² This paper is not concerned with the very real problem presented by persons who make threats to specific individuals, typically spouses in the context of a marital or custody dispute. See, for instance, Walker, Nigel, Sentencing: Theory, Law and Practice, London, 1985, p. 364Google Scholar.

³ See, for instance, Walker, pp. 108–12.

⁴ They universally support the second approach as against the first. However, as pointed out below, it is open for them to support the third approach ahead of the second.

⁵ The term ‘reductivism’ appears to have been coined by Walker, Nigel, who confesses to its being ‘rather ugly’. Sentencing in a Rational Society, Harmondsworth, 1977, p. 18Google Scholar. Reductivism ‘holds that the justification for penalising offences is that this reduces their frequency’. Walker, Nigel, Punishment, Danger and Stigma, Oxford, 1980, p. 26Google Scholar. Walker points out that this term is more precise than ‘utilitarianism’, ‘since utilitarianism is really concerned with maximizing the sum of human happiness, and could conceivably argue that penalties severe enough to make a real impact on the frequency of, say, motoring offences would generate more unhappiness than they would prevent’. Punishment, pp. 26–7. Certainly, the benefits of reducing crime, or perhaps rather, of reducing the total harms of crimes and punishments taken together, need not necessarily be viewed in utilitarian terms.

⁶ On the distinction between these two types of deterrence, see Ten, C. L., Crime, Guilt, and Punishment, Oxford, 1987, p. 7Google Scholar.

⁷ That is, cases in which a particular rehabilitative programme requires that the offender be incarcerated for longer than the normal term. Although rehabilitation is a sentencing aim in its own right, it is treated here purely as a reductivist means.

⁸ See Ellis, Anthony, ‘Recent Work on Punishment’, Philosophical Quarterly, xlv (1995)Google Scholar. These various ways of attempting to reduce crime raise different epistemological issues. The main problem with deterrence and rehabilitation lies in ascertaining the extent to which the required type of effect has been achieved. Marginal judgements are notoriously difficult to make. The problem with incapacitation is quite different. Prisons clearly incapacitate, whatever else they do (ignoring the complications of crimes committed in prison, or planned in prison and committed outside). The difficulty lies in identifying those likely to commit serious crimes, in determining who is ‘dangerous’.

⁹ And, furthermore, if protection is not outweighed by other considerations (for instance, rehabilitation), which generally point towards a shorter sentence (despite the possibility mentioned in n. 7).

¹⁰ This term is also used by Nigel Walker. See, for instance, ‘Dangerousness and Mental Disorder’, Philosophy, Psychology and Psychiatry (Royal Institute of Philosophy Supplement: 37), ed. Griffiths, A. Phillips, Cambridge, 1994Google Scholar.

¹¹ H. L. A. Hart, drawing on Anthony Flew and Stanley Benn, defined ‘the standard or central case of “punishment”’ in terms of five elements which one writer summarized as follows: ‘hard treatment intentionally inflicted on a person who has offended against a legal rule, by an authority constituted by the relevant legal system’. Primoratz, Igor, ‘Punishment as Language’, Philosophy, lxiv (1989), 187CrossRefGoogle Scholar. See ‘Prolegomenon to the Principles of Punishment’, Punishment and Responsibility, Oxford, 1968, pp. 4–5Google Scholar.

¹² The quarantine analogy is not as uncontroversial as might be indicated here. It is defended by Schoeman, F. D. in ‘On Incapacitating the Dangerous’, American Philosophical Quarterly, xvi (1979)Google Scholar, reprinted in Sentencing, ed. H. Gross and A. von Hirsch, New York, 1981, but rejected by Floud and Young, pp. 39–45. Another analogy is remand, in so far as it is used for protective purposes, to prevent the prisoner from reoffending, as opposed to administration of justice purposes, to prevent him from absconding.

¹³ Walker, , Sentencing in a Rational Society, p. 19Google Scholar, Bentham, Jeremy, An Introduction to the Principles of Morality and Legislation, ed. Burns, J. H. and Hart, H. L. A., London, 1982, ch. 15, paras. 11–13Google Scholar.

¹⁴ The Oxford History of Prison: The Practice of Punishment in Western Society, ed. Morris, Norval and Rothman, David, Oxford, 1996Google Scholar.

¹⁵ Floud and Young, p. 20. See also Schoeman.

¹⁶ Ten, pp. 135–6.

¹⁷ For graphic illustrations, see Walker, Nigel, ‘Dangerousness and Mental Disorder’, p. 183Google Scholar, Meek, John, ‘The Revival of Preventive Detention in New Zealand 1986–93’, Australian and New Zealand Journal of Criminology, xxviii (1995), 225CrossRefGoogle Scholar, ‘Judge's Lenience Leads to Tragedy’, Melbourne Age, 02 17, 1996Google Scholar.

¹⁸ On the other hand, it may be argued that it is part of an official's professional capacity to be able to resist such pressure. Expert evidence may be of considerable assistance here. See White, Bob and Haines, Fiona, Crime and Criminology: An Introduction, Melbourne, 1996, p. 53Google Scholar.

