1 The ‘compromise’ position arose out of Rawls', John piece ‘Two Concepts of Rules,’ Philosophical Review 64 (1955)CrossRefGoogle Scholar, and found its most persuasive statement in Hart, H.L.A. ‘Prolegomenon to the Principles of Punishment’ in his Punishment and Responsibility (Oxford: Clarendon Press 1968)Google Scholar. Other versions are offered by Packer, Herbert L. The Limits of the Criminal Sanction (Stanford, CA: Stanford University Press 1968)Google Scholar, Gross, Hyman A Theory of Criminal Justice (New York: Oxford University Press 1979)Google Scholar, and Williams, Glanville Textbook of Criminal Law second ed. (London: Stevens & Sons 1978)Google Scholar, among others.

2 See Goldman, Alan H. ‘The Paradox of Punishment’ Philosophy and Public Affairs 9 (1979)Google Scholar.

3 See the various contributors to ‘Symposium: The New Retributivism,’ Journal of Philosophy 75 (1978). See as well Davis, Lawrence H. ‘They Deserve to Suffer’ Analysis 32 (1972)CrossRefGoogle Scholar; Murphy, Jeffrie G. ‘Marxism and Retribution,’ Philosophy and Public Affairs 2 (1973)Google Scholar; and his Retribution, Justice, and Therapy (Boston: Reidel 1979); Card, Claudia ‘Retributive Penal Liability,’ American Philosophical Quarterly Monograph Series 7 (1973)Google Scholar; Haag, E. van den Punishing Criminals: Concerning a Very Old and Painful Question (New York: Basic Books 1975)Google Scholar; Hirsch, Andrew von Doing Justice: The Choice of Punishments (New York: Hill & Wang 1976)Google Scholar; Finnis, John Natural Law and Natural Rights (Oxford: Clarendon Press 1980)Google Scholar; and Nino, C.S. ‘A Consensual Theory of Punishment,’ Philosophy and Public Affairs 12 (1983)Google Scholar.

4 It was brought to my attention by a referee of this journal that it is important in characterizing transparent persuasion to distinguish between offering transparent reasons that actually do support the moral judgment and offering reasons that one believes supports the judgment. I agree, and have phrased the requirement of transparent persuasion in this latter way throughout. Nonetheless, it is not clear whether Duff is concerned to separate these two possibilities. He writes, for example:

What is crucial here is that I am trying to bring Jasper to behave as he ought to behave: it is the belief that his present conduct is morally wrong that both motivates and justifies my attempt to get him to behave different. In such a case the reasons which I offer him for behaving differently should properly be identical with the reasons which justify my belief that he ought to behave differently and my attempt to persuade him to behave differently. (51)

Here Duff moves quickly from justification that flows from my mere belief to reasons which (actually) justify my belief. Now although, as we shall see, Duff does not think that moral assertions are rationally decidable, he does - because he must- think that we can identify reasons for our, or our society's, moral claims that are clearly applicable to the justification for these claims. Nonetheless, as a moral requirement, surely all that is expected of us as sincere moral critics is that we persuade by employing what we take to be transparent reasons.

It should also be remarked that transparent reasons differ from the motives we may have for engaging in moral criticism. As the referee reminded me, our motives often depend on the kind of relationship that exists between us. I will likely be moved to morally criticise my son or daughter, a relative, or close friend, by different motives than those that move me to criticise a complete stranger. Moreover, the reasons that justify these motives may well be different (and indeed, some motives may be transparent while others are not). Nonetheless, if Duff is correct, the reasons I may legitimately employ by way of moral criticism must be of the same sort in either case.

5 It should be remarked that there are many activities that might be called acts of ‘blaming.’ For example, to blame someone for something may merely be a way of ascribing responsibility, or it may be a way of informing a third person of our attitudes toward the agent, or again it may be a way of informing the agent herself what we think she did something wrong. It is only this last sense that Duff has in mind.

6 Duff's use of the terms ‘penance’ and ‘repentance’ is somewhat confusing. Penance, on occasion, merely describes a specific form of suffering (recognized guilt), on others, a form of moral understanding, and on others still, a way of disowning or disavowing my wrongful conduct. Repentance, on the other hand, Duff typically takes to be an active renunciation of past offences, an act which is assisted by penance (see 246-54). I have taken the liberty of treating penance as the suffering one experiences once one recognizes one's guilt. Thus understood, one can speak, as Duff often does, of ‘imposed penance.’ Repentance, on the other hand, I treat as a state of understanding that counts as a motive for action, specifically reform.

7 The most powerful and persuasive - though unfortunately little known - challenge to the intelligibility of the retributivist intuition is Árdal's, Páll piece ‘Does Anyone Ever Deserve to Suffer?’ Queen's Quarterly 91/2 (1984)Google Scholar.

8 I borrow the term ‘desert-locution’ as well as the list of examples from Ted Honderich ‘Punishment, the New Retributivism, and Political Philosophy’ in Griffiths, A. Phillips ed., Philosophy and Practice (Cambridge: Cambridge University Press 1985)Google Scholar.

9 Duff's ‘most promising’ retributive theory is derived from works (cited supra note 3) by Murphy, Finnis and von Hirsch.

10 I depart slightly from Duff's terminology here because he is slightly inconsistent in his use of ‘coercion’ and ‘manipulation.’ He writes, for example, that ‘When penance is imposed on her, she is certainly coerced…’ (253, my emphasis) although a few lines later he states that ‘For her imposed penance then still addresses her as a rational moral agent: it seeks to engage and arouse, but not to manipulate or coerce, her understanding and her judgment…’ (ibid., again my emphasis). To resolve this, I use ‘coercion’ to refer to any essentially persuasive technique which does not violate autonomy. It seems to me that there is nothing odd about saying that ‘my understanding was coerced’ by a knock-down argument to which I had no satisfactory response. Duff's principle distinction is, I think, retained, despite this modification of his terminology.

11 I owe this point to my colleague Páll Árdal.

12 I take it that this is the tack taken by Ronald Dworkin in his discussion of ‘the obligations of community’ in Chapter Six of his Law's Empire (Cambridge, MA: Harvard University Press 1986).

13 See especially Unger, R.M. Knowledge and Politics (New York: The Free Press 1975)Google Scholar, Chapter 1. Other critics of liberalism who are similarly tempted to draw this inference are Sandel, Michael J. Liberalism and the Limits of Justice (Cambridge: Cambridge University Press 1982)Google Scholar and (although less overtly) Arblaster, Anthony The Rise and Decline of Western Liberalism (New York: Blackwell 1984)Google Scholar and Shapiro, Ian The Evolution of Rights in Liberal Theory (New York: Cambridge University Press 1986)Google Scholar.

14 Supra note 13. I say this in part because such a view would seem to commit one to the dubious proposition that to understand a moral position is, necessarily, to accept and adopt it.