'Comments on Gregg Caruso’s Rejecting Retributivism: Free Will, Punishment, and Criminal Justice'

  • Kelly McCormick orcid logo (Texas Christian University)



Keywords: free will skepticism, retributivism, quarantine model, preemptive incapacitation, respect for victims

How to Cite:

McCormick, K., (2023) “'Comments on Gregg Caruso’s Rejecting Retributivism: Free Will, Punishment, and Criminal Justice'”, Journal of Practical Ethics 10(2). doi:



Published on
31 Mar 2023
Peer Reviewed

In Rejecting Retributivism Gregg Caruso offers ambitious arguments for thinking that our current retributive system of criminal punishment should be abandoned. First, Caruso offers six powerful reasons for rejecting retributivism itself, on the grounds that legal punishment cannot adequately be retributively justified. Second, Caruso proposes, develops, and defends the public health-quarantine model, arguing that it provides us with a more normatively adequate system for dealing with criminal behavior than retributive punishment.

The arguments in this book are rich, and here I will focus on the implications of the public health-quarantine model itself. More specifically, I will raise a dilemma for this model concerning whether (1) it is intended as an ideal model, whereby we ought to radically overhaul our approach to criminal behavior and fully embrace a shift to prevention (rather than sanction) as a foundational aim, or (2) it is intended as a more pragmatic model, whereby we ought to go as far as we can toward revising our approach to criminal behavior while still working within the bounds of our actual criminal justice practices as we find them. If the former, I will argue that concerns about respect for the human dignity of offenders linger despite Caruso’s efforts to defuse them. But, if the latter, then the model recommends outcomes safeguarding the human dignity of offenders that are likely to further undermine the dignity of their victims. In other words, the public health-quarantine model is faced with a troubling tension regarding our respect for offenders and our respect for victims, and it is not obvious how this tension might be resolved.

The Public Health-Quarantine Model

When it comes to arguing against retributivism, Caruso opts for breadth, offering six powerful reasons for thinking that retributive punishment cannot be adequately justified. First is an appeal to the truth of free-will skepticism, which Caruso calls the Skeptical Argument. The justification of retributive punishment depends on the assumption that criminals are (or at least can be) deserving of blame in the basic sense for their criminal behavior. But, if free-will skepticism is true then no one is ever deserving of blame in the basic sense for any of their behavior. So, the truth of free-will skepticism entails that retributive punishment cannot be justified, and thus retributivism should be rejected.

However, even for those unconvinced by arguments for full-blown free-will skepticism, there is a second powerful reason for rejecting retributivism generated by mere uncertainty about whether or not agents are ever deserving of blame in the basic sense. Caruso calls this the Epistemic Argument. The Epistemic Argument highlights the fact that retributivism requires intentionally inflicting harm on wrongdoers, and that intentional harm carries with it a high epistemic standard of proof. Thus, even if the arguments for free-will skepticism fall short, the mere fact that both libertarians and compatibilists face powerful, unresolved objections is sufficient to generate the conclusion that retributive legal punishment falls far short of the high epistemic bar needed to justify its intentionally inflicted (and often severe) harms.

The remaining reasons that Caruso offers for rejecting retributivism move beyond concerns about free-will skepticism and all proceed granting the assumption that ‘the requisite capacity for control is in place and basic desert could be secured’ (16). These further arguments—the Limited Effectiveness Argument, the Misalignment Argument, the Poor Epistemic Position Argument, and the Indeterminacy in Judgment Argument—compose a family of pragmatic concerns about the prospects for retributivism to meet its own aims and sustain a coherent and effective system of legal punishment. When combined with normative and theoretical concerns about the justification of retributive punishment raised by both the Epistemic Argument and the Skeptical Argument, these six arguments from Caruso offer a powerful case for rejecting retributivism.

But then what ought we to do about criminal behavior? Here Caruso supplements a model proposed by Derk Pereboom—the quarantine model—bolstering it empirically with careful attention to recent work on the social determinants of health and criminal behavior. At the heart of this model is an analogy between criminals and the carriers of dangerous diseases. Insofar as we are sometimes justified in incapacitating the latter for reasons of public safety, we can also be justified in incapacitating the former. However, Caruso emphasizes that the kind of justification at issue is not consequentialist, and instead the quarantine model appeals to the right to self-defense. This deontological approach is more resilient against objections than other nonretributive options and requires the quarantine model to take on board further commitment to a principle of least infringement. In accordance with the right to self-defense, such a principle will ensure that criminal sanctions be proportional to the danger posed by an individual. As for self-defense and defense of others, the least restrictive measures should always be taken to protect public health and safety.

