, Henham

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

Child Victims and Sentencing Policy for Violent and Sexual Crimes

Ralph Henham

Copyright © 1997 Ralph Henham.
First Published in Web Journal of Current Legal Issues in association with Blackstone Press Ltd.

Summary

This paper analyses the evidence relating to child victimisation in violent and sexual crimes and suggests the common perception that social protection is achieved through the sentencing policies of the courts appears both simplistic and inaccurate. Subsequent evaluation of the protection actually offered by existing legislation and appellate guidance confirms the lack of any coherent sentencing policy to deal with child victims. A number of suggestions for improvement are made.

Introduction

In contemporary Britain it is widely assumed that children,⁽¹⁾ some of the most vulnerable members of our society, receive adequate protection from the most dangerous offenders through the sentencing policies of the courts. The main purpose of this paper is to explore the extent to which this common perception is supported by the available evidence. The paper begins by examining official statistics and victim survey information in an attempt to establish the extent of child victimisation in legally-relevant offence categories. Legislative provisions and appellate sentencing guidance are then analysed to reveal the purpose and scope of child-related sentencing policy. The paper concludes with some suggestions to improve the consistency and relevance of sentencing guidance in this important area of sentencing policy.⁽²⁾

Statistics on child victimisation

Although much has been written about children as victims of crime (see Morgan and Zedner 1992a, Zedner 1997), it is difficult to establish the extent to which child victims are represented in the legally-derived categories used to compile the annual criminal statistics,⁽³⁾ apart from those categories covering essentially child-specific crimes such as infanticide, child destruction, abandoning a child under two years, child abduction, unlawful sexual intercourse with a girl under 13, unlawful sexual intercourse with a child under 16 and gross indecency with a child. The number of offences recorded by the police (excluding infanticide) falling into these offence categories during 1985-95 indicate that the majority of such offences do not display any consistent pattern of annual increase or decrease (Home Office 1995, Tables 2.15, 2.16). The exceptions are child abduction, which reflects a clear upward trend, rising 81 per cent between 1991 and 1995, and unlawful sexual intercourse with a girl under 16, where offences recorded have been falling more or less consistently since 1985. Conversely, recorded offences of gross indecency with a child have been rising more or less consistently since 1985. It is, nevertheless, instructive to compare the percentage change for offences recorded in each category between 1985 and 1995. The figures show, for example, substantial increases for abandoning a child under two (+77 per cent), child abduction (+248 per cent) and gross indecency with a child (+103 per cent) whilst percentage reductions exist for unlawful sexual intercourse with a girl under 13 (-40 per cent) and unlawful sexual intercourse with a girl under 16 (-53 per cent). Although detailed analysis of the variables responsible for these changes exceeds the remit of this paper, it is probable that the decrease in recorded offences of unlawful sexual intercourse with a girl (whether aged under 13 or 16) is, at least, partially explicable by changed reporting patterns among children, resulting in many such offences being reported as rape. In the normal course of events it would be impossible to obtain any reliable estimate of child victimisation in rape cases from the notifiable offences recorded by the police. However, Watson (1996) has recently published findings which report survey data on victims of violent crime recorded by police forces in England and Wales from 1990 to 1994 and incorporate valuable information on child victimisation. For example, between 1990 and 1994 an average of 22 per cent of rape victims were aged between 10 and 15 (4 per cent aged under 9) with the highest victimisation rate (59 victimisations per 100,000) being in the 10 to 15 age group. Watson opines that the high victimisation rate for offenders in the 10 to 15 age range is partially explicable by reporting factors⁽⁴⁾ and that many victims under 16 are more willing to report offences than many adult victims due to parental influence. As regards indecent assault (including buggery)⁽⁵⁾ Watson reports that over two-thirds of male victims of indecent assault and buggery were under 16, with 43 per cent aged between 10 and 15 and 27 per cent aged under 10. Although female victims of indecent assault tended to be older than male victims, approximately half of female victims were aged under 16, with 17 per cent under the age of 10. Watson also notes that female victimisation rates for recorded indecent assault are substantially higher than for male victims in every age group, but, both males and females in the 10-15 age group were most at risk (with 66 male victimisations and 327 female victimisations per 100,000 population).⁽⁶⁾ Finally, Watson examined the age distribution of male and female victims of recorded violence against the person⁽⁷⁾ and, perhaps not surprisingly, found that two-thirds of victims were male, with 14 per cent aged under 16, and for females, the proportion of child victims under 16 was 12 per cent.⁽⁸⁾

