Skip to content

Oxford Uehiro Prize in Practical Ethics: Can it be Wrong For Victims to Report Crimes?

  • by

This essay was the winning entry in the graduate category of the 6th Annual Oxford Uehiro Prize in Practical Ethics.

Written by University of Oxford student, Maya Krishnan.



Late one night in Managua, Nicaragua, a man punched Leslie Jamison in the face and then ran away with her camera. Jamison called the police. Forty minutes later, a police truck pulled up with a man in the back. A sense of discomfort informs Jamison’s subsequent narration of the incident in her essay collection, The Empathy Exams (2014). Jamison found herself occupying a morally fraught role: that of a white American in Nicaragua who got the police to try to hunt down a likely significantly poorer man. Had she done something wrong by calling the police?

Jamison’s experience calls attention to an under-investigated area in moral philosophy: the ethics of reporting crimes. While those few philosophers who have addressed this topic have sought a basis for duties to report crime or misconduct,[1] the ethics of reporting crimes can also inform in the other direction. For in practice, the criminal justice system is often a central mechanism through which injustice is perpetuated against disadvantaged people.[2]

This paper asks whether victims of crimes sometimes ought not report those crimes to the authorities, and answers in the affirmative. The paper identifies two factors that victims of crimes should consider when deciding whether to report. These factors are (1) the excess punishment that a sentence for a “guilty” verdict would impose over and above a just sentence, and (2) the relative degrees of privilege that the victim and the perpetrator enjoy.

Part One: Clear Disproportionality in Punishment

There are cases in which the consequences of reporting a crime are so high that it is clear that the victim of a crime should not report. Attending to a clear case will help identify principles that can be used in more difficult cases.

Consider the example of theft laws in Saudi Arabia, according to which thieves can have fingers or hands amputated as punishment for theft (Metz 1993, p.202). Suppose that a shoplifter, S, steals a phone from the shop of a merchant, M. The harm that M sustains in forfeiting the phone and forgoing the enforcement of M’s rights is vastly exceeded by the harm S would suffer in having S’s fingers or hand cut off. M should not report the incident of shoplifting to the police.

This case suggests that one factor that victims must consider in reporting crimes is whether the punishment that the perpetrator stands to receive is proportionate to the crime. It is widely held that proportionality in sentencing is a necessary condition on the justice of a sentence.[3] While views regarding what exactly proportionality requires varies among theorists, it is relatively uncontroversial that the severity of a crime correlates with the severity of deserved punishment. If the punishment that a perpetrator stands to receive is very disproportionate, and the victim sacrifices relatively little by forgoing reporting, then the victim should not report the crime.

While the current literature on sentencing in philosophy of law focuses on the question of what sentences the law or judges should impose,[4] the case discussed here makes clear that proportionality is not just a concern for lawmakers and judges. If M were to report the crime of S, then M would be the cause of inappropriately imposed and foreseeable harm to S. Although the criminal justice system would also be a cause in this and similar cases, a victim of crime would not thereby absolve themselves of blameworthiness for harming the perpetrator.

Part Two: Harder Cases

  • The Dilemma

In the type of case considered above, the harms that the perpetrator stood to suffer from the justice system were vastly in excess of a just sentence. Furthermore, the harm that the victim would suffer in leaving the crime unaddressed seemed to be small. It was therefore relatively clear that reporting the crime would be wrong. But there are harder cases. Suppose that the victim has suffered significant harm, and the punishment, while disproportionate, is not wildly disproportionate. In order to clarify the principles at work in such cases, I now consider a simplified and abstract example that is not linked to any particular set of laws or legal system.

Suppose that P is the sole perpetrator of a crime for which V is the sole victim. And suppose that the just punishment for P’s particular criminal act would have severity of n units (where the severity of a punishment is a function of its nature and duration, and corresponds to an amount of harm brought about by the punishment). However, due to the nature of the laws or the criminal justice system in the society of which P and V are part, P will be given a sentence of severity n+m if convicted (where m > 0). So if P is convicted, P ’s sentence will be m units in excess of a just sentence. Suppose as well that the harm to V, if there is no involvement by the criminal justice system and the crime therefore goes entirely unpunished, is o. So if V does not report the crime, V will sustain o units of harm more than is fair. This suffering could be understood in different ways — for instance, as the harm of an unaddressed injury to V’s dignity, or as the psychological discomfort caused by the lack of redress.[5]

The previous section considered cases in which the value of m was high and the value of o was low, and in which it correspondingly seemed appropriate for the victim to suffer o instead of the perpetrator suffering n+m. But what about a case in which m = o? That is, what about the case in which the harm the victim would face from not prosecuting and securing a conviction is equivalent to the harm that the perpetrator faces in excess of a just sentence?

It might seem obvious that in such cases, P is the one who should suffer in excess. After all, P is the perpetrator and V is the victim. But since ex hypothesi P’s committing the crime only makes it the case that P deserves a punishment of severity n, it is not obvious why P should also sustain the additional m units of suffering.

One might hold that, by acting culpably, P is the one who causes it to be the case that there are m units of suffering that one of either P or V must sustain, and that P is therefore the one who ought to sustain the extra suffering.

