Jonathan Glover's review of
Punishment: The Supposed Justifications
by Ted Honderich
At least since the eighteenth century, the philosophy of punishment has been a scene of activity and even progress. There is now, if not moral agreement, at least far greater clarity of thought about the various principles that can be cited in answer to the questions that arise about the morality of punishment. Other social practices and institutions have had less attention. There has been no comparable progress in the philosophies of war or of income distribution. And the philosophy of marriage and the family scarcely exists. The energy devoted to the principles of punishment does not simply stem from the fact that Bentham developed utilitarianism with penal questions in mind. The field's popularity comes also partly from the same causes as its success. 'The justification of punishment' seems a manageable topic. A rather small number of competing moral principles can be applied to a social practice that can be considered in relative (though only relative) isolation from other aspects of society. The practice can be criticized in the light of the principles, and it can at the same time be used to test their plausibility.
In his rational and civilized book, Dr. Honderich contributes to further progress in this field. He has found himself, as he says, 'dealing with the several large moralities which come into clear conflict over punishment and may, indeed, be appraised by way of it'. In this book, the interaction is largely one way. Honderich's own basic moral position is stated early, and emerges unscathed at the end. Alternative views are stated, and are criticized from the standpoint of his own beliefs. His moral system is a two-valued one: it consists of utilitarianism combined with a belief in distributive justice. He obviously shares the view he ascribes to 'many men' on p. 9, when he says that 'their impartial evaluations of actions begin from two fundamental attitudes, one having to do with securing human welfare and avoiding its contraries, the other with the distribution of welfare and its contraries'. Later he says (p. 37) that 'we have taken the position that individuals should not be deprived or made to suffer, no matter what they have done, if insufficient good in the form of human welfare or equality comes of their experience'.
In the course of the book, Honderich develops guiding penal principles based on his two fundamental values, and contrasts these with traditional principles based on retribution, deterrence and reform. These three traditional justifications are given a chapter each. One at a time, each is examined to see whether or not it is plausible on its own as a sufficient justification of punishment. This piecemeal method of considering one question at a time is characteristic of the book's direct and workmanlike approach. But it has here the disadvantage that it leads to a distortion of emphasis. Does any philosopher these days believe that retribution alone, deterrence alone, or reform alone is all that need be considered in justifying punishing someone? All the most plausible theories involve appeals at least to some combination of these factors. But all more sophisticated theories of this kind receive rather brief treatment together in a single chapter called 'Compromises'.
The discussion of retributivism is unusual. The version presented of the theory makes it appeal to consequences of punishment, as well as look back to the culpable behaviour that constituted the offence. It is made part of the 'interpretation' (p. 23) of the claim that a man deserves a particular penalty that 'the penalty will give satisfactions equivalent to the grievance caused by his action'. This seems scarcely plausible even as part of an analysis of what is ordinarily meant by claims about desert. No doubt some who demand retribution do sometimes want the satisfaction of seeing the guilty man punished, but their desires are surely irrelevant to whether or not the punishment is deserved. If they all died, would he no longer deserve the punishment ?
Honderich's own moral attitudes are so enlightened that he seems to find purely backward-looking retributive beliefs too incredible to consider. He is charitable to the retributivist in a way that turns retributivism in part into a rather feeble utilitarianism. The view that desert, analysed in this way, is a sufficient condition of punishment is then easily dismissed. As the book argues, when we punish the guilty, the resulting satisfaction of those with grievances will normally in utilitarian terms be 'purchased at too great a cost'.
The discussion of deterrence is very much more convincing. It is immediately pointed out that mere appeals to deterrence cannot justify giving whatever punishment is necessary to deter people. We would again be in danger of obtaining benefits at too great a cost in misery. Instead, the far more plausible principle of 'economical deterrence' is formulated. According to this, two conditions must be satisfied for a punishment to be justified. It must prevent more distress than it causes, and it must deter more effectively than any less severe punishment.
Honderich discusses the traditional objection to deterrent theories that they would justify victimization: the 'punishment' of innocent people. After dismissing some versions of this objection that are too crude (and after saying that in some conceivable circumstances victimization could be justified) he agrees that even the principle of economical deterrence could justify some morally unacceptable victimizations. Hence economical deterrence is not a sufficient justification for punishment.
