Michael Tonry, ed., One-Eyed and Toothless Miscreants (Oxford: Oxford University Press, 2020) ix + 249 pp

When Polonius informs Hamlet he will pay a troupe of travelling performers well, indeed that he will pay them “according to their desert,” Hamlet snaps back: “God’s bodykins, man, [pay them] much better. Use every man after his desert, and who should’scape whipping?”

Hamlet’s reply, no doubt mediated through his famously melancholy worldview, makes a general point about desert: we usually deserve much less than we get. Or as one evangelical Christian put it to me, “We pretty much all deserve Hell anyway. What more we get than that is gravy.”

Several essays in this collection dip into the idea of a God’s eye view from which we could discern what people really deserve; that there is some “recording angel,” who keeps score, and really does know who should be punished and who should be rewarded and how much (see, e.g., Tonry vii). But this metaphor contains the seeds of a more global skepticism about desert. Maybe God knows what we really deserve, but how do we? Are Hamlet (and my evangelical friend) correct that we all really deserve whipping, even Hell?

The overwhelming impression that one gets from this book is that we don’t really know what it would mean to punish people “proportionally.” One looks in vain for anything like a sustained push among the very smart authors collected here to show that we can figure out what proportional punishment actually is (and the focus is mostly on punishment not reward1); only Richard Lippke tries.2 Some contributors basically admit the emptiness of proportionality as a principle, but end up trying to find a limited role for it anyway. This is philosophy as damage control.

Don’t get me wrong. This is an important collection of essays, although perhaps not important in the way it was intended to be. In his introduction, Michael Tonry says that the idea for the book began when he and Douglas Husak “realized that neither of [them] had any clear understanding” of what it meant to say a particular punishment was “proportinate to a particular crime” (vii).3 This collection is the result, and I am not sure that either Tonry and Husak can honestly say that they succeed in getting any clearer on what proportionality in punishment is. Indeed, in Tonry’s essay, he declares his “bottom line” is that “[n]either retributive theories in general nor ‘the principle of proportionality’ can satisfactorily explain how much punishment (‘hard treatment’ in the disciplinary argot) should be imposed for a particular crime or how much and what kinds of suffering any particular offender deserves to endure” (4). This deflationary conclusion of the book is a feature, however, and not a bug. Some books convince you, some exasperate you, and some books make you think; this is a book that makes you think hard about proportionality in punishment.

My review has three parts. In the first part, I try to spell out why proportionality seems so tempting as a principle. We act as if we have a rough sense of when punishments are too much or too little, and this lulls us into believing that deep down, there is something called proportionality in general that we need to live up to when we punish people. We think we know proportionality and disproportionality when we see it, to paraphrase Justice Stewart on pornography (Tonry 10). Part two, relying on the essays in this collection, shows the belief in anything like a strong proportionality principle is almost entirely mistaken. Part three discusses the implications of this conclusion. I suggest that we ditch proportionality, or at least most talk of proportionality. I also think the failure of proportionality tells us something deep about retributivism’s serious problems as a philosophy of punishment.

I.

What makes it seem like we do have a grasp on what proportionality is? It stems, I think, from our strong intuitions about cases that fall on the extremes, or at least on the perceived extremes.4 Take the example made famous by Justice Powell in the United Supreme Court in Rummel v. Estelle. “A statute that levied a mandatory life sentence for overtime parking,” Powell wrote, “might well deter vehicular lawlessness, but it would offend our felt sense of justice.”5 Powell’s point here does not just pump the intuition of disproportionality, but also shows a common feature of many claims about proportionality: they are also subterranean arguments for retributive as against deterrence theories of punishment. We might be able to deter parking violations if we gave life sentences for parking in front of a fire hydrant, but such a punishment is ridiculously and obviously wrong. That is why we should punish not based on expected deterrent value, but on “desert.” Only desert can limit unjust punishments like that of life in prison for a parking violation.

But the intuition can run the other way, too. We could cite any number of rather recent examples, from the outrage at Brock Turner’s sentence, to the lack of any punishment for any number of police officers in the killing and assaulting of Black people. Part of the moral triumph of the #metoo movement must be in the sense of “just deserts,” and of appropriately harsh punishments for moral monsters like Harvey Weinstein. We know a slap on the wrist when we see it, and that too is a matter of proportionality in the response, in the same way that we can see a lack of proportionality in disproportionately harsh punishments.

