I’ve recently been reading Michael Zimmerman’s book The Immorality of Punishment. As one might guess from the title, the book presents a sustained critique of the moral justifiability of punishment. It does in an engaging and minimalistic style, one that is replete with formal arguments but devoid of all footnotes. I would recommend it highly, particularly if read alongside David Boonin’s denser and more scholarly (in the sense that it engages more with the available literature) work The Problem of Punishment.
Anyway, in an effort to better understand his critique, I thought I might provide an overview of the arguments Zimmerman marshals against consequentialist theories of punishment on the blog. A consequentialist theory of punishment is one that justifies punitive acts and practices (more on the distinction between these in a moment) by reference to their consequences. Usually, this amounts to the claim that punitive practices can reduce/deter harmful conduct and that since this is a good thing, all things considered, it is morally justified.
Unsurprisingly, Zimmerman begs to differ. I’ll try to explain why over the next few posts. This particular post is purely introductory in nature. It outlines what Zimmerman means when he refers to punishment, what he means when he refers to moral justification or the lack thereof, and sketches the overall goal of his critical project. Subsequent posts will look at particular consequentialist justifications of punishment and explain why Zimmerman thinks they fail.
1. What it Punishment?
Interestingly, nearly ever book or article I’ve ever read on the topic of punishment starts by carefully delineating the set of practices picked out by the term “punishment”. This is understandable: if we are going to challenge or support the practice of punishment it is a good idea to know exactly which practice(s) we are talking about. Nevertheless, I find the careful attention paid to the definition of punishment to be interesting in that it suggests that many authors fear their arguments will be misconstrued because those tasked with its interpretation will have a tendency to misapply the term. This fear may be well-founded. I admit that I myself am not entirely sure whether a person who is obliged to pay out compensatory damages is properly considered a punishee or not.
This is roundabout way of saying that Zimmerman is very particular about the kind of practice he is challenging in his work. Indeed, he dedicates his entire first chapter to the definition of (legal) punishment (something that Boonin also does, incidentally), explaining why this definitional task is necessary and carefully formulating and reformulating a particular definition.
I shall not summarise every stage in Zimmerman’s definitional machinations. Instead, I shall skip straight to the last stage and provide his final, preferred definition of legal punishment, which is (pg 20):
Legal Punishment: Person A legally punishes Person B if and only if A acts on behalf of the state in such a way that:
(1) he harms the punishee;
(2) this harm is intended by the state;
(3) this harm is believed by the state to be fitting - in particular, fitting to the fact, perhaps in conjunction with some other facts, that the punishee is associated with some offence;
(4) he thereby expresses the state’s disapproval both of the offence and of the offender; and
(5) he thereby acts in some legally official capacity.
This definition has many significant features, which Zimmerman discusses at some length. The first three conditions set out what we may call the core requirements of punishment: intended harm that is fitting. The last two, while no doubt important in the legal setting, are peripheral to this core. The definition is also interesting in that, despite the fact that five conditions are set out, it is surprisingly open-ended. This is particularly noticeable in condition (3). Notice how the fittingness relationship alluded to in this condition is not cashed out in terms of the offender’s responsibility for the offence. It is merely cashed out in terms of the offender’s association with the offence. Similarly, notice how condition (5) does not specify a particular group of legal officials with the authority to punish.
One could get bogged down in these technical aspects of the definition all day, but let’s not do that here. Let’s simply accept it and move on to consider what Zimmerman is trying to do in his book.
2. Zimmerman’s Basic Thesis about Punishment
Put somewhat crudely, Zimmerman’s basic thesis — the one that is developed at length throughout his book — is that punishment is very unlikely to ever be morally justified. To make this a little less crude, we need to do two things. First, consider the different forms that moral justification (or the lack thereof) can take. Second, consider the differences between punitive acts, practices and institutions.
On the first of these matters, Zimmerman distinguishes between something that is partly morally justified and overall morally justified. Something is partly morally justified when there is at least one moral reason that counts in its favour or when no moral reason counts against it. Something is overall morally justified when, all things considered, the weight of moral reasons comes down in its favour. Correlatively, something is partly morally unjustified whenever there is at least one moral reason counting against it. And something is overall morally unjustified whenever the weight of moral reasons counts against it. Obviously, to say that something is only partly morally unjustified is weaker than saying that it is overall morally unjustified.
