1. By “punitive,” I mean “inflicting or intended to inflict punishment; retributive, punishing.” I leave aside its other meaning, namely, “Of a tax or other charge: extremely high, severe; (also) prohibitive, damaging” (Oxford English Dictionary). In conversation I have found that some understand “punitive” to mean “like punishment.” I leave aside that definition, too.

2. John Kleinig, The Ethics of Policing (1996), at 99–100. According to Kleinig, the police should not inflict punishment for the same reasons that Locke thought that private individuals should not carry out justice: they do a poor job of it. On this view, the properly drawn implication is not that punitive police action is unjust for the procedural reason that it is insufficiently downstream in the criminal justice process, but that punitive police action is in fact likely to be unjust; implicitly in this argument, it might not be unjust.

3. Strictly, the central claim of my argument need not be that the police actions I focus on are punishment. My central claim is that some police actions fall under the normative standards that also govern punishment. That is, they fall under of the normatively relevant elements of any conception of punishment. In my view, the distinction is extremely fine, but my argument can honor it. For instance, it may be the case that “imposed by a legal court-like authority” is a nonnormative element of the definition of punishment. Thus, I allow that, as some will plausibly claim, the normal application of the words “punishment” and “punitive” involve courts, prisons, and so forth. In fact I doubt that such a definition is useful, but there are some things one might say in support of it: we in fact hesitate to describe punishment outside of the state as punishment proper; it is vigilantism, ersatz justice in place of legitimately administered punitive treatment. In any case, the claim of this paper is that the coercive and reprobative structures of the police, with their claims to legitimacy, are relevantly similar to those of courts, and so the relevant sort of legal authority is present. That is, I require only that the court-like element of the definition (if it exists) must be normatively inert in this context. However, I leave aside this complication in the main text of the paper, since if I am correct, the concept of “punishment,” so conceived, is an arbitrarily limited subject of normative study, and we would more usefully reconceptualize it in order it to include police action.

4. Williamson, Tom & Bagshaw, Peter, The Ethics of Informer Handling , in Informers: Policing, Policy, Practice (Billingsley, Roger, Nemitz, Teresa & Bean, Philip eds., 2d ed. 2013)Google Scholar, cited in Steve Hewitt, Snitch!: A History of the Modern Intelligence Informer (2010), at 29.

5. Dunnighan, Colin & Norris, Clive, Some Ethical Dilemmas in the Handling of Police Informers , 18 Pub. Money Mgmt. 21, 24 (1998)Google Scholar.

6. Id. at 24.

7. Fyfe, Nicholas & Sheptycki, James, International Trends in the Facilitation of Witness Co-operation in Organized Crime Cases , 3 Eur. J. Criminology 319 (2006)Google Scholar; Hewitt, supra note 4; Harfield, Clive, Police Informers and Professional Ethics , 31 Crim. Just. Ethics 73 (2012)Google Scholar; Dunnighan, Colin & Norris, Clive, Risky Business: The Recruitment and Running of Informers by English Police Officers , 19 Police Stud. Int'l Rev. Police Dev. (1996)Google Scholar. A classic study of this and related practices is Goldstein, Joseph, Police Discretion not to Invoke the Criminal Process: Low-Visibility Decisions in the Administration of Justice , Yale L.J. 543 (1960)CrossRefGoogle ScholarPubMed.

8. Rich, Michael L., Brass Rings and Red-Headed Stepchildren: Protecting Active Criminal Informants , 61 Am. U. L. Rev . 1433, 1442 (2012)Google Scholar. See also Alexandra Natapoff, Snitching: Criminal Informants and the Erosion of American Justice (2009).

9. Kleinig, John, Punishment and the Ends of Policing , in Liberal Criminal Theory: Essays for Andreas von Hirsch (Simester, A. P., Neumann, Ulfrid & Bois-Pedain, Antje du eds., 2014)Google Scholar.

10. See Stillman, Sarah, The Throwaways , New Yorker , Sept. 3, 2012 Google Scholar; Kuo, Susan S., Official Indiscretions: Considering Sex Bargains with Government Informants , 38 U.C. Davis L. Rev. 1643 (2005)Google Scholar.