In summary, therefore, the somewhat incomplete picture painted by official statistics of recorded crime appears to suggest consistent and significant increases in cases of child abduction and gross indecency with a child, while alarmingly high victimisation rates have been noted by Watson in recorded rape and indecent assault cases in the 10 to 15 age group. However, it is well known that there are significant limitations in the ability of official statistics to adequately delineate levels of actual crime (see Maguire 1997) and it is, of course, largely for this reason that the British Crime Survey was originally developed and first carried out in 1982. Although wounding offences (not common assault) are comparable with official police figures in the British Crime Survey, no specific information on sexual offences is collected and, in any event, the British Crime Survey only includes victims over the age of 16.⁽⁹⁾ Nevertheless, the 1992 survey (exceptionally) included an additional sample of 1,350 young people aged 12-15 who were questioned on a range of topics including their experiences of victimisation away from home, fear of crime, and contacts with and attitudes towards the police (Maung, 1995). In her report, Maung analyses (inter alia) the nature of victimisation and notes that victims of assault usually cited perpetrators of the same age and sex as themselves, and school-related assaults in particular were frequently perpetrated by other young people, often fellow pupils.⁽¹⁰⁾ However, in the case of harassment (other than sexual) 24 per cent of incidents were regarded more seriously when committed by adults. More significant, in the present context, was the nature of victimisation in sexual harassment cases where, as one would expect, incidents were regarded much more seriously and victims were more frightened. Sexual harassment cases were most likely to come to the attention of the police,⁽¹¹⁾ although Maung points out that incidents perceived as 'more serious' by 12 to 15 year olds were more likely to be communicated to adults, notwithstanding victims often felt that they ran the risk of not being believed and (or) having their freedom curtailed. Maung concludes that since most incidents against young people are not perceived as 'crimes' by the victims and are perpetrated by people of the same age it is questionable whether such victimisation is capable of resolution by the criminal justice system, notwithstanding the incidents may result in harm or distress for the victim. Certainly, repeat victimisation may lead to the perception that a certain level of victimisation is 'normal' and part of everyday life. Maung does not suggest, however, that a similar conclusion should be drawn in the case of incidents perceived as 'more serious' by victims, such as sexual harassment.

Of the many criticisms directed at large scale victim surveys, such as the British Crime Survey, is their failure to explore the reasons for multiple victimisation, and, to appreciate the fact that much criminal behaviour, such as the sexual and physical abuse of children, is intimately connected to ongoing relationships between offenders and victims. Substantive 'new realist' criticism has centred on the British Crime Survey's failure to adequately account for the phenomenology of victimisation, particularly in the context of sexual assaults on women (see Maguire 1997, pp 169-172). In consequence, much alternative qualitative research focusing on specific types of criminal behaviour and particular victim groups has been produced (see further Zedner 1997, pp 590 - 591). In the present context recent work by Morgan and Zedner (1992), Anderson et al (1994) and Hartless et al (1995) on child and early teenage victimisation has provided important information about the extent and impact of child victimisation. For example, Hartless et al questioned a sample of 208 11-15 year old young teenagers from an inner city Glasgow school regarding their victimisation experiences and offending behaviour.⁽¹²⁾ In common with Anderson et al's Edinburgh study, 37 per cent of respondents had been assaulted during the past year, and, as regards sexual offences, 5 per cent (4 per cent in Edinburgh) of boys reported being touched by males and 17 per cent of girls, in both Glasgow and Edinburgh. In general terms, 82 per cent of the sample reported at least one victimisation during the past year, with a mean of four victimisations in the previous year for the whole sample. As expected, females were more likely to be victims of harassment and sexual offences. Consistent with Maung's study, of those victims in the sexual offences group who could, or would, provide details of the offender, 82 per cent described the offender as an adult, and 72 per cent maintained that it was a stranger. Hartless et al conclude that, overall, victims were unlikely to report victimisation experiences, although some 80 per cent of them were regarded as 'serious' in that victims identified the offender as an adult stranger.

The British Crime Survey and local or group victimisation studies, therefore, confirm and expand information obtained from official statistics by suggesting a high degree of victimisation for both violent and sexual crimes involving child victims, and, have contributed substantially to the development of an ethnography of victimisation in particular social and cultural settings. This de-mystification of particular victimisation experiences, victim perception and the reasons and reporting rates for violent and sexual crimes against children, supports the view that there is extensive under-reporting of such crimes in specific areas (up to 80 per cent, for example, in Glasgow according to Hartless et al), and, as we have seen, Watson has, in any event, demonstrated that the highest victimisation rates in the 11-15 age group are for recorded offences of rape and indecent assault. Notwithstanding that a serious social problem clearly exists regarding the 'real' extent of child-related violent and sexual crime, the role of the sentencing court, and sentencing policy in particular, appears equivocal. The common perception that social protection is achieved through the sentencing policies of the courts appears both simplistic and inaccurate when viewed against a background of extensive under-reporting of child-related violent and sexual crime, and, more specifically, the possible reasons for that phenomenon. It is evident that, apart from the most important reason given for not-reporting to adults (incident 'too trivial'), some of the other most supported reasons suggest negative perceptions regarding the relevance, ability and effectiveness of the criminal justice system to address the offending behaviour in question (see Maung 1995, Table A3.5). For example, the reason 'no-one could do anything' although ostensibly directed at adults, essentially reflects a perception that, ultimately, there is little point in involving the police. Similarly, the reason 'frightened of offenders' may imply a future reluctance to give evidence, or, a perceived lack of protection offered by the police or eventual sentencing measures. The remaining reasons are more obviously related to the victim's immediate social and cultural milieu.⁽¹³⁾ Certainly, no impression is gained that the courts are ultimately perceived as providing a distinct measure of protection for child victims, since there are many more immediate and directly relevant considerations in the decision not to report, for example; how would the police react if they became involved? Nevertheless, the immediate perception that punishment, deterrence and public protection are paramount and attainable objectives in sentencing child-related crime might more readily increase the willingness of child victims (or their parents) to report violent and sexual crimes.⁽¹⁴⁾

The remainder of this paper is, therefore, devoted to evaluating the extent to which public protection is actually reflected in current sentencing policy regarding child victims of violent and sexual crimes.