Reform is similarly dismissed as a sufficient justification for punishment or for compulsory 'treatment'. The treatment might cause more distress than its consequences would justify; it might contribute to the growth of undesirable non-rational methods of persuasion; a policy of treatment might consume more of society's resources than a utilitarian could justify. For all these reasons, it is argued that considerations other than reform alone are relevant to justifying either punishment or 'treatment' for crime.
In the chapter on 'Compromises', some persuasive arguments are advanced against views proposed by Ross, Mabbott, and Armstrong. There is also a critical discussion of the main contribution to the philosophy of punishment in recent years: that of Hart. Honderich denies the value of separating questions of general justifying aim, of amount of punishment, and of the proper recipient of punishment. He prefers to consider as a single question 'what, if anything, can morally justify the practice of punishment?', arguing that there is no reason why treating the problem as a single one need blind us to the variety of aspects of punishment. It is hard to see exactly what is at issue here. It is clear that we have to decide whether or not to set up or maintain a system of prisons, fines, etc., and that when we have such a system we have the further problems of deciding how and when it is to be used. A prime purpose of Hart's separation of these questions was presumably to show that we do not have to appeal to exactly the same values in answering these questions. If we think that in deciding how to use the penal system we ought to avoid unfair punishment, we do not have to invoke fairness as one of our reasons for setting up the system. Honderich says that by treating the problem as expressible as a single question, we need not 'fall into the confusion of thinking that a single-reason answer to the moral question about punishment will do' (p. 142). But if this is so, it is not clear that there is anything of substance at stake in this disagreement over how many questions there are.
There is a more substantial disagreement between Honderich and Hart over the question of the requirement that, in the absence of strong enough reasons to justify overriding this principle, we ought only to punish those who were responsible for their acts at the time of breaking the law. Hart's utilitarian arguments for this principle (that its acceptance will reduce insecurity, etc.) are accepted, but his objections to the unfairness of punishments that break the principle are rejected. It is clear that there is a fundamental difference of attitude here. Hart believes that desert, in the normal 'backwarcJJiooking' sense rather than in Honderich's special sense of the word, is relevant to the fairness of a punishment. He believes in retributive justice to the extent of wanting in general to reject as unfair any punishments that are undeserved. Honderich, consistently with his view that acts ought to be judged solely by their consequences for the maximization and equalization of happiness, rejects even this negative appeal to retributive justice. He also adds, in a somewhat inconclusive chapter on free will, that his own approach, excluding even this sort of direct reference to desert, has the advantage of not presupposing anything about our freedom to act differently from how we do act.
The book's general conclusion is that 'what can justify punishment is the reduction of distress at an economical rate, usually to be obtained by imposing penalties on those responsible for offences, and the fact, if it is one, that the practice contributes to the avoidance of greater inequality' (p. 173).
There is something very attractive about a book that approaches punishment from so uncompromising a consequentialist standpoint. After a period in which the advocates of utilitarianism have often been timid and defensive, trying to explain away differences between their own and more conventional moralities, it is good to see a radical utilitarian critique of a major social institution. The book's principles are applied to a number of questions not mentioned in this review, including strict liability and exemplary penalties. There is also an interesting concluding chapter on the question of what sorts of act ought to be compelled or prohibited by law, which discusses Mill's views and the 'enforcement of morals' debate.
There are some general criticisms that can be made of the book. One (relatively minor) criticism concerns the style. The arguments are honestly and clearly stated, but the writing is often cumbersome and at times ungrammatical ('I have, I am afraid, already said about as much as I am able to bring into literalness and clarity in support of my contention' (p. 129). More serious is the total absence of argument in support of Honderich's own views of justice as equality. Why should we accept that a just distribution of welfare is an equal one irrespective of desert (in the normal sense) ? And it would have been good to see rather more argument in support of the view that utilitarianism needs to be supplemented by a principle of equality rather than left on its own or supplemented by other quite different principles.
But it is good to have the implications for punishment of one plausible and attractive set of values clearly spelled out, as they are in this sane, thorough and socially radical book.
Jonathan Glover's review of Punishment: The Supposed Justifications (Hutchinson, 1969) appeared in the journal Inquiry in June, 1970. The book was enlarged and published again under the title Punishment: The Supposed Justifications Revisited (Pluto Press, 2006). There is a review of this edition by John Williams.
HOME to T.H. website front page
HOME to Det & Free front page