11. As Michael Moore says, “the giving of punishment is dangerous to virtue.” A Tale of Two Theories, 28 Crim. Just. Ethics 27, 42 (2009). This fact infects the entire criminal justice system, from criminal courts to prosecution services to policing to prison officers.

12. “‘When I use a word,’ Humpty Dumpty said in a rather scornful tone, 'it means just what I choose it to mean—neither more nor less.’—‘The question is,’ said Alice, ‘whether you can make words mean so many different things.’—‘The question is,’ said Humpty Dumpty, ‘which is to be master—that's all.’” Lewis Carroll, Through the Looking-Glass (1871). For discussion see MacKay, Alfred F., Mr. Donnellan and Humpty Dumpty on Referring , 77 Phil. Rev. 197 (1968)CrossRefGoogle Scholar.

13. Thus, John Rawls refers to deprivations that have the aim of general deterrence but are meted out on those who are known to be innocent as “telishment.” Rawls, John, Two Concepts of Rules , 64 Phil. Rev. 3 (1955)Google Scholar.

14. Hart, H. L. A., Prolegomenon to the Principles of Punishment , in Punishment and Responsibility: Essays in the Philosophy of Law (2008)CrossRefGoogle Scholar.

15. Analogously, see Robert Nozick on the proper allocation of healthcare resources and barbering resources. Some argue that healthcare resources should be distributed only according to health needs, and that anyone who denies this has misunderstood the nature of healthcare. Nozick responds by noting that this argument may apply to any other good, and that in so applying it, one realizes the argument's circularity: one still requires a reason to conceive of the good in the way that one does. Anarchy, State, and Utopia (1974), at 233–235.

16. In further support of this point, consider the reasoning of the European Court of Human Rights in Welch v. UK (1995). The Strasbourg court was asked to consider whether or not the United Kingdom had imposed a criminal penalty. The court's conclusion rested on the reasoning that the criteria for criminal punishment and penalties have “autonomous meaning.” That is, it is not necessary or sufficient for an act to be punitive that it is labeled as “punishment” by the authority carrying it out; rather, courts should look to the nature of the act itself. Similarly, as Hale LJ wrote, dissenting, in R (Smith) v. Parole Board (2005): “to the person concerned it is experienced as punishment, whatever the authorities may say.” Cited in Andrew Ashworth & Lucia Zedner, Preventive Justice (2014), at 158. Thus, in deciding the applicability of Article 7 of the European Convention on Human Rights’ prohibition on retrospective punishment, courts must decide for themselves the relevant meaning of its constituent concepts. In the U.S. context, the legal position may be different:

Kolber, Adam J., Unintentional Punishment , 18 Leg. Theory 1, 5 (2012)Google Scholar, citing Kansas v. Hendricks, 521 U.S. 346 (1997). However, it may be that by “punitive intent,” we should understand “intention to impose deprivation or censure”—in which case, the position is one that I deal with in Sections III.C and III.D. In any case, legal decisions are not decisive for my purposes. The courts are seeking a definition in a legal context; they are seeking, for example, as the court was in Welch, to understand the proper implementation of the procedural element of Article 7 ECHR. They are not required to consider the wider question of the reach of the normative standards that are applicable to punishment in general. Furthermore, it is possible in any case that the courts misapply the concepts. The argument here is normative, not legal. So we cannot resolve this issue by looking at the state of the law.

17. Scheid, Don E., Note on Defining “Punishment ”, 10 Can. J. Phil. 453 (1980)CrossRefGoogle Scholar.

18. In a broader context, Antony Duff offers the point in the following way: “[The] whole process of investigation and trial can involve burdens much like those imposed by punishment: loss of freedom, for those detained for questioning or pending trial; loss of money, in lost earnings or legal costs; serious intrusions on one's time; the shame or embarrassment of being thus investigated, tried and exposed to public scrutiny; and the sense of being subjected to the forceful disciplinary power of the state. Such burdens also fall on those who are in the end not punished.” Antony Duff, Punishment, Communication, and Community (2003), at xiii.

19. Human Rights and the Moral Responsibilities of Corporate and Public Sector Organisations (Tom Campbell & Seumas Miller eds., 2004), at 184.

20. The case of the management of minor infractions by way of an implied threat to arrest has a similar structure. Instead of bringing an individual within the formal criminal justice system, a person is given the chance merely to move on, surrender contraband, or accept a reprimand.