Legislation, appellate sentencing guidance and child-related crime

Legislation

Where offenders are regarded as 'dangerous' child victims of violent and sexual crime derive no particular concessions or consideration from the relevant legislative provisions. Dangerous offenders are potentially subject to a plethora of sentencing legislation which is not only irrational and lacking in any overall strategic policy objective, but also, largely in consequence, substantially defective in achieving any widely recognised sentencing objective such as retribution, deterrence or social protection. In the first instance, the Criminal Justice Act 1991 abandoned any attempt to define potentially dangerous offenders and devolved this responsibility onto the Court of Appeal. Currently, dangerous offenders may receive either commensurate sentences under ss1(2)(a) and 2(2)(a) or protective sentences under ss1(2)(b) and 2(2)(b) of the Criminal Justice Act 1991 (see Henham 1996a, von Hirsch and Ashworth 1996) which are based upon principles of selective incapacitation (see Mathiesen 1990). Protective sentences under the 1991 Act display several weaknesses and anomalies which may be summarised as follows:- Firstly, the potential exists for offenders to be adjudged dangerous and receive protective sentences in circumstances where they do not have previous convictions for offences of a similar nature if the nature and circumstances of the offence indicate a strong likelihood that the offender may commit similar offences in the future; Crow and Pennington (1994) 16 Cr App R (S) 409. This produces an even weaker basis for accurate risk prediction than actuarial and predictive assessments which do, at least, rely on past behaviour as an important predictor of a future dangerous propensity. Some protection was provided for the offender in Crow and Pennington itself where the Court of Appeal insisted that the sentence should bear a 'reasonable relationship' to the offence for which it was imposed, although long enough to provide the necessary measure of public protection. Secondly, because sentences under s2(2)(b) must not exceed the statutory maximum for the offence, offenders may receive sentences which are only marginally longer than the commensurate sentence for that offence, but, inadequate to provide the measure of public protection demanded by the offender's perceived dangerousness.(15) The operation of the totality principle also ensures that the totality of the sentence is not disproportionate to the nature of the offending behaviour.⁽¹⁶⁾ Little wonder that Thomas (1993) was prompted to comment that many protective sentences are doubly ineffective since they are both too severe to be considered commensurate and, coincidentally, too short to provide effective public protection. Finally, where a discretionary life sentence is imposed on a violent and sexual offender, the sentencer must state in open court what proportion of the sentence has been imposed to reflect the seriousness of the offence under s2(2)(a) (see Practice Direction (Crime: Life Sentences) [1993] 1 WWLR 223). The Parole Board is currently obliged to review such sentences once offenders have served the commensurate part of their sentence, and, the offender may then be released on licence provided the Board is satisfied that it is no longer necessary for the protection of the public that the prisoner should continue to be confined (s34(4)(b) Criminal Justice Act 1991). However, offenders sentenced under s2(2)(b) who may have ceased to be dangerous have no corresponding right to have their sentences reviewed under the early release provisions contained in Part II of the 1991 Act, and must instead normally wait until they have served half the original term imposed by the court.⁽¹⁷⁾ In addition to the 1991 Act provisions, those violent and sexual offenders suffering from the requisite degree of mental abnormality may be subject to a disposal under the Mental Health Act 1983 (see Birch (1989) 11 Cr App R (S) 202) or, if otherwise appropriate, a discretionary life sentence subject to the restrictive criteria in Hodgson (1967) 52 Cr App R 113.

The introduction of automatic life sentences for serious violent and sexual offenders convicted a second time in s2 Crime (Sentences) Act 1997 eschews proportionality and selectivity in favour of the principle of collective incapacitation which may have marginal incapacitative effects (see Mathiesen 1990), in addition to serious ethical shortcomings. Since indecent assault is not punishable with life imprisonment and, therefore, does not qualify for an indefinite sentence (see Crime (Sentences) Act 1997 s2(5)) many sexual abusers of children will escape the new automatic life sentence. As Thomas (1996) correctly points out, the most persistent and harmful child abusers rarely commit offences punishable with life imprisonment. We have seen that those offences mostly frequently committed against children, namely indecent assault (maximum sentence 10 years) and gross indecency with a child (maximum sentence 2 years) are increasing consistently, yet they do not qualify for an automatic life sentence. Paradoxically, unlawful intercourse with a girl under 13, a notifiable offence whose incidence has declined over the last 10 years, is included in the definition of a 'serious offence' and, therefore, counts as a qualifying offence for the purpose of an automatic life sentence. Finally, although rape clearly counts as a qualifying offence for an automatic life sentence, the fine distinctions of gravity carefully drawn by Lord Lane C J in Billam (1986) 8 Cr App R (S) 48 (see also Ranyard, Hebenton and Pease 1994), which already treat the crime as aggravated by the fact that the defendant has previous convictions for rape or other serious violent or sexual offences, will be significantly compromised.⁽¹⁸⁾

It may, therefore, be concluded that the interests of child victims of violent and sexual crimes continue to be severely compromised by fundamental weaknesses in predictive techniques and the consequent tendency for contemporary legislative provisions to require judicial resolution of the dangerousness decision. However, not only have the judiciary been invested with the apparent ability to resolve the seemingly intractable problem of achieving the correct balance between harm (past and present) and future risk, no legislative machinery has been provided to facilitate this task. It has fallen to the Court of Appeal to endeavour to provide the necessary sentencing guidance which is considered more fully in the following section.

