Punishing Violence

Antonia Cretney, Gwynn Davis

• Publisher: Routledge
• Year: 2002
• Print ISBN:
  9780415098397
• eBook ISBN:
  9781134859351
• Topic: Medicine
• Subtopic: Health Care Delivery

It is a common perception that violent crime is on the increase and social surveys record a growing fear of victimisation among the public. Yet not all violence is criminalised, and much criminal violence still goes unreported.
Punishing Violence examines the series of decisions - by victims, police officers, prosecutors and courts - which determine whether or not violent behaviour is criminalised.
Antonia Cretney and Gwynn Davis examine the relationships underpinning violence, the reasons for violent acts and the factors militating against successful court prosecutions. In doing so, they provide an authoritative account of the reality of assault and identify a serious dislocation between the purposes of victims and the purposes of the justice system in the treatment of violent crime.

Chapter 1

Introduction

Violent incidents comprise but a tiny fraction of the encounters which we have with our fellow men and women. Most of us subdue our violent impulses virtually all of the time. We are effectively socialised into a non-violent culture. But it is readily apparent that this is not true of all, and it is perhaps true of very few of us in extremis. Interpersonal violence is for this reason a matter of abiding academic and media interest. There has of late been a perceived increase in the prevalence of violent conduct, and while there may indeed have been some increase in real terms this perhaps also reflects a greater public awareness and recognition of violence as a problem meriting criminal sanction. It is possible that we have a greater sensitivity towards violence than our forebears, and a greater reluctance to tolerate its use in public or private.¹ Two questions then follow: does this greater sensitivity encourage those on the receiving end of violence to bring their victimisation to the attention of police and courts?; and, second, what is the response of criminal justice agencies to those assaults which are brought to their notice? In this book we shall focus upon a certain type of behaviour (interpersonal violence) in order to explore the definition of that behaviour—first by victims, and second by the police and courts—and, thereafter, to monitor the response to that behaviour, either within the framework of criminal prosecution or independently of it.² Even as we write we read of a major teaching union opting to bring private prosecutions for assault against schoolchildren who have attacked its members;³ of complaints by the Prison Officers’ Association that serious assaults by inmates upon officers and fellow prisoners have not led to criminal proceedings;⁴ and of Crown Prosecution Service reluctance to level serious charges against drunk drivers who kill.⁵ Does this indicate a failure on the part of agents of the state to respond effectively to flagrant breaches of the criminal law? If so, what are the reasons for this failure?

‘Assault’ has a technical, legal definition as well as a lay meaning. Legally speaking it covers a number of offence categories ranging from ‘common assault’ (punishable with a maximum penalty of a £2,000 fine and/or six months’ imprisonment) to ‘wounding or causing grievous bodily harm with intent’ (maximum sentence: life imprisonment).⁶ In addition there are various sexual assaults, of which the most serious is rape, which again carries a maximum sentence of life imprisonment. We do not take these legal definitions as our starting point. We are concerned with any physical violation through the use of force, although we should make clear at this point that we have focused principally upon non-sexual violence. Physical integrity is in general highly valued (Ashworth 1991:277) and the violation of that integrity, through the application of force, is profoundly unwelcome. Assault, therefore, although subject to quite complex legal definition, is something to which we all respond intuitively.

This is not to say that there are no problems of definition, and it is inevitable that any attempt to measure the extent of violent conduct will encounter difficulties. Nonetheless it is important to make some attempt to do this and the obvious place to start is with the official record of offences of violence against the person recorded by the police. The figures reveal a dramatic increase in the number of violent offences over the past 20 years, a trend also observed in other Western countries. Thus in 1971 there were 47,000 recorded offences of violence against the person in England and Wales; in 1990 the figure had risen to 184,700; in 1991 it was 190,300; and in 1992 it was 201,800.⁷ While the method of recording may have changed somewhat, rendering the figures not precisely comparable, there is no denying that an increase of over 400 per cent in 20 years is very dramatic. The rise has been most significant in respect of the less serious violent offences (Walmsley 1986:53), although one may also note that robbery, which is essentially a crime of violence (Ashworth 1991:340) and can give rise to very serious injury, also shows a striking increase. There were in fact 20,300 robberies recorded in England and Wales in 1981; 45,300 in 1991; and 52,900 in 1992.⁸ In general we can say that violent offences appear to have increased more rapidly than other crimes, and this is reflected in the fact that violent offenders form an increasing proportion of the prison population.⁹ The official statistics also reveal an apparently higher prosecution rate for assaults than for most other offences. Some 75 per cent of recorded violent crime is said to be cleared up—referred to by one Home Office Minister as ‘a very high detection rate indeed’.¹⁰