21. Not all amalgams of unattractive options are themselves cases of hard treatment. For example, the choice “obey the law, or go to prison” limits one's options, but it seems a stretch (depending on one's view of the legitimacy of the state) to say that this is an imposition of a disadvantage. This remains the case even if one would benefit from unpunished criminality.

22. E.g., David Boonin, The Problem of Punishment (2008), at 12–28.

23. Robert Nozick, Philosophical Explanations (1981), at 380.

24. Due to Kolber, supra note 16, at 3.

25. A further option is that state punishment is constituted by some element of the judge's intention, say, the given number of years of incarceration, but not by any specific facts about that incarceration. This view has problems of its own: a long period of intentionally lenient and comfortable incarceration would count as a heavy punishment. Furthermore, we would need some method or justification for the selection of only a subset of the judge's intention, and in providing this, we either modify the definition of punishment (“intentional deprivation measured by years of incarceration”), or locate the relevant type of intention outside of the mind of the judge—which would be to adopt option (3).

26. Kolber, supra note 16. See also Kolber, Adam J., Against Proportional Punishment , 1141 Vand. L. Rev. (2013).Google Scholar

27. Jonathan Bennett, The Act Itself (rev. ed. 1995), at 222.

28. For useful discussion see Francis Delaney, Neil, Two Cheers for “Closeness”: Terror, Targeting and Double Effect , 137 Phil. Stud. 335 (2008)CrossRefGoogle Scholar.

29. See Quinn, Warren S., Actions, Intentions, and Consequences: The Doctrine of Double Effect , Phil. Pub. Aff. 334, 341ff. (1989)Google Scholar.

30. This would be an expansion of Rawls's category, mentioned in note 13 above.

31. Victor Tadros, The Ends of Harm: The Moral Foundations of Criminal Law (2011).

32. “Punishment is a conventional device for the expression of attitudes of resentment and indignation, and of judgments of disapproval and reprobation, on the part either of the punishing authority himself or of those ‘in whose name' the punishment is inflicted.” Joel Feinberg, The Expressive Function of Punishment, 49 Monist 397, 400 (1965).

33. A central defense of this view of punishment is Andrew von Hirsch, Censure and Sanctions (1996).

34. Husak, Douglas, Why Criminal Law: A Question of Content? , 2 Crim. L. Phil. 99 (2008)Google Scholar.

35. Here is another way in which police censure functions: it is possible for the censure that is formally attached to one transgression to attach, de facto, to another. In concerted preventive investigations, police will seek to identify any possible infraction, however minor, carried out by somebody who they suspect is carrying out a more significant crime. This is a case in which police use their powers in a way that is plausibly legitimate, and in a way that expresses public censure for actions other than those that provide the formal institutional authority for those powers. How else can we understand a police decision to deploy their resources in a determined way, so that somebody is arrested for something, anything? The expression of censure is not fully captured by the censure involved in punishing or arresting someone for their misclaimed benefits or their failure to maintain their car; it also exists in the public decision to align resources in such a way that the individual will be placed under scrutiny that is costly to him or her, and that decision about the alignment of resources is made in response to a wrong that it is believed the individual has committed. It is difficult to construe this as anything except hard treatment carrying with it a public expression of censure for crimes that have gone, formally, unpunished. The formal censure involved in investigation and prosecution for minor infractions can also be informal censure for major infractions.

36. Even though jurisdictions vary with regard to the circumstances in which a person's criminal record may be revealed publicly, not all in the literature on expressivist or communicative theories make this distinction. E.g., Husak, supra note 34, at 18; Kessler Ferzan, Kimberly, Beyond Crime and Commitment: Justifying Liberty Deprivations of the Dangerous and Responsible , 96 Minn. L. Rev. 183 (2011)Google Scholar. For further discussion of the distinction between stigma and censure, see Hadjimatheou, Katerina, Criminal Labelling, Publicity, and Punishment , 35 Law & Phil. 567 (2016)CrossRefGoogle Scholar.

37. Skillen, A. J., How to Say Things with Walls , 55 Philosophy 509 (1980)CrossRefGoogle Scholar.

38. Cordner, Gary, Community Policing , in The Oxford Handbook of Police and Policing (Reisig, Michael D. & Kane, Robert J. eds., 2014)Google Scholar, at 165.