Appellate guidance

The main purpose of the Court of Appeal's sentencing principles is to provide guidance to Crown Court judges on the wide range of offences commonly dealt with by the Crown Court (Thomas 1979, Ch 1), but, the Appeal Court is also regarded as an important mechanism for shaping sentencing policy in the lower courts and for providing a degree of internal consistency (for criticism, see Wasik 1981). Its limited success in this respect was partially⁽¹⁹⁾ responsible for the development of the Magistrates' Association Sentencing Guidelines (see Wasik and Turner 1993), yet the latter do not provide any guidance on a wide range of sexual offences, such as indecent assault and gross indecency, which have particular relevance to children,⁽²⁰⁾ and, are rising rapidly. As regards offences at the more serious end of the penal severity scale, the Court of Appeal has been consistently criticised for failing to provide principled, relevant or comprehensive sentencing guidance (see Ashworth 1995, 1997 and Ashworth and von Hirsch 1997), and, it may be argued that the advent of the Attorney General's reference procedure under s36 Criminal Justice Act 1988 has made no material difference in this respect, since it was until recently limited to indictable offences, and, has failed to provide important guidance in cases on the borderline between custodial and non-custodial sentences (Henham 1994, Shute 1994).

The analysis which follows considers Court of Appeal sentencing guidance across a wide range of violent and sexual offences relevant to children:

(i) Homicide and non-fatal violence

As regards involuntary manslaughter, the range and combination of circumstances surrounding this offence vary almost indefinitely (Adler and Polk 1996). This is clearly reflected in the sentencing cases dealing with the manslaughter of young children, yet, it is equally true that the Court of Appeal regards public protection as the overriding sentencing principle; as Stocker L.J. put it in Bashford (1988) Cr App R (S) 329:

However, there is also an appreciation of the public's perception of just deserts and the need to deter acts of unofficial retaliation:

Consequently, the correct sentencing approach requires a strong emphasis on punishment and deterrence and, additionally, as appropriate, an element of denunciation to mark the public abhorrence of this sort of behaviour (see Drake J in Ali (1988) 10 Cr App R (S) 59). Although prison sentences in the range of three to seven years would appear appropriate where there is no question of deliberate grave violence, sentences at either end of this scale may exceptionally be justified, turning on the degree and length of exposure to the violence offered(see Thomas 1979, pp 79-83 and Lord Taylor C J in Tickle (1990) 12 Cr App R (S) 395). However, as in Brannan (1994) 16 Cr App R (S) 766, fine distinctions are drawn regarding those actions causing death and, I would suggest, a more rational, and, ultimately consistent approach, would result from narrative sentencing guidelines describing the full range of possible factors and their potential impact on penalty levels.⁽²¹⁾ The same comments apply to offences of violence against children, and, cruelty or neglect,⁽²²⁾ despite the Court of Appeal's efforts to provide more substantive form to its guidance in these areas. Where repeated acts of violence are concerned Lord Lane C.J. was again unequivocal in setting out the Court's approach in Sims - Rees (1985) 7 Cr App R (S) 120:

Nevertheless, the Court frequently experiences extreme difficulty in balancing the need to reflect serious acts of violence and callous ill-treatment of very young children against the strong mitigation provided by immature and inadequate parents frequently sharing grossly overcrowded accommodation with their children (see for example, Mahieu and Mahieu (1988) 10 Cr App R (S) 265, David (1994) 15 Cr App R (S) 867, Burtenshaw (1994) 16 Cr. App. R. (S) 227). However, the balance of competing interests may be resolved against the child victim where the parent responsible is not the child's father by an appropriate reduction in what would otherwise be the commensurate sentence (see Houghton (1991) 13 Cr. App. R. (S) 11). It may be argued that where the primary sentencing objective is public protection, child victims should be entitled to expect 'equality of consideration' in the determination of commensurate sentences and the usual 'progressive loss of mitigation' principle should be suspended.⁽²³⁾ All child-related offences should, therefore, be treated as equivalent to serious crimes (such as rape) where the gravity of the crime is regarded as having virtually extinguished any discount for mitigating circumstances. In any event, clear exposition and uniformity is lacking in the Court of Appeal's present approach:

The exact nature and differences in approach referred to are not elaborated and, although occasional judgments consider the significance of specific factors in depth,⁽²⁴⁾ there is no comprehensive sentencing guidance. The position is repeated with regard to cruelty and neglect offences. Despite the careful evaluation of specific factors by Auld J. in Ace (1990) 12 Cr. App. R. (S) 533, most cases stress the seriousness of the offence, eschewing the analysis of principle and approach in favour of determining the impact of a narrow range of factors upon sentence.⁽²⁵⁾ The parent's paramount duty is to protect his or her child (see Lord Taylor C J in Emery (1992) 14 Cr App R (S) 394), and the Court has been influenced by the fact that, occasionally, the victims of the offence may suffer more by the penalty than the perpetrator of the offence (see R v S [1996] 2 Cr App R (S) 256).