The middle cases are somewhat of a muddle (see Duus-Otterstrum at 47). We know when things are too much or too little and we call them disproportionate, but it gets harder to say positively when things are just right, that is, when punishments are proportionate. We talk about things like an eye for an eye or a tooth for a tooth, but this had better be a metaphor, or else—as Tonry points out, citing Hegel (9)—we should see justice in a lot of toothless and blind miscreants. Better, perhaps, to see “an eye for an eye and a tooth for a tooth” as Augustine helpfully explained: as a limiting principle on excessive punishment, viz., don’t revenge yourself on someone by taking more than an eye or more than a tooth, if all they’ve taken is one of your eyes and one of your teeth.6 Kant thought that taking a life for a life was an obvious (indeed, required) kind of proportionality in punishment, but this has proven a controversial example.

Nonetheless, we seem to be pointing out something real when we complain that a punishment is too much or too little, and then we almost feel like we can fill in the missing middle cases. If our intuitions are especially strong about the bad punishments, doesn’t that mean–doesn't that have to mean—that we are relying on something to get a sense of what proportional punishments would be? More than one of the chapters cites the familiar-to-professors example of grading fairly (e.g., Matravers 79, 93): don’t we, when we grade exams, have at least a rough sense of proportionality, of giving out better grades to exams that deserve them, and worse grades to exams that deserve them?

II.

The problems with defining what would count as proportionality in punishment almost come spilling out of this book (see, especially Ryberg). Consider the following questions, which can be multiplied (see Matravers 77–79 for a good summary):

In order to say that the worse the crime is, the more severe the punishment should be, we would have to say what makes a crime bad, or at least what makes one crime worse than another. Do we know? If someone robs a person five times, is that worse than if they slap the person one time? What is worse: shooting at someone and missing them, or kidnapping them for several days? For that matter, what is it that makes even one murder worse than another? Is it worse to kill the judge in your pending case, or a witness in that same case? Is it worse to poison someone, or to shoot them?

But we can go further than simply the nature of the crime, and the harm caused, because we also have to think of the nature of the culpability for that crime. Is it worse to negligently create some major harm, or to deliberately cause some minor harm (Ryberg 58)? Or consider the problem of youth and responsibility. It seems plausible to say that the same crime committed by a person who is 16 is less bad than if the person committing it was 25, but how much less?7 (Ryberg 60).

And then there is the problem of what should go into evaluating someone’s blameworthiness for punishment. Does it matter that this is the second or third time that they have committed the crime (that is, is there a so-called “recidivist” premium)? And what about the idea of “whole life culpability,” which Adam Kolber discusses at length in his chapter? Suppose a person has had a rotten upbringing, and has suffered horribly already. Should he be punished less than someone who had a relatively trouble-free childhood and early adulthood?

There are also obvious problems measuring the severity of punishments, as Doug Husak elaborates in his fine contribution. When we say someone who has done a very bad thing deserves a comparatively harsh punishment, how do we rank the harshness of punishment? Which is worse, being in jail for three years, or permanently losing custody of your children? If someone adapts well to prison life, is her thirty year sentence somehow disproportionately lenient when compared to someone who never adjusts to living in a cage, and for whom every day in prison is torture?8

This leads me to a further observation about many discussions of proportionality. We may tell ourselves that we know how one punishment is worse than another, because we take it for granted that the default punishment for offenses is prison, and we know that the relevant unit of greater suffering is more time (see e.g., Lippke, 136). A one year sentence is worse than a two year sentence, all else being equal. But things aren’t equal. We’ve already considered the problem of differences in subjective experience of punishment, but there are other problems too. Not all prisons are the same, not by a long shot: in some prisons you are more likely to be raped, or to be put into solitary confinement (so-called “administrative detention”), and as we recently saw, in some prisons, you were more likely to get sick with COVID-19 and die. All of these distinctions get downplayed when we just fix on the unit of time and forget everything else that makes incarceration awful, as if the only thing inmates lost by being in prisons was time.

