¹ The distinction was originally introduced by Mackie, J., Persons and Values (Oxford, 1985), pp. 207–8.Google Scholar

² For a critical discussion of different versions of positive retributivism see, for instance, Ten, C. L., ‘Positive Retributivism’, Social Philosophy & Policy 7 (1990), pp. 194–208.CrossRefGoogle Scholar

³ This distinction is meant only as a rough way of categorizing costs, here serving the purpose of directing attention to different parts of the work that is required to fulfil an obligation to punish. Thus, whether the sentencing by the court is placed in one category or another is not important.

⁴ Avio, K. L., ‘Economic, Retributive and Contractarian Conceptions of Punishment’, Law and Philosophy 12 (1993), pp. 249–86CrossRefGoogle Scholar, at 264–5.

⁵ Obviously, this does not imply that modern retributivists tend to support the penal systems in the precise forms in which they exist today (for instance, many retributivists have argued for leniency in punishment compared to existing penal levels in many countries). But this does not change the fact that retributivists usually believe that a state-driven punishment system is justified.

⁶ See, for instance, F. Heidensohn and L. Gelsthorpe, ‘Gender and Crime’, or Hoyle, C. and Zedner, L., ‘Victims, Victimization, and Criminal Justice’, both in Oxford Handbook of Criminology, ed. Maguire, M., Morgan, R. and Reiner, R. (Oxford, 2007), pp. 381–42Google Scholar and 461–95 respectively.

⁷ Obviously, this view does not exclude the possibility that the obligation to track down and apprehend criminals may also be sustained for other reasons (e.g. out of consideration for the victims). For a discussion of the relation between the obligation to punish perpetrators and the obligation to track them down and apprehend them, see also Ryberg, J., ‘Racial Profiling and Criminal Justice’, The Journal of Ethics 15 (2011), pp. 79–88.CrossRefGoogle Scholar

⁸ For the use of the phrase pro tanto see, for instance, Kagan, S., The Limits of Morality (Oxford, 1991), p. 17.CrossRefGoogle Scholar

⁹ It should be recalled that – and I shall return to this in the final section – in real life resources are scarce, which means that there are good reasons to believe that positive retributivist prescriptions cannot simply be fulfilled without being confronted with the resource priority challenge.

¹⁰ von Hirsch, A. and Ashworth, A., Proportionate Sentencing (Oxford, 2005), pp. 12 and 27CrossRefGoogle Scholar.

¹¹ For instance, deontologists of the Kantian school might – like Kant himself – subscribe to the view that there exists a positive obligation to assist poor people. Or they might subscribe to a range of special positive obligations to protect and support family and friends. See, for instance, Kagan, S., Normative Ethics (Oxford, 1998)Google Scholar, ch. 4.

¹² Lippke, R. L., ‘Retributive Parsimony’, Res Publica 15 (2009), pp. 377–95CrossRefGoogle Scholar, at 383. Lippke's main point is that retributivists can urge reductions in the use of punishment in order to conserve scarce resources for other valuable social purposes. But obviously, even if this is correct, it does not resolve the problem considered here, namely, how the obligation to punish should be weighed against other obligations when resources are limited.

¹³ It is a fact that many retributivists subscribe to the existence of thresholds in relation to deontological constraints. Obviously, one might do the same in relation to positive obligations (that is, by holding that there is a positive obligation as long as the negative consequences of fulfilling it are below a certain threshold; if the threshold is transgressed, the obligation no longer exists). However, it should be noticed that the mere reference to thresholds does not succeed in solving the problem addressed here, namely, how priorities between different obligations should be set. For instance, if I have a positive obligation A and a positive obligation B and cannot fulfil both, and if the negative consequences of not fulfilling either A or B do not amount to a transgression of the thresholds for each obligation (or transgress the thresholds of both obligations), then it still remains an open question whether I should fulfil A or B.

¹⁴ Hart, H. L. A., Punishment and Responsibility (Oxford, 1968), p. 9.Google Scholar

¹⁵ Cahill, M. T., ‘Retributive Justice in the Real World’, Washington University Law Review 85 (2007), pp. 815–70Google Scholar, at 833.

¹⁶ For a more thorough discussion of this type of implication, see Ryberg, Jesper, ‘Punishment and Desert-Adjusted Consequentialism’, Retributivism Has a Past – Has it a Future?, ed. Tonry, M. (New York, 2011), pp. 86–100.CrossRefGoogle Scholar

¹⁷ See, for instance, Dolinko, D., ‘Retributivism, Consequentialism, and the Intrinsic Goodness of Punishment’, Law and Philosophy 16 (1997), pp. 507–28Google Scholar; or Dolinko, D., ‘Restorative Justice and the Justification of Punishment’, Utah Law Review 1 (2003), pp. 319–42.Google Scholar

¹⁸ See, for instance, Davis, M., ‘Punishment Theory's Golden Half Century: A Survey of Developments from (about) 1957 to 2007’, The Journal of Ethics 13 (2009), pp. 73–100.CrossRefGoogle Scholar