(ii) Sexual offences

The many reported cases of rape involving young children display a consistent application of the Billam sentencing guidelines which refer specifically to aggravating factors particularly relevant to child victims.⁽²⁶⁾ However, as in those cases involving adult victims, sentencers have displayed a considerable lack of sensitivity where child victims of rape are concerned (see Dalrymple, 1993), and there have been a significant number of rape cases referred under the Attorney General's reference procedure for 'unduly lenient' sentenced contained in S36 Criminal Justice Act 1988.⁽²⁷⁾ Reference decisions on rape have similarly seen a consistent application of the Billam guidelines,⁽²⁸⁾ but, as with most reference cases, the Court of Appeal has also given discount for the 'double jeopardy' factor without giving a clear indication of its extent, thus restricting the value of the case as future guidance. Further, public misconception of the purpose and efficacy of the reference procedure is compounded by the test of undue leniency propounded by Lord Lane C.J. in Attorney-General's Reference (No. 4 of 1989) [1990] 1 WLR 91 who stated that a sentence is unduly lenient 'where it falls outside the range of sentences which the judge, applying his mind to all the relevant factors, could reasonably consider appropriate'. This test was equated with the principle that public confidence in the judicial system would be lost if the error of principle committed by the trial judge was not corrected. However, the equation of public confidence with judicial errors of principle severity restricts the Court of Appeal's power to use the reference procedure constructively. Public opinion is mistaken in regarding the sentencer's task as one of primarily reflecting the public perception of undue leniency with due regard to sentencing convention, whereas, in reality, the approach adopted by the Court of Appeal is based primarily on sentencing convention, since it equates public perceptions of justice with the need to prevent judicial errors of principle.

Nevertheless, the Attorney-General's reference procedure has managed to produce some useful sentencing guidelines in incest cases,⁽²⁹⁾ where none previously existed, and also provided important guidance on the sentencing of offenders convicted of the buggery of young children, where no previous authority existed.⁽³⁰⁾ However, as concerns rape, I would argue that there is insufficient weight given to the trauma experienced by child victims (see Morgan and Zedner 1992a, 1992b) in the Billam guidelines, and sentencing guidelines developed specifically to deal with the unique experiences and circumstances of child rape victims should be developed. The need for specific sentencing guidance is even greater for cases of indecent assault on young children.⁽³¹⁾ Given its increasing numerical significance, it is alarming that there are no comprehensive sentencing guidelines dealing with the sentencing of child-related indecent assault,⁽³²⁾ notwithstanding that attention has been drawn in individual cases to specific factors, such as the vulnerability of young children staying with their father following marital breakdown (see Helliwell (1987) 9 Cr App R (S) 357), and, the desirability that sentencers should receive factual information as to the impact of the offending behaviour upon a victim.⁽³³⁾ This confusion of approach to sentencing indecent assault cases generally is aptly illustrated by Townsend (1994) 16 Cr App R (S) 353, where the Court of Appeal concluded that it had been wrong to state in Neem (1993) 15 Cr. App. R. (S) 18, on similar facts, that, as a matter of principle, an indecent assault case of that kind should not carry a custodial sentence on first conviction. The Court considered that it was at liberty to pass a non-custodial sentence in such circumstances although Neem (and its subsequent application in Chagan (1994) 16 Cr App R (S) 15), established that such a course was wrong in principle. Although not strictly binding precedent, the Court failed to explain why it chose to contradict a decision it had reached only a few months previously, thereby undermining its own authority, and, its own previously expressed desire to achieve a consistent approach in sentencing. Finally, although Lawton L.J. laid down sentencing guidelines for cases of unlawful sexual intercourse with a girl aged under 16 in Taylor (1977) 64 Cr App R (S) 183, few would not now admit that some of his Lordship's comments are wholly inappropriate in a contemporary context, for example:

In conclusion, it is also important to point out that there has been a consistent increase in sentence lengths for sexual offences in recent years (Sampson 1994, pp 53-58), and, as Sampson explains, it is impossible to know whether this results from some change in the nature of the offences themselves, or, a deliberate policy on the part of sentencers. However, Sampson correctly concludes that there is no evidence to support the latter proposition, especially given the manifest complexities of sentencing, rather, such evidence as does exist indicates a tendency by judges not to follow sentencing guidelines in a disproportionately large number of cases.⁽³⁴⁾ Nevertheless, the dominant trend is for sentence lengths to reflect more accurately the gravity of serious sexual offences. The explanation offered by Sampson that the increase results from public pressure upon sentencers rather than any real or increased understanding of the nature of the offences themselves is persuasive, but difficult to substantiate.