This seems plausible.[6] But while P did act culpably, P’s acting culpably is not what is causally explanatory of there being an extra m=o units of suffering that P or V must sustain. What explains that is the flaws of the laws or the criminal justice system. For if P were to have committed the same crime in a society with just laws or a just legal system, P would have been assigned a sentence of severity n. It therefore seems that P’s culpable action is explanatory of P being assigned a sentence, but not of there being an extra m=o units of suffering that either P or V must sustain. But it is the extra m=o units we seek to assign to one of P or V. It is not, therefore, clear that P stands in any special relation to the m=o units of suffering, such that it would clearly be appropriate for P to sustain that harm.

The basic difficulty is that we are trying to find out who should suffer the extra m=o units. But it is the flaws of the laws or the criminal justice system, rather than P or V themselves, that explains why one of either P or V stands to receive not merely a burden but an unjust burden. Neither P nor V is at fault, or even responsible, for the fact that there are an extra m=o units of suffering that must be disbursed. However, neither “the system”, nor the lawmakers who put the relevant laws into place, are available to shoulder the burden. The resulting dilemma can only arise in non-ideal circumstances: Which of P or V ought to take on a burden that neither should have to take on?

(b) Responsibility for structural injustice

Here we have a case of structural injustice, or injustice in which it seems that no particular individual is to blame. The classic analysis of responsibility for structural injustice is Iris Marion Young’s “collective responsibility” account, according to which those who contribute to structural injustice have “an obligation to join with others… in order to transform the structural processes to make their outcomes less unjust” (2011, p.95).

But while the kind of injustice that has generated the extra units of suffering is plausibly structural, the collective political response does not work because it has the wrong time-horizon. The relevant change in sentencing procedures would require years if not decades to take effect, but the victim has to make a decision in the near future. Moreover, it is the victim who faces the decision whether to prosecute, rather than a collective or a group.

(c) Individual Responsibility in an Unjust World

The solution that I propose appeals to an extension of the notion of strict liability to address liability for injustice. Strict liability is imposed by law when someone violates an obligation not to injure another party (Gardner 2001). Strict liability comes apart from desert (i.e. what an agent deserves) (McMahan 2017), and does not require negligence or wrongdoing (Gardner 2001).

I propose the following principle: in general, those who are relatively privileged are liable to sustain harms imposed by injustice. General strict liability would then provide a second consideration that is relevant for victims of crimes to consider. In the simplified case in which m=o and all else is equal, the more privileged of P or V is liable to sustain the harm of injustice. Liability therefore provides a normative basis for assigning a burden to one of P or V.

The key problem facing this solution is to provide a justification for holding that the relatively privileged are liable to sustain harms imposed by injustice. The strategy I now pursue in addressing that problem is to identify a moral basis for strict liability and appeal to a version of that moral basis in the case of injustice. Although the moral basis for strict liability has been called into question (see Nagel 1976), Tony Honoré has famously defended its moral basis by arguing that strict liability is about holding agents responsible for “engaging in risk-creating activity” (1999, p.101). According to Honoré, to act is to make a “bet” regarding the benefits and risks of so acting, and strict liability is justified insofar as it secures a fair distribution of risk among agents (p.26, p.80).

When one acts under conditions of injustice, one’s actions contribute to the maintenance and perpetuation of unjust conditions (Young 2011). Those who are privileged are, statistically speaking, likely to benefit from the injustice under which they act. If one is committed to holding that the probabilistic harms for individual actions should be fairly distributed among beneficiaries and those who stand to be harmed, one should also endorse an analogous principle for statistical harms. On this view, those who are privileged are liable to sustain harm in cases in which doing so would benefit someone who is less privileged. Moreover, this principle of general strict liability is needed to sustain the integrity of individual strict liability. There cannot be a fair distribution of risk among individual agents if better outcomes are, on the whole, being assigned to members of some groups at the expense of members of others.

Moving beyond the consideration of simplified cases in which m=o, one can say that in general the relative degree of privilege of victim or perpetrator is a consideration that informs the morality of reporting crimes. If the sentence is unjust, then when weighing the excess harm to the perpetrator against the harm to the victim, the relative privilege of the victim and the perpetrator informs how much weight to provide to the interests of each.


The scope and strength of the demands that general strict liability makes on particular individuals requires further investigation. Furthermore, a more complete treatment of this topic would provide a systematic account of the circumstances under which being strictly liable for injustice would give rise to a normative basis for assigning burdens.

It would be worthwhile to embark on a more complete investigation of strict liability for injustice. Strict liability is a promising theoretical resource for addressing structural injustice, insofar as it assigns burdens in cases in which individuals are not at fault. Strict liability is particularly helpful because it assigns burdens to individuals. For while structural injustice as such may not be the responsibility of individuals as such, structural injustice creates circumstances in which individuals are, as a matter of fact, the only available ontological unit (so to speak) for discharging the relevant burdens. The cases described in this paper thus suggest a limitation of a recent trend within analytic social and political philosophy: that of holding that theories of structural injustice (and apt responses to it) should operate primarily at a non-individual level (see Schwartzman 2006, Young 2011, Haslanger 2015).