39. In further support of this point, it is notable both how narrowly full enforcement statutes have been interpreted, and also that they have not been universally adopted. There is reason to view such statutes only as “broad statements of purpose and not as rigid requirements.” Kleinig, John, Selective Enforcement and the Rule of Law , 29 J. Soc. Phil. 117, 121 (1998)CrossRefGoogle Scholar.

40. According to one UK study, “one third of all crimes cleared up by the police involve . . . informers.” Informers: Policing, Policy, Practice (Roger Billingsley, Teresa Nemitz & Philip Bean eds., 2013), at 5. It has been suggested to me that this is likely to be true for serious crime, but not crime in general; in any case, it remains a highly significant practice. See note 4 above for references concerning the ubiquity of the possibility of prosecution as a motivation for informers.

41. For development of what this means from a natural law perspective, see Thomas V. Svogun, The Jurisprudence of Police: Toward a General Unified Theory of Law (2013).

42. Jeffrey Reiman objects directly to informant recruitment by way of a threat to prosecute: “. . .it is a power which we would not grant to police over all citizens, since if we wanted that we would pass a law requiring all citizens to give police whatever information the police want.” Against Police Discretion, 29 J. Soc. Phil. 132, 137 (1998). This argument is unsuccessful: that we do not grant police this power over all citizens is consistent with our granting it over some.

43. It might be argued that consistently following the argument of this paper would commit one to the view that prosecutorial power also can fall under the heading of punishment, especially where it is applied in order to obtain a plea bargain. This issue will depend on further understanding the category of punishment. It is plausible that a normal arrest is not punishment, since it aims at facilitating the courts’ procedures and so is not in itself censorious. A similar case may be made for normal use of prosecutorial discretion. In the United Kingdom the Crown Prosecution cannot increase or decrease charges in accordance with value of getting a conviction, and the option of offering to drop some charge in order to obtain a guilty plea is not directly available. I discuss plea bargains further below.

44. A classic assertion of this and other such principles is in Hart, supra note 14.

45. Tadros, Victor & Tierney, Stephen, The Presumption of Innocence and the Human Rights Act , 67 Mod. L. Rev. 402 (2004)CrossRefGoogle Scholar.

46. For discussion of the idea of a direct moral grounding of criminal justice procedures, see Tomlin, Patrick, Extending the Golden Thread? Criminalisation and the Presumption of Innocence , 21 J. Pol. Phil. 44 (2013)CrossRefGoogle Scholar.

47. Calvin Jeffries, John & Stephan, Paul B., Defenses, Presumptions, and Burden of Proof in the Criminal Law , 88 Yale L.J. 1325, 1327 (1979)CrossRefGoogle Scholar. Cited in Tomlin, supra note 46, at 49.

48. See Laudan, Larry, Is Reasonable Doubt Reasonable? , 9 Legal Theory 295 (2003)CrossRefGoogle Scholar.

49. Rakoff, Jed S., Why Innocent People Plead Guilty , N.Y. Rev. Books, Nov. 20, 2014 Google Scholar.

50. Gov.UK, Criminal Court Statistics (Quarterly): January to March 2016, https://www.gov.uk/government/statistics/criminal-court-statistics-quarterly-january-to-march-2016. Thanks to Antony Duff and an anonymous Legal Theory reviewer for urging this point on me.

51. Bowers, Josh, Punishing the Innocent , 156 U. Pa. L. Rev. 1117 (2008)Google Scholar.

52. A different type of plea bargain involves an early plea of guilty to the charges brought, with the expectation of a reduced sentence, and a commitment from the prosecutor not to seek a higher sentence.

53. Richard L. Lippke, The Ethics of Plea Bargaining (2011), at 221.

54. Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness (rev. ed. 2012), at 84–89. See also Rakoff, supra note 49.

55. Natapoff, supra note 8.

56. Rich, supra note 8.

57. He states: “such cases fall outside the scope of this Article.” Id. at 1461 n.154.

58. Husak, Douglas, Preventive Detention as Punishment? , in Prevention and the Limits of the Criminal Law (Ashworth, Andrew, Zedner, Lucia & Tomlin, Patrick eds., 2013), at 179 Google Scholar.