Conclusion

The evidence presented in this paper has described the serious incidence of child victimization and suggested that there are significant weaknesses in the protection offered by existing legislation and appellate guidance to child victims of violent and sexual crimes. This stems largely from the lack of any coherent sentencing policy dealing with child victims, and, a fundamental failure to recognise children as a separate victim category requiring special protection. It has, therefore, been argued that such a co-ordinated strategy could be developed by the Court of Appeal through the production of specific guideline judgments to deal with certain types of child-related crime (for example, rape and indecent assault). I suggest that it would also be helpful if the test of 'undue leniency' propounded by Lord Lane C.J. in Attorney-General's Reference (No. 4 of 1989) were to be re-formulated. As mentioned, the existing test is essentially concerned to ensure that judicial errors of principle are corrected rather than with any wider requirement to equate the reasonableness of a sentencer's approach primarily with public perceptions of justice. If the concept of undue leniency were to be expanded to include "failure to protect the public from substantial risk" it would greatly enhance the ability of the Court of Appeal to supply guidance in cases which involved the sentencing of dangerous offenders, and, as a consequence, provide a further opportunity for the Court of Appeal to produce guidelines for specific offences. In addition, it would ensure that the relative penal severity of child-related crimes is established and sustained within existing tariffs as part of a coherent sentencing policy. Increasing penalties through the introduction of mandatory minimum sentences for serious violent and sexual crimes in the Crime (Sentences) Act is likely to exacerbate the trauma for child victims.⁽³⁵⁾ Therefore, the development of criteria and guidance to achieve continuity of policy and practice in the sentencing of violent and sex offenders against children is now an urgent priority.⁽³⁶⁾

Bibliography

Alder C M and Polk R (1996) 'Masculinity and Child Homicide' British Journal of Criminology 396.

Ashworth A (1993) 'Victim Impact Statements and Sentencing' Criminal Law Review, 498.

Ashworth A and Hough M (1996) 'Sentencing and the Climate of Opinion' Criminal Law Review 776.

Ashworth A and von Hirsch A (1997) 'Recognising Elephants: The Problem of the Custody Threshold' Criminal Law Review 187.

Hartless J et al (1995) ' More Sinned Against Than Sinning: A Study of Young Teenagers' Experiences of Crime' British Journal of Criminology 114.

Henham R (1994) 'Attorney General's References and Sentencing Policy' Criminal Law Review 499.

Henham R (1996a) 'Sentencing Policy, Appellate Guidance and Protective Sentencing' Journal of Criminal Law 424.

Henham R (1996b) 'Back to the Future on Sentencing: The 1996 White Paper' Modern Law Review 861.

Home Office (1996) Criminal Statistics for England and Wales 1995, Supplementary Tables, London: Government Statistical Service.

Hood R and Shute S (1996) 'Protecting the Public: Automatic Life Sentences, Parole and High Risk Offenders Criminal Law Review 788.

Maguire M (1997) 'Crime Statistics, Patterns and Trends: Changing Perceptions and their Implications' in Maguire M, Morgan R and Reiner R (eds) The Oxford Handbook of Criminology, 2nd edition (Oxford: Clarendon Press).

Maung N A (1995) Young people, victimisation and the police: British Crime Survey findings on the experiences and attitudes of 12 to 15 years olds, Home Office Research Study No. 140 (London: HMSO).

Morgan J and Zedner L (1992a) Child victims: crime, impact and criminal justice (Oxford: Clarendon Press).

Morgan J and Zedner L (1992b) 'The Victim's Charter: A New Deal for Child Victims' Howard Journal of Criminal Justice 294.

Mirlees-Black C, Mayhew P and Percy A (1996). The 1996 British Crime Survey England and Wales, Home Office Statistical Bulletin, Issue 19/96 (London: Government Statistical Service).

Ranyard R, Hebenton B and Pease K (1994) 'An Analysis of a Guideline Cases as Applied to the Offence of Rape' Howard Journal of Criminal Justice 203.

Sampson A (1994) Acts of Abuse: Sex Offenders and the Criminal Justice System (London: Routledge).

Shute S (1994) 'Prosecution Appeals against Sentence - The First Five Years' Modern Law Review 745.

Straw J (1996) Honesty, Consistency and Progression in Sentencing (London: The Labour Party).

von Hirsch A (1986) Past or Future Crimes (Manchester: Manchester University Press).

von Hirsch A and Ashworth A (1996) 'Protective Sentencing Under Section 2 (2) (b): The Criteria for Dangerousness' Criminal Law Review 175.

Wasik M (1981) 'Sentencing Disparity and the role of the Court of Appeal' Justice of the Peace 348.

Wasik M and Turner A (1993) 'Sentencing Guidance for Magistrates' Courts' Criminal Law Review 345.

Watson L (1996) Victims of Violent Crime Recorded by the Police, England and Wales, 1990 - 1994, Home Office Statistical Findings, Issue 1/96 (London: Government Statistical Service).

Yates C (1992) 'The Legal Reality - Sentencing and the Intra-Familial Sex Offender' in Bradley G and Lamplugh D (eds) Sentencing Sex Offenders: The Current Crisis and Proposed Resolutions. Report of a Conference set up by the Suzy Lamplugh Trust in conjunction with the Criminal Bar Association, HMP Blundeston.

Zedner L (1997) 'Victims' in Maguire M, Morgan R and Reiner R (eds) The Oxford Handbook of Criminology, 2nd edition (Oxford: Clarendon Press).

1. The term 'children' is used to refer to those individuals under the age of 16. Back to text.

2. I do not suggest that child victims have any greater claim to be protected through the sentencing policies of the courts than any other victim group. Back to text.

