwood et al, 1977; Ericson, 1993; Matza, 1969).

The consistent nature of these types of findings have led a number of senior police managers, under pressure to deliver efficient policing services, to be increasingly questioning of the value of the specialised role of the crime investigator, institutionalised in the police service in the figure of the detective and the Criminal Investigation Department (CID). Further support for such revisionary thinking was provided by a number of serious miscarriages involving detective work, where corruption and serious misconduct by officers was demonstrated to have taken place (Maguire and Norris, 1992; Reiner, 1992). For a time the Metropolitan Police Service was reported to be seriously considering disbanding the detective branch and returning the investigative function to uniform officers, as a way of breaking the notion of the CID as constituting a 'firm within a firm' (Hobbs, 1988). Such measures were also seen as a way of decreasing the risks of future miscarriages of justice. Eventually though, such radical innovations were not adopted, although many forces now only allow officers to serve for a fixed term in the detective branch before having to return to uniform.

This solution is, though, symptomatic of the attitudes of certain senior officers within the police service, who have recently tended to place more emphasis on 'problem solving' police skills than investigative skills, in the belief that the former are more important. One of the complaints made by the senior detectives in my study was that the trajectory of their career paths meant that very few officers nowadays have the luxury of spending considerable time in an investigative role, which would allow them to acquire the skills required to be effective in terms of investigating complex and difficult cases. These sorts of skills they felt were institutionally devalued. It can be argued then, that in the process of adopting such an approach towards detective skills, a number of senior policy makers have neglected to consider that an effective investigator may have to be possessed of certain specialised knowledge and skills.

If we understand crime investigation work as being first and foremost information work, then the role of the effective investigator is not simply concerned with processing this information. Detectives have to be able to interpret, analyse and critically evaluate the content of a piece of information and the reliability of the source, if they are to be able to negotiate the complex trust issues through which a piece of information is constructed as useful to them. The ability to produce and construct information in a form that can be used to inform investigative activities, whilst being conterminously legally valid, can be identified as constituting the core skills of the crime investigator. Just because specialised detective skills are not important in solving most crimes, does not mean that they do not influence the success or otherwise of a small number of the more difficult to solve crimes. The management of large-scale crime inquiries then, requires not only a Senior Investigating Officer who is aware of procedural and organisational guidelines, but an individual who has the capacity to think critically about the information and evidence that is collected, and to construct solutions to various types of investigative problems.

The Symbolic Qualities of Murder Investigation

The dramatic qualities inherent to the crime of murder, and the unequivocable harm that it denotatives, creates a demand for effective state action in relation to these sorts of incidents from the public. In responding to such crimes, the police role is aligned with public demands for unambiguous action, as a consequence of which the police organisation is at least temporarily legitimated and the links between the police, state, law and morality are symbolically displayed, through the ritual of the murder investigation. What the public furore over the failings of the Stephen Lawrence investigation and other cases such as the murder of Rachel Nickell demonstrates, is the concern shared by the majority of the public that the police should be able to solve these particularly serious and high profile crimes. Their failure to do so may cause much harm to the institutional legitimacy attributed to the police by the public.

There is no doubt that mistakes were made in the investigation into the murder of Stephen Lawrence, but the failings of Macpherson and his colleagues to comprehensively understand the nature of some of the key problems involved in this particular area of police work seems to have led them to misunderstand the causes of a number of the problems that they identify. Consequently, an opportunity for improving the quality of service provided by the police when investigating murder may well be missed.


1The project was conducted with a police force in the South of England. The principal sources of information were ethnographic 'observer as participant' methods and semi-structured and informal interview. In total about 5 months of observational data was collected. As a condition of access to the data, assurances were provided that the police force concerned or any individuals connected to the cases studied would not be identified. This agreement is maintained in the paper.

2Source: Criminal Statistics England and Wales 1994, Table 4.8.


DALY, M and M. WILSON (1988) Homicide. New York: Aldine de Gruyter.

ERICSON, R. (1993) Making Crime: A Study of Detective Work (2nd edition). Toronto: Toronto University Press.

GEBERTH, V. (1995) Practical Homicide Investigation. New York: CBC Press.

GREENWOOD, P., J. CHAIKEN and J. PETERSILIA (1977) The Criminal Investigation Process. Lexington, Ma.: D.C. Heath.

GUDJONSSEN, G. (1992) The Psychology of Interrogations, Confessions and Testimony. Chichester: Wiley.

HOBBS, D. (1988) Doing the Business. Oxford: Oxford University Press.

INNES, M. (1999a) Investigating Murder: The Police Response to Criminal Homicide. Unpublished Ph.D. dissertation, London School of Economics.

INNES, M. (1999b) 'Professionalising the police informant: the British experience', Policing and Society, (in press).

McCONVILLE, M., A. SANDERS and R. LENG (1991) The Case for the Prosecution. London: Routledge.

MACPHERSON OF CLUNY, Sir William (1999) The Stephen Lawrence Inquiry: Report of an Inquiry by Sir William Macpherson of Cluny. London: The Stationery Office; <>.

MAGUIRE, M. and C. NORRIS (1992) The Conduct and Supervision of Criminal Investigations. (Royal Commission on Criminal Justice Research Study No. 5) London: HMSO.

MANNING, P. (1980) The Narc's Game. Cambridge, Ma.: MIT Press.

MATZA, D. (1969) Becoming Deviant. Englewood Cliffs, N.J.: Prentice Hall.

POLK, K. (1994) When Men Kill. Cambridge: Cambridge University Press.

REINER, R. (1992) The Politics of the Police (2nd edition). Hemel Hempstead: Harvester Wheatsheaf.

ROCK, P. (1998) After Homicide: Practical and Political Responses to Bereavement. Oxford: Clarendon Press.

SIMON, D. (1991) Homicide: A Year on the Killing Streets. London: Coronet Books.

Copyright Sociological Research Online, 1999