ers.”35 Surely to the extent that we want to rely on intuitions— and I agree
with Rawls in thinking that we rely on these only as part of an overall
network of principles, theories, and judgments— the judgments on which
we rely ought to be “considered judgments,” that is, judgments that are
the fruit of some testing and reflection, not the ones that are maximally
designed to bypass reflection. Moore denies that his theory is emotion-
based, but what he means by that is that it is not an expressive theory that
justifies punishment by its ability to vent emotions. That’s true. But at
the level of justification, the view does appear to be emotion- based, and
based on just those emotions that we are subjecting to critical scrutiny.
Everyday Justice
187
Indeed I think Moore’s theory may also be inconsistent. He announces
that his form of retributivism does not rely on the lex talionis. But then he uses in justification intuitions that probably rely on exactly that. At least he has not attempted to show that there is a class of reliable intuitions
supporting his theory that do not at some level rely on the lex talionis, or what I have called “magical thinking.”
As to Moore’s abstract contention that wrongdoing is sufficient for
punishment,36 don’t we have to ask, “Why?” And doesn’t he owe us
something in return? He tries to avoid the “why” question, appealing
to the intrinsic appeal of his view, but it does appear that we can’t do
away with that question altogether. When a child does wrong, we don’t
just punish away without thinking of consequences. Indeed we typi-
cally think of consequences above all: we want a child who will become
a moral and rational adult, and we choose strategies reasonably aimed
at that end. We seek out expert theories, and try to figure out what treat-
ment best produces good results. If some type of punishment is shown to
be useful, we might choose that. But we do it because it is a good route to
our goal. A parent who sent a child to her room “because she deserves it,”
and who had nothing more to say about whether this was a good way to
produce adult virtue, would not be my favorite type of parent. The revo-
lution in the punishment of children that has taken place since the time
of Dr. Spock is built on clinical and empirical data: harsh treatment does
not work. If parents did not care about what works, but simply thought
wrongdoing sufficient for punishment, they would not care about the
data, and change would not have occurred. Does Moore think that par-
ents are wrong to care about what works? (Of course I think that even if
harsh punishments did work one should not use them, but at least at that
point we would have a real debate.)
Moore will certainly reject this example, on the grounds that children
are not fully rational, and therefore not fully responsible moral agents.
I doubt this, particularly when we think about older children and young
adults who are our children. And I have already argued in chapter 4 that
retribution is a bad way to deal with adults who are our children. But
let us in fairness to Moore think instead about friends one of whom has
wronged the other, but who remain in an ongoing relationship. (And
remember that Moore thinks moral desert sufficient for retribution.)
Surely, as I’ve argued in chapter 4, piling on punishment just on grounds
of desert is not an unproblematic or obvious strategy. It seems to me that
anyone who deals with such a situation merely in terms of backward-
looking blame, without focusing a major part of her attention on the
future, is likely to lead a lonely and unhappy life. In human relations
generally, we are always wrong to neglect the future; it is the only thing
we can change, and it is in effect the only life we have.
188
Anger and Forgiveness
On the whole, I greatly prefer Morris’s overall approach to justifica-
tion. Although I do not think it succeeds in this particular instance, it is
at least built upon principles for which he offers appealing arguments.
Moore’s bare retributivism seems to latch onto some powerful emotions,
but those emotions are precisely the ones whose rationality and useful-
ness we are calling into question.
It is instructive to ponder two more nuanced interventions in the
retributivism debate that one might dub “borderline retributivism”: those
of R. A. Duff and Dan Markel.37 The “core retributivist thought,” accord-
ing to Duff, is that “what gives criminal punishment its meaning and the
core of its normative justification is its relationship, not to any contin-
gent future benefits that it might bring, but to the past crime for which
it is imposed” (3). In terms of my argument, this is a bad starting point,
both because it focuses on punishment to the exclusion of other strate-
gies for addressing crime and because it refuses to focus at all on the
future. Nonetheless, as Duff elaborates the nature of the backward focus
he has in mind, the proposal has attractive features: the core idea is that
of a “calling to account,” and its significance is political, expressing core political values. All this I can agree with, and I have already said that it is important to focus on the past to just this extent: it is right and important that the truth about the past be publicly acknowledged. Capabilities are
intrinsic goods, so a society that aims to protect them for all should take
note of violations: an intrinsic damage to social welfare has occurred. But
for the most part the reasons for such a focus will be forward- looking: we
are trying to build a society in which capabilities are protected, so we
need to note the problems in our way. Moreover, I’ve argued that a decent
society needs to engender bonds of trust and related moral sentiments
among citizens, and citizens will not trust political institutions if they
feel that political principles are just words on paper, proclaiming the
importance of certain capabilities while in practice their violation is taken lightly.