¹⁹ This is certainly not the only objection. A further objection concerns the extension of civil detention to dangerous persons generally irrespective of whether they can be classified as dangerous offenders or not. (This could be regarded as a form of net-widening additional to that just mentioned.) In ‘Dangerous Offenders, and the Morality of Protective Sentencing’, I argue that, subject to perhaps insuperable objections concerning the identification of such persons, their civil detention is in principle justified (424). A record of violent crime is evidentially significant for this purpose – past violent conduct is the best predictor of future violence, as Walker, Nigel is keen to point out (‘Dangerousness and Mental Disorder’, pp. 186–7)Google Scholar. I maintain, however, that such a record is not morally significant. This view is defended against the Floud Report's contention that, once a person has committed sufficient serious offences to warrant being classified as ‘dangerous’, he should lose the benefit of what the Report refers to as the presumption of harmlessness.

²⁰ As noted in n. 7, rehabilitation may be treated as an end in its own right, but here we are concerned with it only as a reductivist means. Denunciation too is regarded sometimes as an end in its own right, and sometimes as another reductivist means.

²¹ This is not to deny that there may equally be harsh incapacitative measures. Execution, deportation and castration are graphic examples. (Other non-harsh incapacitative measures include licence cancellation.) Note that by concentrating on ‘hard’ rehabilitative measures, it is not denied that ‘soft’ rehabilitative measures can be highly manipulative in character, being paternalistic and infringing rights. Also, there may be deterrent and ‘hard’ rehabilitative value implicit in the mere denial of liberty, an un-intended reductivist benefit over and above the intended incapacitative value.

²² Intuitively, it seems, the more unpleasant they are, the greater their deterrent value. However, confirming such correlation through empirical evidence is another matter. It is quite another matter again whether making social harm reduction institutions more unpleasant increases their ‘hard’ rehabilitative value.

²³ Of course, whether the necessary resources are provided is another matter. Indeed, it seems, resources should be provided to finance monetary compensation for loss of income and career interruption as well.

²⁴ The latter type of constraints, of course, are observed only by those criminal justice systems which respect proportionality, and hence reject extended sentences.

²⁵ Such a social harm reduction system bears similarities with a criminal justice system reorganized on lines recommended by Barbara Wootton. It is beyond the scope of the paper to explore the points of comparison, but for a concise summary, see Ten, ch. 6.

²⁶ See, for instance, A. von Hirsch, Past and Future Crimes.

²⁷ Consider, for instance: Rawls's, John rule-utilitarian theory of punishment (‘Two Concepts of Rules, Philosophical Review, lxiv (1955))Google Scholar; H. L. A. Hart's theory, which distinguishes between the question of the general justifying aim of punishment, and the question of distribution (of who to punish, and how much to punish) (Punishment and Responsibility, esp. ch. 1); and C. L. Ten's ‘compromise theory’, according to which retribution and reduction are both necessary to the justification of punishment in every individual case (Ten, p. 79).

²⁸ D. J. Galligan, for instance, holds that ‘[w]e need to draw an essential distinction between retribution as a moral principle which we might employ in our personal lives, and retribution as a social principle which justifies institutionalised state coercion of individuals’. ‘The Return to Retribution in Penal Theory’, Crime, Proof and Punishment: Essays in Memory of Rupert Cross, ed. Tapper, C. F. H., London, 1981, p. 153Google Scholar. See also Goldman, Alan H., ‘Toward a New Theory of Punishment’, Law and Philosophy i (1982), 60–1Google Scholar.

²⁹ For instance, Galligan writes: ‘…modern penal measures especially of a custodial kind … typically involve, over and above the loss of liberty, impositions which are injurious, brutalising, often painful, and deeply degrading. These might logically be separate from the official punishment, but while they are inherent in any modern penal system, they might be considered as part of punishment and taken into account in any adequate explanation of that practice’. Galligan, p. 154.

³⁰ Hart, p. 4.

³¹ Ibid., p. 15.

³² Ibid.

³³ Furthermore, such deterrent benefit may be overall disadvantageous, if it functions to cancel out a greater potential rehabilitative benefit.

³⁴ von Hirsch, A. discusses the ‘interdependence’ of desert and deterrence (Doing Justice: The Choice of Punishments (Report of the Committee for the Study of Incarceration), New York, 1976, pp. 49–55)Google Scholar, but the relation between them investigated here is quite different.

³⁵ And presumably ‘hard’ rehabilitation as well.

³⁶ See n. 19.

³⁷ As Galligan points out against the likes of Hart and Rawls: ‘[t]o many people there is a deeply embedded intuition that part at least of the general purpose of criminal justice is the correction of wrongs, quite apart from any consequential contribution to crime control’. Galligan, p. 151. Ten, C. L. similarly holds that ‘retributive considerations’ can provide ‘positive reasons in support of punishment’. Ten, p. 84Google Scholar.

³⁸ See Lyons, David, The Forms and Limits of Utilitarianism, Oxford, 1965, and Ten, pp. 67–71CrossRefGoogle Scholar. As Ten puts it: ‘[b]ut why would anyone, motivated solely by utilitarian considerations, act in accordance with a rule even on those rare occasions when it is known that obeying the rule will fail to produce the best consequences?’. Ten, p. 70.