That said, there are grades of moral seriousness that need to be taken into consideration when assessing the moral status of a particular act type or token. As Zimmerman sees it, something can be overall morally unjustified, without it being seriously wrong. And similarly, something can be only partly morally unjustified, and nevertheless seriously wrong. Zimmerman gives the example of a small business owner tasked with distributing bonuses at the end of the year. He asks us to suppose that the overall morally preferable decision is to distribute on the basis of talent, i.e. the most talented workers get the most money. But he then asks us to imagine that the business owner refuses to do this in order to avoid conflict and instead gives everybody an equal bonus. Given the assumptions, what the business owner did was overall morally unjustified, but it was not seriously unjustified. Degrees of seriousness in unjustifiability become particular important when Zimmerman presents his overall conclusion about punishment in the final chapter of his book. I won’t be looking at that chapter here, however.
This brings us to the second matter, namely: the distinction between punitive acts, practices and institutions. To illustrate the distinction, Zimmerman uses the analogy of taxation. There can be particular acts of taxation wherein a person has some percentage of the income they earn or the money they pay collected by the government. There can also be practices of taxation, which in effect amount to a sustained and systematic sequence of acts of taxation. Finally, there can be institutions of taxation — tax departments, officials, advisers, courts and so on — that oversee the implementation of the particular acts and practices.
The same is true of punishment. The intentional and fitting imposition of harm on an individual in response to a particular offence is an act of punishment. A sustained and systematic set of such fitting impositions of harm is a practice of punishment. And a variety of official bodies and persons might be necessary to oversee the implementation of those acts and practices.
Why does any of this matter? Well, it obviously matters when it comes to the interpreting the kind of argument that someone is offering against punishment. The problem is that people tend to conflate acts, practices and institutions in their everyday talk about punishment. So when we are presented with an argument against punishment, we must ask: is this argument about particular acts of punishment, the state-sanctioned practice of punishment, or about the institutions of punishment?
Zimmerman’s work is interesting in that it is an attack on the justifiability of all three. In fact, it is more complex than this because it speaks to both the partial and overall moral justifiability of all three. Consequently, Zimmerman argues for six theses. In the interests of brevity, I will condense these down to just two:
Partial Unjustifiability Thesis: It is highly likely that any act, practice and/or institution of punishment — past, future or present — is very seriously partly morally unjustified. In other words, it is highly likely that there is at least one very strong moral reason (possibly more) that counts against punitive acts, practices and institutions.
Overall Unjustifiability Thesis: It is highly likely that any act, practice and/or institution of punishment — past, future or present — is overall morally unjustified. In other words, it is highly likely that the weight of moral reasons count against punitive acts, practices and institutions.
3. How to Assess Consequentialist Theories
The preliminary work is now done. We know what is meant by “punishment” and we understand the nature and the targets of Zimmerman’s arguments. We can now move on to look at the arguments he offers against consequentialist theories of punishment. As I said at the outset, a consequentialist theory of punishment is any theory that aims to justify an punitive act, practice or institution by referring to its likely consequences. In most cases, this amounts to the claim that punishment is justifiable because it reduces instances of harmful conduct.
As Zimmerman notes, there are thought to be three mechanisms through which punishment can achieve this end.
Rehabilitation: By being intentionally harmed, the offender can learn the moral disvalue of the conduct in which he/she engaged and thereby improve their moral character. This will reduce the likelihood of their engaging in further acts of wrongdoing.
Incapacitation: By being intentionally harmed through involuntary detention or other means, the offender can be prevented from engaging in further acts of wrongdoing.
Deterrence: By intentionally harming people if they engage in wrongdoing, both they and others will be deterred from engaging in further acts of wrongdoing.
One thing should be noted about each of these justifications: although they have been carefully defined here to necessarily involve punishment, this need not be the case. In order to punish a person, one must intentionally harm them. But it seems plausible that one could rehabilitate, incapacitate or deter without necessarily doing this. So the key question for defenders of punitive rehabilitation, incapacitation and deterrence is whether they are more likely to be effective than their non-punitive equivalents.
To put this another way, when it comes to assessing the merits of each of the consequentialist theories of punishment, the key question to be asked is a contrastive one: is this act, practice or institution more justifiable than an alternative act, practice or institution with a similar end in mind?
We’ll see if we can answer that contrastive question with regards to the rehabilitative theory of punishment in part two.