It is the central contention of this paper that in some cases, crimes should not be reported. Such cases can include instances in which victims sustain significant harm as a result of the perpetrator’s actions. The injustice of the criminal justice system makes reporting a crime into a moral dilemma for victims. In the sort of world we occupy, victims considering whether to report should consider the interests of those they expose to the unfairness of the criminal justice system.



Alexander, Michelle and Cornel West, (2010). The New Jim Crow: Mass Incarceration in the Age of Colorblindness. The New Press.

Belloni, Frank and Jacqueline Hodgson, (2000). Criminal injustice: an evaluation of the criminal justice process in Britain. Palgrave Macmillan UK.

Bottoms, Anthony, (1995). “The philosophy and politics of punishment and sentencing.” From The Politics of Sentencing Reform, ed C.M.V. Clarkson. Clarendon Press.

Chiao, Vincent, (2015). “Mass Incarceration and the Theory of Punishment.” Criminal Law and Philosophy.

Delmas, Candice, (2014). “The Civic Duty to Report Crime and Corruption.” The Ethics Forum, 9:1.  

Duff, R.A., (2001). Punishment, Communication and Community. Studies in Crime and Public Policy. Oxford University Press.

Feinberg, Joel, (1990). The Moral Limits of Criminal Law, Volume IV: Harmless Wrongdoing. Oxford University Press.

Gardner, John, (2001). “Obligation and Outcomes in the Law of Torts.” From Relating to Responsibility: Essays in Honour of Tony Honoré on his 20th Birthday. Hart Publishing.

Haslanger, Sally, (2015). “Social structure, narrative and explanation.” Canadian Journal of Philosophy.

Honoré, Tony, (1999). Responsibility and Fault. Hart Publishing.

Jamison, Leslie, (2014). The Empathy Exams. Graywolf Press.

Kolber, Adam, (2013). “Against Proportional Punishment.” Vanderbilt Law Review, 66.

McMahan, Jefferson, (2017). “Proportionate Defense.” From Weighing Lives in War, ed. Jens David Ohlin, Larry May, and Claire Finkelstein.

Metz, Helen Chapin, (1993). Saudi Arabia: a country study. Federal Research Division: Library of Congress.

Nagel, Thomas, (1976). “Moral Luck.” Proceedings of the Aristotelian Society, Supplementary Volumes, 50.

Sadurski, Wojciech, (1985). “Distributive Justice and the Theory of Punishment.” Oxford Journal of Legal Studies. 5:1.

Sakala, Leah (2014). “Breaking Down Mass Incarceration in the 2010 Census: State-by-State Incarceration Rates by Race/Ethnicity.” Prison Policy Initiative. <>.

Sawyer, Wendy and Peter Wagner (2019). “Mass Incarceration: The Whole Pie 2019”, Prison Policy Initiative. <>.

Schwartzman, Lisa, (2006). Challenging Liberalism: Feminism as Political Critique. Penn State University Press.

von Hirsch, Andrew, (1992). “Proportionality in the Philosophy of Punishment.” Crime and Justice, 16.

Young, Iris Marion, (2011). Responsibility for justice. Oxford University



[1] Candice Delmas argues that “one has a prima facie duty to report criminal offenses that are harmful or intrinsically wrong” (2014, p.57), which can be defeated when the conduct reported is not morally wrong. Elizabeth Harman has argued that victims of sexual misconduct have an obligation to report the misconduct in order to protect other victims (which may include reporting in a civil rather than criminal context). Feinberg holds that there is a “requirement to report” that is “a civic duty presupposed by our legal system and implicitly recognized by it in many ways” (1990, p.245, cited in Delmas).

[2] See Sadurski (1985), Bottoms (1995), Belloni and Hodgson (2000), Corston (2007), Alexander and West (2010), Sakala (2014), Chiao (2015), Sawyer and Wagner (2019).

[3] See e.g. von Hirsch (1992), Duff (2001), Corston (2007). For dissent see Kolber (2013).

[4] See above.

[5] I suppose for the sake of simplicity that the case, if reported, will lead to prosecution and a conviction. While in general this does not hold as a matter of fact, dropping this simplification would not change the analysis. The moral permissibility of an action (i.e. reporting a crime) should not depend on the chance that the action fail to achieve its aim. The uncertainty attendant on what sentence a judge might impose (given that in practice, there is often latitude in determining sentence severity) generates additional complexity that would need to be addressed in a more complete treatment of this topic.

[6] There are ways of modifying my view to account for this point, if one rejects the argument about causal explanation that I provide in order to suggest that P’s status as perpetrator is not clearly relevant to whether P or V should sustain the extra m units of suffering. One might say that there is a “perpetrator premium” or some extra amount of suffering beyond a just sentence that P is liable to sustain in light of being the perpetrator, in recognition of the fact that P but not V acted culpably (call this premium q). In that case, one can set up a new version of the dilemma I present here by asking what the victim should do if the punishment P stands to receive is n+m+q and n+m+q=o, to which I would provide the same solution that I present for cases in which q is not taken into account and n+m=o.

At the Author’s request, comments have been disabled on this post.

Share on