Vol. 32 No. 8 (September 2022) pp. 98-101

REJECTING RETRIBUTIVISM: FREE WILL, PUNISHMENT, AND CRIMINAL JUSTICE, by Gregg Caruso. New York: Cambridge University Press, 2021. pp.389. Cloth: $110. ISBN: 978-1108484701 Paper $32.99 ISBN: 978-1108723480.

Reviewed by Whitley Kaufman. Department of Philosophy. University of Massachusetts Lowell. Email:

Gregg Caruso, a philosophy professor at SUNY Corning, calls for a total revamping of the criminal justice system in the United States. Caruso shifts it entirely away from its current basis in retributive justice (the idea that punishment is justified because wrongdoers deserve it) to what he calls a “public health-quarantine model, in which the purpose of criminal justice is not punitive, but aims primarily at crime prevention. The public health model involves a focus on social justice and addressing the social inequities that lead people into criminal behavior, including inequality and discrimination. It would, Caruso argues, lead us to adopt a system that is very different from the highly punitive system we have today. It would entail that sentences be much lighter and that prisoners be treated with dignity and respect, making their prison existence as close to a normal existence as possible, consistent with the goals of preventing crime.

Caruso uses Norway as a model for a better criminal justice system. Norway has explicitly abandoned the punitive or retributive aim in favor of a wholly rehabilitative and preventive system, and one in which sentences are far lighter than in the U.S., and prisoners are treated far better. Caruso points out that far from leading to an explosion in crime, Norway has one of the lowest crime rates in the world and one of the lowest recidivism rates. In contrast, the United States has one of the highest incarceration rates in the world, but also a high crime rate and a high recidivism rate. Caruso predicts that if we were to adopt such a system, it would cost far less and lead to less crime overall. The current system with its focus on making criminals pay, he thinks, ends up distracting us from the far more important goal of preventing crime in the first place.

Criticism of the United States’ system of criminal punishment is, of course, nothing new. There seems to be a widespread consensus among academics (though not among politicians or the general public) that our system is far too harsh and punitive, and at great cost not only to the taxpayer, but also to the inmates and their families with little benefit to show for it. Criticism of the retributive justification of punishment is also widely shared with a number of recent books arguing that there is no coherent moral basis for the idea of retribution, and that it is merely a form of vengeance. Unfortunately, most of these criticisms (and Caruso’s as well) ignore what is the only morally plausible account of retributive punishment, that defended by Nietzsche: retribution is a defense of the honor of the victim (I explain this theory in detail in Honor and Revenge: A Theory of Punishment (Springer, 2012)). But even on Nietzsche’s account, it seems undeniable that criminal punishment in the United States is wildly excessive.

What is distinctive about Caruso’s attack on retributivism is that it is based on metaphysics, specifically the philosophical issue of free will. Caruso belongs to the school of free will skepticism, a small but vocal movement among philosophers that defends the position that humans do not have free will, and therefore cannot be held morally responsible for their actions. They cannot be blamed for their misdeeds, or praised for their good deeds. In addition to the free will skepticism argument, Caruso presents five other types of argument against retributivism: the moral principle that those who intentionally inflict harm on others must provide strong justification for doing so (and retributivists have failed to do so); the failure of retributive punishment to achieve social goals such as crime prevention; the lack of fit between legal criteria of blameworthiness and moral criteria; the fact that the state is not well suited to judge desert (even if such a thing existed); and the fact that judgments of proportionality in punishment are highly subjective and indeterminate and hence unfair to those being punished. However, the free will skepticism argument is the most distinctive, original, and the one that Caruso is apparently most invested in.

Caruso’s attack on the idea of free will involves a version of the classic determinist argument (though Caruso rejects the label ‘determinism’ as misleading). As Caruso explains, “who we are and what we do is ultimately the result of factors beyond our control…and because of this we are never morally responsible for our actions” (p.35); at least the type of moral responsibility that would justify blaming or praising someone for their actions. If so, the retributive justification for punishment cannot be accepted, for it presumes that people are morally responsible for their actions. We do not have the god-like power to create ourselves. What we do is determined by our character and our circumstances, neither of which we have the power to choose. Caruso’s chapter defending free will skepticism, the longest chapter in the book, is a detailed response to the philosophical debate on free will, and is highly technical and will be hard going for those not familiar with the literature. Nonetheless, the overall point is fairly clear: we do not have ultimate control over our actions, and our actions cannot fairly be blamed when we do wrong. Hence a criminal justice system that is grounded in the idea that criminals deserve to suffer because they have done wrong must be rejected. As Caruso summarizes his position, “to hold people truly or ultimately responsible for their actions…would be to hold them responsible for the results of the morally arbitrary, for what is ultimately beyond their control, which is fundamentally unfair and unjust” (p.107). If so, we need to replace our current legal system with a justification that “does not rely on desert” (id.).