3. The basic offence groups included in the notifiable offences recorded by the police are; Theft of/from vehicles, other theft, burglary, criminal damage, violence (including robbery), fraud and forgery, sexual offences and other. These categories give no indication whatsoever regarding the incidence of child-related crime. However, notifiable offences are largely coterminous with offences triable in the Crown Court (summary offences are not included) and, as such, it is claimed that they include the most serious crimes for which the vast majority of prison sentences are passed. Nevertheless, it is difficult to claim that many included incidents are any more serious than those not included, such as common assault, child cruelty and indecent exposure. Notwithstanding, notifiable offences provide the necessary starting point, since sentencing principles and legislation are clearly related to the same legally-derived categories. Back to text.

4. These were, for example, television campaigns and police force initiatives which encourage reporting of sexual offences against children. The majority of victims in this age group were aged 15. Back to text.

5. Watson confirms that long term trends show the number of recorded offences of indecent assault and buggery have increased at a rate of about 5 per cent per year over the period 1984-1994. The rate of increase for buggery exceeds that for indecent assault. Back to text.

6. The explanation given for the high victimisation rate is the same as for rape cases. Back to text.

7. Watson also examines rates for robbery and theft from the person, pp. 10, 11. Back to text.

8. In this case the victimisation rate for both sexes shows that the 16-24 year old age group were most at risk. Back to text.

9. See Mirless-Black, Mayhew and Percy 1996, p 12. In the light of subsequent discussion it is interesting to note that one third of serious crimes and two thirds of woundings were not reported to the police although regarded as serious by victims aged 16 or over. Back to text.

10. The wide definition of assault included much behaviour which, although technically a legal assault, would not normally be regarded as meriting further intervention by agents of the criminal justice system. Back to text.

11. Around 30 per cent of incidents involving 12 to 15 year olds were either reported to the police or another. Back to text.

12 The authors comment that existing surveys fail to look at the relevance of victimisation in relation to participants' current experiences and individual biographies. Are young males unafraid of street attack because the fear or actuality of it has not been part of their childhood experience? And, do young women, conversely, fear sexual attack precisely because it has been part of their individual or group youthful lives? Back to text.

13. It is interesting to note that the perception that 'no-on could do anything' increases with the victim's age. Back to text.

14. The extent to which such perceptions are a function of social and cultural values ultimately determined by social structure (see Hall et al 1978), or, whether they simply imply the need for a pragmatic educative programme (see Hough 1996, Ashworth and Hough 1996) is a contentious issue beyond the scope of this paper. Back to text.

15. See, for example, Attorney General's Reference (No. 4 of 1993) (Bingham) [1993] Crim. L R 795, Coull (1993) 15 Cr App R (S) 115, Apelt [1994] Crim LR 75, Lyons [1994] Crim LR 145. Back to text.

16. See Lord Taylor C.J. in Mansell (1994) 15 Cr App R (S) 771. In Grabbidon and Bramble [1997] Crim LR 137 the Court of Appeal resolved the conflict between Chapman (1994) 15 Cr App R (S) 844 and Christie (1994) 16 Cr App R (S) 469 (see Thomas 1996) by deciding that to achieve the correct balance when imposing a sentence under S2 (2) (b), the sentencer should calculate the appropriate commensurate sentence as if S2 (2) (b) did not apply, and then add an appropriate enhancement for the protection of the public. Back to text.

17. Existing arrangements for parole and early release are abolished under Part II, Crime (Sentences) Act (1997) but the provisions have not been implemented. For analysis of the White Paper which preceded the Act, see Henham (1996b). Back to text.

18. Proposals to change the supervision and sentencing of sex offenders were contained in Chapter 8, 1996 White Paper. Extended release supervision orders for sex offenders are contained in s20 Crime (Sentences) Act 1997 but have not been implemented, although the Home Secretary indicated in a Parliamentary Written Answer (Official Report, 30 July 1997, cols 261, 262) that alternative provisions would be included in the forthcoming Crime and Disorder Bill). Back to text.

19. A major contributing factor is that the Court of Appeal's decisions are necessarily a function of the limited range of appeals presented to it. Back to text.

20. Examples of other offences falling into this category include; indecency with children under 14 (Indecency with Children Act 1960 s1), cruelty to a child (Children and Young Persons Act 1933 s1), abduction of a child (Child Abduction Act 1984 ss1 & 2). Back to text.

21. A move in the right direction occurred in Attorney-General's Reference (No. 33 of 1996) (Latham) [1997] Crim. L.R. 140 when the Court of Appeal produced sentencing guidelines for cases of manslaughter resulting from the use of a knife carried in a public place with a view to it being used as a weapon. Back to text.

22. Offences under these headings include cruelty to persons under 16 (Children and Young Persons Act 1933 s1), abandoning a child under two years (Offences against the Person Act 1861 s27). Back to text.

23. The principle or notion of progressive loss of mitigation recognises that the culpability of the offender increases with successive convictions. Modern desert theorists have accepted that individual culpability is variable but limited to the extent that this mitigation should not override the principle that proportionality must be maintained between the punishment and the gravity of the offence. The notion of a 'ceiling' for the offence thus places a limit on the importance attached to the repetition of conduct by asserting that the overall penalty should be proportional to the gravity of the conduct manifested in the present offence. See von Hirsch, 1986.Back to text.