It is acknowledged, not least by those responsible for gathering the figures, that criminal statistics contain many distortions. Increases in recorded crime probably reflect changes in reporting and recording practices at least as much as they do changes in the underlying crime rate. Taking first the willingness of victims (and others) to bring offences to police attention, it is probable that offences against the person, such as wounding, are among the least well reported.¹¹ Even victims who do report an assault may do so reluctantly (Shapland et al. 1985:16), making it less likely that the offence will be recorded or an investigation pursued. Our own contact with victims living in conditions of severe deprivation (for which see Chapter 2) suggested a high level of unreported violence. Heavy drinkers, especially, may become involved in an altercation and perhaps collapse in the street, unaware that they might lay claim to being an assault victim and certainly with no intention of reporting the incident to the police.¹² As a corollary, it is possible that the recent rise in the number of woundings and other assaults reflects a greater willingness to report violence, at least in certain contexts. Most obviously there has been a marked increase in the number of ‘domestic’ assaults reported to the police: this almost certainly reflects a growing perception on the part of women that assaults in the home can and should be brought to police attention.¹³

The other difficulty in calculating the true level of assault arises from variations in police recording practices in respect of those incidents which are brought to their attention. There is evidence of significant non-recording (or ‘cuffing’, in police argot) of some categories of assault, notably those in which the police anticipate difficulty in prosecuting. One survey found that nearly a quarter of assault and wounding cases were written off (for statistical purposes) by the police, usually because the complainant did not wish the offence to be prosecuted (Sparks et al. 1977:145). Those researchers observed that cases written off in this manner ‘were almost invariably family disputes’. Another survey, this time of assault victims who sought hospital treatment, revealed that only one-quarter of these assaults were recorded by police although half the victims interviewed claimed police awareness of the incident (Shepherd et al. 1989). The researchers noted that assaults which occurred in the street or in clubs or discotheques were unlikely to be recorded by the police. Overall, assaults on female victims were even less likely to be recorded than were assaults on males, again reflecting the down-playing of ‘domestic’ violence.

Just as a greater readiness to report assault probably accounts for some part of the rise in the official figures, it is likely that a greater police willingness to record assaults upon ‘unpromising’ victims has also contributed (Mayhew et al. 1993:22). In Bristol, for example, each of the three police divisions which cover the city now records about a thousand complaints of ‘domestic’ assault each year (Cunningham 1990). This, in the words of a local Detective Chief Inspector, reflects a policy decision taken within the police force in 1988 ‘to intervene, rather than to ignore most complaints’ (Cunningham 1990).

While there is good reason to suppose that certain categories of assault are more likely to be reported than hitherto and, having been reported, are more likely to be recorded by the police, we should not assume that the official record, even now, provides anything approaching a true picture. The disincentives to reporting and recording are too strong for that to be the case.¹⁴ This encourages a search for alternative measures, notably the various crime surveys which have proved such a powerful tool in calculating the true level of criminal activity. On the basis of these data it has been estimated that only a quarter of criminal offences find their way on to police records. Furthermore, the rate of non-recording appears to be highest in relation to assaults and thefts from the person (Sparks et al. 1977:157). The 1982 British Crime Survey suggested that only 23 per cent of violent offences and 11 per cent of robberies appeared in police crime records (Hough and Mayhew 1983). Latest British Crime Survey figures suggest that 17 per cent of robberies and 20 per cent of woundings are recorded by the police, compared with an 86 per cent recording rate for theft of motor vehicles.¹⁵ This is consistent with survey data from other countries: the strongest correspondence between survey figures and offences recorded by the police is in relation to vehicle theft; there is a moderate correspondence for burglary; and a very poor correspondence for assault.¹⁶

As well as providing a truer measure of criminal activity, surveys offer an alternative means of comparing crime levels in different countries. One technique involves conducting victim surveys by telephone at the same time in several countries, with the same questions being asked. This has revealed England and Wales to have one of the lowest rates of non-sexual assault, well behind Australia and the USA.¹⁷ Surveys also provide an alternative means of calculating trends in violent conduct, with the latest figures being compared with results for previous years. Overall, violent crime recorded by the British Crime Surveys rose by 8 per cent between the two surveys of 1982 and 1988,¹⁸ while the 1992 survey revealed a further increase. The number of woundings rose from 507,000 in 1982 to 626,000 in 1992—an increase of 12 per cent, while the number of common assaults rose from 1,402,000 to 1,757,000—an increase of 25 per cent.¹⁹