What is particularly striking about this argument is that it involves taking the result of a metaphysical debate and using it to recommend concrete – and quite massive—social reforms. I am unaware of any similar case where an abstruse and complex philosophical problem such as free will is made the basis for practical reform in society, let alone the radical changes endorsed by Caruso. It is, however, even more strange that Caruso writes as if the debate is settled, and that the “only reasonable” position is free will skepticism. In fact, this is quite misleading. A recent survey of philosophers, the Chalmers survey, found that only 11% of those surveyed agree with Caruso that we should reject free will. Now, it is of course possible that the other 89% of philosophers are simply being unreasonable. But, it is more charitable to assume that the debate is unsettled and premature to suggest that we are ready to make social recommendations based on this issue. Even worse, one could say that the overwhelming majority of philosophers reject free will skepticism, and thus Caruso’s attack on retributivism must be rejected, at least at present.

Moreover, the many philosophers who reject free will skepticism (myself included) are hardly being unreasonable. Many of them have argued that Caruso’s assumption that free will requires “ultimate” responsibility is misguided; all that it requires is the rational capacity to make moral choices. This is the difference between someone being not guilty by reason of insanity (lacking that capacity) versus someone who is guilty because they have that capacity. Caruso insists that this is not enough, since that does not constitute ultimate control over our behavior. But, why should we assume that free will requires such a thing? In fact, Caruso cites no evidence that the law requires anything like “ultimate” control over one’s behavior, nor does he give a convincing case that moral responsibility requires ultimate control.

Caruso has another strategy in hand, however: even if the free will skeptics have not convinced more than a small minority of philosophers, still we should accept it as the default position on the grounds of the burden of proof. Since punishment involves inflicting massive harm on many people, the burden of proof should be on those who advocate punishment to show that it is morally and metaphysically justified. Now, this argument is not unreasonable; given the nature of punishment, its advocates surely need to do a better job showing how it is justified. And there is a good case that they cannot do so, at least with respect to the extremely harsh punitive regime in the United States.

However, it is more than a little problematic to apply this burden of proof argument to a “metaphysical” question, such as the existence of free will. Metaphysical questions are, by their nature, the sort of question for which there is no prospect of solution. Nor, is there a clear method by which a metaphysical question can be resolved:how exactly could one ‘prove’ that free will is real? And prove it to whom? Caruso’s attempt to shift the burden of proof serves as a way of getting around the fact that his argument has, so far, been rejected in the arena of philosophy. Surely, the reasonable position here is to demand that free will skepticism achieves clear acceptance among philosophers before putting it to work in massive social reform.

In fact, Caruso’s argument faces a further and even more daunting problem. The position he defends is a moral one: retributive punishment is morally unjustified, and we have a moral duty to cease practicing it. But, does this not assume the very notion of moral responsibility, and indeed free will, that he has already rejected? His argument presumes that we have it in our power to make moral choices, and presumably that we are accountable and blameable if we fail to do so. What is this if not moral responsibility? Caruso would answer that this is a weaker form of moral responsibility, not “ultimate” moral responsibility. But, it is far from clear just what it would mean to have a non-ultimate form of responsibility. And, in any case, the retributivist can say that this is all he needs as well: that the criminal has the power to make moral choices, and is accountable for his actions in the same way we are accountable for any injustices in our punitive practices. One suspects that the free will skeptics are trying to have it both ways: to deny moral responsibility for purposes of retributive punishment, but insist on moral responsibility for purposes of eliminating retributive punishment.

In the end, however, even if one rejects his free will skepticism argument, much of Caruso’s critique of our current punitive system can be accepted. He makes a powerful case that our current practices are fundamentally unjust, and that there are alternatives that would be much less harmful to criminals and much more beneficial to society. Even if one does not reject retributivism outright, Caruso is surely right that in America, we have seen retributivism run amok by not only in making sentences far too harsh, but also in distracting attention from the crucial and arguably far more important goals of crime reduction. One might suspect that America’s obsession with blaming and harshly punishing wrongdoers is a way to avoid dealing with the larger problem of social and racial inequities that cause crime in the first place. In calling attention to this problem and pointing out alternatives (the case of Norway is a powerful example), Caruso has done a very important service in support of a social reform movement that is urgently needed.

© Copyright 2022 by author, Whitley Kaufman.