Where Caruso’s model builds most significantly on the standard quarantine model is in its call to ‘situate the quarantine analogy within the broader justificatory framework of public health ethics’ (22). Once we have done so, this model moves us away from the usual focus on sanctions and requires instead that we take active preventative steps to minimize ‘outbreaks’ of criminal behavior in the first place. As Caruso points out, quarantine is only needed when ‘the public health system fails in its primary function,’ and such a model would render prevention the primary function of the criminal justice system (22).

In practice, this shift to a more preventative approach to criminal behavior requires closer analysis of the social determinants of criminal behavior. While much of the book is devoted to discussing a wealth of empirical work highlighting the ways in which the social determinants of criminal behavior overlap heavily with the social determinants of health, Caruso emphasizes that foremost among them are

how social inequities and systemic injustices affect health outcomes and criminal behavior, how poverty affects health and incarceration rates, how offenders often have preexisting medical conditions including mental health issues, how homelessness and education affect health and safety outcomes, how environmental health is important to both public health and safety, how involvement in the criminal justice system itself can lead to or worsen health and cognitive problems, and how a public health approach can be successfully applied within the criminal justice system. (22)

The upshot of identifying these overlapping social determinants is that Caruso’s model takes social justice to be a ‘foundational cornerstone to health and safety’ (22). The public health-quarantine model therefore requires active attempts to identify and remedy the serious threat to both public health and safety posed by social inequities such as racism, sexism, and poverty. Caruso ultimately offers a preventative model for dealing with criminal behavior that has the potential to be both more humane and more effective than our current retributive system.

Involuntary Mental Health Therapies and Preemptive Incapacitation

However, as stated at the outset there is a dilemma looming for Caruso’s public health-quarantine model. If successfully executed, the public health-quarantine model has the potential to drastically reduce criminal behavior. By identifying and remedying the social determinants of health and criminal behavior, we would ideally find ourselves in circumstances where the need for actual incapacitation would become rare. On the face of things, this seems like a distinct merit. However, here I will suggest that this outcome comes at a high cost for at least some likely offenders.

To see why, consider two of the eleven specific policy proposals for implementing the public health-quarantine model that Caruso himself suggests:

  1. (10) Researching more effective interventions and rehabilitation strategies for psychopathy.

  2. (11) Making use of big data, neuroscience, and other predictive technologies to aid in identifying, tracking, and predicting violent behavior for the purposes of designing general and local interventions, but prohibit such technology from being used for the purpose of preemptive incapacitation. (Caruso 2021: 266–68; emphasis in original)

First, it is no surprise that Caruso offers a specific policy for dealing with psychopathy. Despite the fact that a relatively small percentage of the overall population suffers from psychopathy, psychopaths compose a disproportionately large percentage of those incarcerated for violent criminal behavior.1 Psychopathy also constitutes a unique determinant of criminal behavior, one unlikely to be mitigated by successful policies intended to remedy various systemic inequalities. As such, any foundationally preventative model for dealing with criminal behavior should offer policies for dealing with psychopathic criminals in particular.

However, it is not clear how such a model can thread the needle between achieving its preventative aims on the one hand and respecting the human dignity of psychopaths on the other. One obvious policy for dealing with psychopathic criminal behavior would be to incapacitate those diagnosed with psychopathy as soon as they have been deemed a significant threat to the safety of others. However, this policy would obviously be at odds with the spirit of Caruso’s public health-quarantine model. Such diagnoses could—and often do—occur at a very young age. A model that recommends incapacitating children or even young adults due to the threat to public safety posed by psychopathy or traits associated with psychopathy would be sharply at odds with Caruso’s repeated assertions that the public health quarantine model does not require incapacitating the innocent (277, 292–96).

Perhaps because he is live to this concern, Caruso does suggest an alternative to the preemptive incapacitation of psychopaths: the use of neurofeedback therapy in correctional settings. While it is beyond my current purposes to dive too deeply into the nature of this therapeutic method, it is a kind of biofeedback and conditioning program, one that has thus far demonstrated ‘reduced recidivism, improved cognitive performance, improved emotional and behavioral reactions, and inhibition of inappropriate responses’ (267).2 While I agree wholeheartedly with Caruso that such findings could have promising implications for the treatment of psychopathy, the concern I wish to raise here is the degree to which those diagnosed with psychopathy would be given a genuine, voluntary choice to undergo such therapy on Caruso’s model. If, for example, the only alternative offered by the state is a lifetime of incapacitation, then the answer seems clearly no. But surely a necessary condition for respecting the human dignity of another is to respect their choices regarding their own mental health and identity. And so, a deeply troubling implication for Caruso’s model taken as an ideal preventative approach to criminal behavior emerges. The public health-quarantine model’s emphasis on prevention rather than sanction leaves it particularly ill-suited to respect the human dignity of those who suffer from one of the most clearly identified mental health determinants of criminal behavior.

