Principles of sentencing - incapacitation

Incapacitation
"Prison works"	Another commonly stated function of the criminal law is the protection of society by the incapacitation of dangerous offenders. Crime is reduced by restricting the offender's opportunity to commit further crimes. "Prison works" was the phrase used by Michael Howard, when he was Home Secretary, at the Conservative Party Conference in 1993. Ministry of Justice provided evidence that of more than 1,700 inmates were let out 18 days early in 2007 in an attempt to ease the pressure on jails. Six had committed a total of eight new crimes after being let out and another 18 had gone missing after failing to report in regularly. Thirty were recalled to jail within a few days.
Children, mentally ill and petty offenders as prisoners	The former head of the Prison Service Martin Narey told the Observer newspaper, in July 2006 that that thousands of children, mentally ill people and petty offenders were in prison that did not belong behind bars. The Home Secretary has announce that he will build more jails, rather than pledge any early cut in numbers. Home Secretary Douglas Hurd, under Margaret Thatcher reduced the number of prisoners by 4,000.
Other sentences that protect society.	Most societies allow for some special form of protective sentencing. The Criminal Justice Act 1991 broadly endorses the concept of sentencing proportionate to the seriousness of the crime but allows longer sentences to be imposed in cases of violent or sexual offences 'if it is necessary to protect the public from serious harm from the offender' (section 2(2)(b) ). Similarly, with the wide array of offences carrying a maximum penalty of life imprisonment (for example, rape, causing grievous bodily harm with intent, aggravated burglary, etc.) the sentencing judge can impose such a maximum penalty upon an offender perceived to be dangerous to society. Finally, it must be remembered that non-custodial protective sentences can be imposed. Thus disqualifying an offender from driving is regarded as one of the most effective means of protecting road-users from dangerous drivers.
Variations in sentencing produce injustice.	Two widely differing examples of protective sentencing will help reveal the nature of, and the difficulties with, utilising the criminal law and the process of punishment for this purpose. In Nicholls (1970) the offender was convicted of one count of indecent assault on a girl aged seven. He had no less than nine previous convictions for indecent assault and attempted rape of children aged between four and nine years of age. The court imposed an extended sentence of ten years' imprisonment on the basis that the offender was a danger to the public. While extended sentences have now been abolished, a court could today still achieve the same result by utilising section 2(2)(b) of the Criminal Justice Act 1991. The other case is a famous and controversial United States Supreme Court decision, Rummel v. Estelle (1980) in which a sentence of life imprisonment for obtaining $120.75 by false pretences was affirmed. The defendant had two previous convictions: fraudulent use of a credit card to obtain $80 worth of goods (nine years previously), and passing a forged cheque of $528.36 (four years previously). He was prosecuted under a Texas recidivist statute making life imprisonment mandatory upon a third felony conviction. A majority in the Supreme Court rejected a claim that life imprisonment was a grossly disproportionate punishment and held that it was legitimate for a state to segregate recidivists from the rest of society 'for an extended period of time'.
What is "dangerous"?	The contrast between these two cases leads us directly to a consideration of two central problems associated with sentencing on such bases. First, how is 'dangerousness' to be defined for these purposes? Most people would agree that Nicholls with his persistent sexual aggression constituted a 'danger' to children. But can the same be said of Rummel? If protective sentencing is to be employed, it must surely be limited to those who are dangerous in the sense of being likely to cause serious harms. Of course, this approach in itself raises problems. While most would agree that death, serious bodily injury and serious sexual assault clearly qualify as serious harms for this purpose, controversy arises in relation to a host of other harms, for example, 'loss or damage to property which results in severe personal hardship' (Floud Committee, 1981). One thing does, however, seem certain. It could not be seriously asserted that the kind of offence Rummel might in future commit could possibly justify the imposition of a sentence of life imprisonment.
Predicting re-offending.	The second problem is one of predicting or identifying dangerousness. How probable must it be that Nicholls or Rummel will offend again before we are justified in imposing an incapacitative sentence? How immediate, how frequent and how specific must the risk be? Numerous research projects have been undertaken aimed at developing accurate criteria for predicting dangerousness. For instance, Greenwood (1982) developed a seven-factor prediction index (based on previous criminal history, drug use and unemployment) to identify persons likely to commit frequent acts of robbery or other violent crimes in future. The problem with this and many other projects is that research has been based on the admissions of imprisoned offenders only, which has prompted the cynical response that this is 'like trying to learn about the smoking habits of smokers generally by studying the smoking activity of residents of a lung cancer ward' (Von Hirsch, 1986). Even more serious is the fact that such projects to date have tended to yield rather too many 'false positives' (classifications of dangerousness that did not materialise).
Punishing people on likelihood of future behaviour.	This last point is critical. A 'false positive' is not just a statistic. It is a human being locked up in prison because society has made a mistake in wrongly predicting him to be dangerous. Accordingly, until such time as more accurate prediction criteria have been developed (if this were ever possible) it must be highly questionable whether punishment for such reasons can be justified. There is a related question of whether it can ever be permissible to punish people on the basis of what they might do in the future.