24. See, for example, Lord Taylor CJ in R v J (1996) 1 Cr App R (S) 20 in which the court asserted that it did not consider that self-inflicted intoxication or lack of judgment when taking drugs was a mitigating factor where an adult in charge of a young child had tortured him intentionally. Back to text.

25. See, for example, Colwell (1993) 15 Cr App. R (S) 323 in which the Court concentrated on the context in which the offence was committed, the impact on the victim and the circumstances of the appellant. In terms of overall approach Steyn LJ simply referred to the need for judicial discretion to be directed towards the issue of what is the minimum term necessary to mark the seriousness of the offence. Back to text.

26. Lord Lane CJ stated that the crime should be treated as aggravated by the following factors (inter aila); the victim is either very old or very young, the effect upon the victim whether physical or mental (in this respect Ashworth refers to the case of Attorney General's Reference (No. 19 of 1992) (McMorrow) (1992) 14 Cr App R (S) 330 where evidence of nightmares and personality changes was admitted on buggery and attempted rape charges involving child victims. Ashworth suggests that the terms of Billam appear to encourage the reception of such evidence which could equally be applied in the case of other sexual and violent crimes, (see Ashworth 1993). His Lordship also indicated that the extra distress which giving evidence can cause the victim should mean that, perhaps more so than in other cases, a guilty plea should normally result in some reduction from what would otherwise be the appropriate sentence (this fact must now be declared in open court following the implementation of Criminal Justice and Public Order Act 1994 s48). Further, the victim's previous sexual experience is irrelevant and, equally, previous good character is of minor relevance. Back to text.

27. For the period 1989-1996 thirty one rape cases had been referred to the Court of Appeal by the Attorney General. If the three cases which were withdrawn are included, this amounts to 11.19% of all cases considered by the Attorney General for referral during this period (Figures kindly supplied by the Legal Secretariat to the Law Officers). Back to text.

28. See, for example, Attorney-General's Reference (No. 7 of 1989) (Thornton) (1990) 12 Cr App R (S) 1 applied in Collier (1992) 13 Cr App R (S) 33. In Attorney-General's Reference (No. 36 of 1995) (Dawson) (1996) 2 Cr App R (S) 50, Lord Taylor CJ specifically stated that the trial judge had been wrong to take into account the fact that the girl (aged 11) had previously been corrupted in reduction of sentence. Indeed, his Lordship went further than Billam in stating, 'We consider that a girl - or a young boy for that matter - who has been sexually abused is more in need of sensitive, and proper treatment by adults subsequently'. Back to text.

29. Attorney-General's Reference (No. 1 of 1989) (1989) 11 Cr App R (S) 409, [1989] Crim LR 925. For interesting analysis see Yates (1992). Back to text.

30.Attorney General's Reference (No. 17 of 1990) (Jones) (1990) 12 Cr App R (S) 572, [1991] Crim. L.R. 487. The Court of Appeal enumerated five factors which a sentencer should take into consideration when determining the correct sentence for this type of case, namely; the overall gravity of the offence, the necessity for punishment of the offender, the necessity to protect the public from the activities of someone who is prepared to sniff solvent and then, having his sexual inclinations aroused, goes and commits this type of offence, the public concern at sexual offences on young children, the effect which a severe sentence may have upon other people who may be minded to act in this way.
Buggery committed after 3 November 1994 (the commencement date of Criminal Justice and Public Order Act 1994 ss143, 144) is charged as rape. This applies whether the victim is male or female. Back to text.

31. A model of sentencing guidance was provided for buggery cases in Willis (1974) 60 Cr. App. R. 146 where Lawton L.J. considered the general approach before carefully listing and discussing the following aggravating factors in detail:- physical injury to the boy, emotional and psychological damage, moral corruption, abuse of authority and trust, and, similarly, the following mitigating factors:- mental imbalance, personality disorders, emotional stress. Back to text.

32. Such general guidance as exists in Bibi (1980) 2 Cr. App. R. (S) 177 and Freeman (1980) 2 Cr. App. R. (S) 194 is lacking in detailed analysis of the correct approach and potential impact of aggravating and mitigating factors. Back to text.

33. See Attorney-General's Reference (No. 2 of 1995) (S) (1996) 1 Cr. App. R. (S) 274. Ashworth (1993) suggests that, although there are substantive counter-arguments of principle and procedural implications, the strongest argument for introducing victim impact statements is where the actual consequences have been held relevant to sentence. The argument of principle is stronger where the offender has selected particularly vulnerable victims (such as children). Back to text.

34. It is interesting to note, however, that Ranyard, Hebenton and Pease (1994) suggest that the Billam guidelines may well have served to increase the average length of sentences imposed in rape cases. Back to text.

35. For example, automatic life sentences will remove the incentive to plead guilty thereby resulting in increased trauma or humiliation for victims giving testimony. Victims may also refuse to testify in cases where close relatives are involved because of emotional pressure, knowing that conviction will result in a life sentence. It should also be noted that automatic life sentences are likely to increase control over only a minority of sexual and violent offenders. (Hood and Shute 1996). Back to text.

36. Proposals aimed at achieving consistency and progression in sentencing by developing guidelines judgments across a range of offences are desirable(see Straw 1996) yet there is no doubt that the introduction of automatic life sentences for 'serious' violent and sexual offences will instead distort the relative gravity of different offences. Back to text.

United Kingdom Journals