Further, this is not the only proposed policy that threatens to undermine the human dignity of potential criminal offenders. In regard to (11)—the use of predictive technology to design general and local interventions—Caruso explicitly tacks on a caveat prohibiting the use of such technology for the purpose of preemptive incapacitation, emphasizing that when it comes to predictive technology, the possibility of false positives looms large, and the burden of proof rests always with those who wish to limit liberty and preemptively incapacitate (268). Further, in regard to concerns about preemptive incapacitation, Caruso later states:

The right of self-defense can only justify limiting one’s liberty when that individual’s actions seriously threaten another’s life, liberty, property, or physical well-being. Since innocent people do not pose such a threat, it would be a violation of the conflict resolution principle to incapacitate them. (277)

Here I admit that I find Caruso’s appeal to innocence in order to block concerns about preemptive incapacitation puzzling, given his commitment to free-will skepticism. As others have persuasively argued (see, for example, Lemos 2018), once we have abandoned the notion of basic desert of blame, everyone is innocent, even those who have engaged in violent criminal behavior. As such, if it is permissible to incapacitate those who have engaged in such behavior, what exactly blocks our justification for preemptively incapacitating others who have not yet caused harm but for whom we have good reason to think such behavior is likely in the future?3 The intuitive prohibition on harming (or, in this case, incapacitating), the innocent seems fundamentally grounded in the fact that the innocent have done nothing to deserve this harm. But as a free-will skeptic, Caruso must exercise caution not to sneak in appeals to basic desert.

Further, it is not at all clear that the alternative deontological principles that Caruso appeals to (for example, the principles of least infringement and normality) will not be defeated by the right to self-defense in at least some cases. Unfortunately for Caruso, his own excellent work identifying the social determinants of health and criminal behavior suggest a particularly troubling path down which this line of inquiry might lead. Without a more robust and principled account of how the public health-quarantine model prohibits preemptive incapacitation, there is a worry this model might actually promote the incapacitation of innocent people who suffer from the very same unjust inequities that render them more likely to engage in criminal behavior. And so, far from offering a more humane alternative to retributivism, the threat of involuntary mental health therapies and preemptive incapacitation suggest that the public health-quarantine model may fare no better on this front, at least when taken as an ideal model that takes the prevention of criminal behavior as its foundational aim.

Rehabilitation, Reintegration, and Respect for Victims

Perhaps one might argue that the above concerns arise only if we take Caruso’s comments about prevention as the ‘foundational cornerstone’ of the public health-quarantine model too literally. Perhaps what Caruso has in mind is not so much an ideal model of how we ought to radically overhaul the criminal justice system but rather a more pragmatic approach. On this reading, rather than aiming to prevent criminal behavior as much as possible, perhaps the idea is simply that we enact common-sense policies for substantially reducing criminal behavior and the need for incapacitation within the confines of many of our current practices as we find them.

The fact that several of the specific policy proposals that Caruso offers emphasize rehabilitation and reintegration of offenders supports this reading. Given that we are unlikely to prevent criminal behavior entirely, or radically overhaul the criminal justice system overnight, in actual practice, the public health-quarantine model requires that we still treat offenders as humanely as possible. This entails that we must also adhere to what Caruso calls a principle of normality whereby when incapacitation is necessary, we must ensure that life inside prison resembles life outside as much as possible, conditions are never stricter than what is required for the security of the community, and no other rights beyond the offenders’ liberty can be infringed upon (279).

Here I will report that I strongly agree with the spirit of Caruso’s view on this front and suspect that we have good reason to actively shift toward more humane practices than our current retributive system regardless of the underlying justification. However, my concern is that some of the policies that Caruso endorses as required by the public health-quarantine model are not always consistent with respect for the human dignity of victims. In fact, for many kinds of violent and sexual crimes, meeting Caruso’s own conditions for respecting the dignity of offenders will be downright inconsistent with also respecting their victims.

If it is not immediately obvious why this is so, consider a possible example, one that is unfortunately all-too common in the actual world. Imagine a case in which the victim, Aaliyah, is a university undergraduate at a small liberal arts college. One night at a party, Aaliyah has something slipped into her drink and she is sexually assaulted by one of the hosts. Let’s consider a version of this kind of case in which Aaliyah reports the crime, agrees to testify in court, several witnesses come forward to support her, her case is rigorously prosecuted by the district attorney, and her assaulter, Brock, is ultimately found guilty. On Caruso’s model, what should the state now do with Brock?