It has been acknowledged, not least by the Home Office researchers principally responsible for the British Crime Surveys, that surveys have many limitations when it comes to calculating the level of violent crime (Hough 1991). Many assaults, Hough suggests, will not be defined by the victim as criminal because they take place in the domestic sphere. Non-domestic incidents may likewise not be defined as criminal even though an offence has clearly been committed. The problem with assault, as compared with an offence such as burglary, is that respondents will apply different criteria when answering so-called ‘screen’ questions such as: ‘Has anyone hit or attacked you in any way?’. A related problem is to be found in the ‘events’ orientation of crime surveys. This assumes that infringements of the law are discrete incidents. But, Hough suggests, some violent offences are better conceptualised as processes rather than events. This is particularly true, it might be thought, of ‘domestic’ assault. And indeed it has been claimed that up to 25 per cent of women have been struck by their partner.²⁰ It has also been suggested that better educated respondents are more likely to report an assault because they have a lower tolerance of this type of offence and are more likely, therefore, to regard it as having crossed the ‘crime’ threshold (Mayhew et al. 1989:5). If it is accepted that people differ in their attitudes towards violence, this creates problems for any attempt to measure the extent of violence by means of survey techniques. It has been suggested, for example, that middle-class men are more likely to be upset by violence than are their working-class counterparts, and that many ‘assaults’ are in fact minor fracas between young men which cause no great distress to the participants or to anyone else. These are matters which we shall explore in this book, but we can say immediately that it is wrong to suppose that there is widespread tolerance of violence on the part of assault victims, or that assaults not reported to the police generally involve incidents which might be characterised as ‘six of one and half-a-dozen of the other’. In order to explore these questions properly we have to go beyond survey data. Even in terms of simple counting, it is necessary to study behaviour rather than to rely on self-reporting on the one hand or police-defined samples of victims on the other.

One measure of behaviour is the deliberate infliction of injury, and this has led researchers to attempt to measure the incidence of assault by reference to the Accident and Emergency (A & E) Department record of assault victims admitted to hospital for treatment. Shepherd et al. (1989) make the case for utilising this data source. They observe that the recording of violent offences by hospitals depends on the presence of injuries perceived to require treatment, rather than on victims’ perception that a crime has been committed. Other advantages are that victims may be interviewed shortly after injury; that A & E Departments serve well-defined communities; and that hospitals are independent of the police. A disadvantage of this method of recording is that even the seriously injured may decide not to attend for treatment—may even be prevented or deterred from doing so by their assailant. But that is a problem common to the police record also. In fact the population of victims seeking hospital treatment may be largely distinct from that known to the police, and distinct also from that which is recorded in the various crime surveys. This is reflected in British Crime Survey data suggesting that most crimes of violence do not result in serious physical injury. The 1984 survey revealed that in only 12 per cent of assaults did the victim seek professional medical attention (Hough and Mayhew 1985), while the 1988 survey indicated that in respect of the more serious offence of wounding only 30 per cent of victims sought medical treatment (Mayhew et al. 1989:37).

By virtue of their providing a partially distinct sample of assault victims, A & E Department records also provide an alternative measure of trends in violent conduct, and one that is not contaminated by changes in reporting behaviour or an increased police presence. Thus Shepherd et al. (1993) report, on the basis of hospital data gathered in Bristol, that the incidence of intentional injury for which treatment was sought at the local A & E Department reached a peak in 1983 and has not increased since. This finding is markedly at variance with the picture presented by police crime statistics. It may be that hospital data, while deficient in many respects, provide a better indicator of trends in violent conduct than do police records. A & E Departments may, at the very least, provide a better measure of assaults occasioning significant injury than do either police records or crime surveys; on the other hand, they are unlikely to provide an accurate measure of less serious violence.

A DIFFERENT KIND OF MONITORING

Gottfredson has suggested that ‘because of the rarity and clustering of serious criminal victimisation…’ crime surveys such as the ones now being conducted on a regular basis by the Home Office are not the best way to study the circumstances surrounding serious offences. He identifies ‘a particular need to “flesh out” the situational aspects of criminal events’ (Gottfredson 1984:31). Sparks et al. were likewise alive to the limitations of victim studies such as their own and recommended further projects which would follow incidents through the justice system, starting at the earliest possible moment:

In this way it would be possible to obtain a more complete picture of both the police and the public’s reaction to crime incidents of various kinds, and to gain a better knowledge of the factors which lead some incidents to be settled informally without being recorded in police statistics of crime.

(Sparks et al. 1977:233).

That is indeed what we have attempted to do. A study of this kind enables the researcher to explore the possibility that even self-report studies arriv...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• List of Illustrations
• Preface
• 1 Introduction
• 2 The Reality of assault
• 3 Harm and reporting
• 4 The victim and the police
• 5 The police and the demands of the legal process
• 6 Police case construction
• 7 Victims in court
• 8 Assault, prosecution and the victim
• Appendix A—Research method
• Appendix B—The cases
• Notes
• Bibliography
• Index