To see how hard all these questions are, take a recent example. On February 2, 2021, the Eighth Circuit Court of Appeals affirmed a sentence of thirty months for Deavia David Brown, who pleaded guilty to being a felon in possession of “ammunition.” How much ammunition? One bullet. The sentence seems disproportionate (as the court itself admitted in its opinion). But why is it, and what would be a proportionate sentence? 5 months in jail? 2 days in jail? A week of house arrest? Supervised probation for a year? Or is the sentence actually appropriate given that Brown was a felon, that is, he had a prior conviction? Is it possible that an even greater sentence would be “proportionate” to what he deserves? The problem isn’t that proportionality theory doesn’t give us a precise answer to what sentence (if any) would be appropriate for Brown, it’s that it gives us no guidance at all.9

III.

Kant thought that retributivism was almost synonymous with proportionality (see also Husak 97).10 Without proportionality, retributive theory is a disaster. Without it, retributive theory becomes the idea that only people who do something bad should be punished, or maybe just the idea that it is disproportionate to punish people who have not done something bad. Charitably, then, retribution without proportionality becomes the innocence principle, viz., it is wrong to punish the innocent. This is a true principle, so we shouldn’t take it for granted, and it is not something that we can get from either deterrence or rehabilitative theories of punishment. But I don’t think it amounts to a theory of punishment on its own.

Nor does proportionality seem to make sense as a principle outside of retribution, so it is not as if we just need to find the theory of punishment that does work with proportionality. We can talk about proportionality in a casual way when it comes to deterrence theory—a punishment is disproportionate, for example, when it is harsher than what is necessary in order to get optimum deterrence. But this is just to repeat deterrence theory; proportionality is not doing any real independent work. Retribution, by contrast, is geared toward having proportionality do this kind of real work. You punish someone according to their desert, and no more (or less) than they deserve. This is very different from the idea that we should punish some people and some crimes more severely in order to pursue some worthy social goal, like public safety.

Matravers and Husak do the best, I think, in trying to minimize the importance of proportionality in theories of punishment and making it less “central” (Matravers 93, Husak 110–113). This is the right move to make. If proportionality cannot do the work of telling us what people deserve, it should rightly lose pride of place in punishment theory. But I do not think that this means thinking of proportionality as just one factor among others in considering what punishments to give. If the requirements of proportionality are mostly “elusive” (Husak 121) then making proportionality only 5% of the equation rather than 50% won’t make those requirements any less “elusive.” The same problem would seem to apply to “weaken[ing]” the strength of the proportionality principle (Husak 122). What Matravers and Husak miss is that what’s left of proportionality is mostly talk, and that invocations of disproportionality, pace Tonry, actually function more rhetorically than substantively (21).

In algebra, proportionality is the relationship between two quantities. Invoking the idea of proportionality in punishment suggests that we have the ability to match crimes with sentences, algorithmically. But as these essays show, it’s not even something we can do with difficulty. We can’t generate the kind of precision that proportionality theory aspires to.

The perceived lack of proportionality in extreme cases seems like something real, and a substantive judgment. But this is mostly an illusion. When we say a punishment is “disproportionate,” this amounts only to the sort of judgment we make when we say a punishment is cruel, or that it is simply “excessive” (in a hand-waving sort of way). Talking in terms of cruelty or excessiveness, in fact, seems to me preferable. Judgments of cruelty and excessiveness do not contain the hints, as proportionality does, of a mathematical exactness. A punishment that is cruel or excessive is just one that is too much and too much of the wrong thing. It is not clear we can be more precise than this.

I confess to be at sort of a loss to find the thick ethical concept that adequately corresponds to the idea that some punishments are too lenient in the way that cruelty and excessiveness correspond to disproportionately harsh punishments.11 It is not clear to me that this is a bad thing. And this leads me to my final thought. Retributive theory’s core intuition is that people deserve to suffer for the wrongs they have done. We can put off for another day deciding how much people deserve to suffer. But the failure to define any robust principle of proportionality should lead us to question this retributive intuition. Maybe people just don’t deserve to suffer.12 Or maybe that’s a job best left to God.13