In part, I raise this as a genuine open question for Caruso, as he does not address it explicitly in the book. But here are some possibilities at least suggested by the policies that are explicitly outlined. First, it might be the case that the model recommends no incapacitation for Brock. Perhaps Brock has been sufficiently cowed by the process of his trial and corresponding guilty verdict to have ‘learned his lesson’, and a judge decides that he no longer poses a sexual predation threat to the public. While consistent with Caruso’s model—and perhaps even recommended by it given its commitments to the principle of least infringement and principle of normality—what might the impact of this outcome on Brock’s victim, Aaliyah, be? One need not exercise much imagination to think that it would be devastating. In fact, this is precisely the kind of outcome that Brock Turner famously received for his sexual assault of Chanel Miller, who publicly and eloquently documented precisely this impact in her victim- impact statement and memoir.4 Far from being an outlier, the outcome of this case seems to be precisely the kind of result that the model would recommend given its heavy focus on the humane treatment of offenders. But how can this possibly be squared with respect for the victim’s human dignity?

Further, when we consider cases like this, Caruso’s remarks on reintegration suggest even more troubling threats to victims’ dignity in actual practice. Consider a version of the case in which Brock is sentenced to a period of incapacitation. Let’s say that Caruso’s principles of least infringement and normality are also implemented, and a judge deems the following appropriate: a one-year sentence of incapacitation at the newest local prison modeled after the Norwegian prison island of Bastoy, where Brock will have access to skiing, tennis, a private beach, and a variety of farm animals to tend to, if he so chooses, in an idyllic, pastoral setting. Brock will also have access to intensive therapy and educational opportunities, though the extent to which he partakes in either is entirely up to him. Importantly, once this year of incapacitation is over and Brock is released, he will be reintegrated back into his community, including the educational community of the small liberal arts college that Aaliyah still attends. To further complicate matters, Aaliyah was a freshman at the time of her assault, and she and Brock share the same relatively small philosophy major. As such, it will now be impossible for Aaliyah to avoid Brock both on campus and in classes while also completing her own coursework for her major.

What should Aaliyah do? Again, one need not consider hypothetical scenarios in order to gain a clear picture of the impact this ‘more humane’ approach to dealing with criminal behavior will have on the victim. In actual practice, victims already find themselves in Aaliyah’s position all too often. In many cases, a victim like Aaliyah is likely to change her major, if not leave the school entirely, regardless of her own preferences, and thus significantly restrict her own liberty. Such victims report feeling disvalued and disrespected when perpetrators like Brock are accepted back and successfully reintegrated into the communities that they share with their victims. But reintegration is not only consistent with but recommended by the public health-quarantine model. So, in practice, this model is likely to result in outcomes that can—and in fact often do—directly undermine our respect for the human dignity of victims.


Where does this leave the public health-quarantine model? Here I hope to have elucidated two areas where a plausible public health-quarantine model—one that we might actually enact in practice, either in an ideal or more pragmatic form—must devote further attention. On the one hand, this model is grounded in persuasive arguments for a criminal justice system that is more humane for offenders and more effective at reducing criminal behavior. However, further argument is needed to ensure that it can promote these ends without doing so at the expense of the human dignity of potential offenders and victims. Given the public health-quarantine model’s appeal to deontological justificatory principles, perhaps these concerns can ultimately be addresses by explicit endorsement of further principles capable of safeguarding our respect for these two groups. This move would render the public health- quarantine model far messier than its current form, but such is often the nature of ambitious philosophical positions intended to improve our treatment of one another in actual practice.


  1. See Hare (2003) and Salekin, Rogers, and Sewell (1996). [^]
  2. See Smith and Sams (2005) and Martin and Johnson (2005). [^]
  3. Perhaps, for example, psychopaths. [^]
  4. See Miller (2019). [^]


Caruso, Gregg. 2021. Rejecting Retributivism: Free Will, Punishment, and Criminal Justice (Cambridge: Cambridge University Press)

Hare, Robert. 2003. The Hare Psychopathy Checklist-Revised, 2nd edn (Toronto: Multi-Health Systems)

Lemos, John. 2018. A Pragmatic Approach to Libertarian Free Will (New York: Routledge)

Martin, George, and Cynthia Johnson. 2005. ‘The Boys Totem Town Neurofeedback Project: A Pilot Study of EEG Biofeedback with Incarcerated Juvenile Felons’, Journal of Neurotherapy, 9: 71–86

Miller, Chanel. 2019. Know My Name (New York: Penguin Books)

Salekin, Randall, Richard Rogers, and Kenneth Sewell. 1996. ‘A Review and Meta- Analysis of the Psychopathy Checklist and Psychopathy Checklist-Revisited: Predictive Validity of Dangerousness’, Clinical Psychological Science, 3: 203–15

Smith, Peter, and Marvin Sams. 2005. ‘Neurofeedback with Juvenile Offenders: A Pilot Study in the use of QEEG-based an Analog-based Remedial Neurofeedback Training’, Journal of Neurotherapy, 9: 87–99