In the article in question, Duff does not say why this “calling to
account” is socially important. But in his earlier book Punishment,
Communication, and Community, Duff says something very like what I
have just said: If law is to mean what it says, “it is committed to censur-
ing those who engage in such conduct.”38 To remain silent would be to
go back on its commitment to its own values. Duff does not explicitly
allude to trust here, but it is surely part of the picture. (But if that is so, then the proposal is inconsistent, to the extent that it officially denies forward orientation but justifies itself in terms of such forward- looking ele-
ments.) Duff also seems right to insist that, while the civil law is focused
on damages, the criminal law is focused on wrongdoing. I agree. And
I also welcome Duff’s insistence that the criminal law should not view
Everyday Justice
189
itself as the agent of the moral law: such moralistic views are not appro-
priate to law in a pluralistic liberal society (14). However, all these prom-
ising moves appear to be undermined when Duff abruptly shifts course,
stating that the “core retributivist thought” is that “the guilty deserve to
suffer (something), and that a proper aim for the criminal law is to subject
them to that suffering” (16). Here he appears to embrace the moralism
he has previously rejected, and he endorses a different view of the core
of retributivism from the one with which he began (and a view that is
much more familiar and less “borderline”). Still, the process of “calling
to account,” as Duff describes it, is actually quite forward- looking and
reform- oriented, no matter what he says: “By imposing a burdensome
punishment, we hope to make it harder for the offender to ignore the
message, to keep his attention focused on what he has done, and to pro-
vide a structure within which he can face up to his wrongdoing and on
what he must do to avoid its repetition” (17). In an important footnote
(45 p.n. 21), he insists again that the purpose of punishment is “to affect
his future conduct.”
On balance, then, Duff’s proposal has promising elements, but in its
attempt to set itself squarely in the retributivist tradition it sells those
elements short.
Markel has basically the view embodied in that last extract from
Duff: the retributivism he favors is a “confrontation” with the offender.
It should focus on communication and not on suffering, its aim should
be to affect future conduct, and its focus should be the wrongful act of a
person, rather than the person him- or herself. Like Duff, he emphasizes
that this conception of punishment is political, and not part of a com-
prehensive ethical or metaphysical doctrine. He therefore explicitly gives
up all views that rest on notions of metaphysical balance or cosmic fit-
tingness. He nonetheless insists that punishment must be “proportional”
to the severity of the offense, apparently because he believes that only
that carefully calibrated severity accurately communicates the perceived
magnitude of the offense.
It seems to me that Markel has more or less abandoned retributivism,
in favor of a forward- looking theory that focuses on the reform of the
offender.39 If that is his goal, he certainly ought to be more interested in
the empirics than he seems to be. Does punishment actually produce
acknowledgment and reform, or does it more often produce hardened
offenders? How might a system that is still a type of punishment really
contrive to communicate a message that would be useful to reform? (This
is Braithwaite’s question, and I think Markel really ought to become a
Braithwaitean.) Are there not lots of other variables that are pertinent to
whether an offender repeats the offense— such as skills training, employ-
ment opportunities, and education? At one point Markel does note that
190
Anger and Forgiveness
subsidized drug and alcohol treatment, and skills training, might reduce
crime (59), but he says that the difference between these programs and
punishment is that they ought to be open to all people, not only those
who have offended (59). It’s not at all clear, though, that punishment of
the traditional sort is the only program particularly suited to criminal
offenders that we could devise. (Perhaps criminal offenders need specific
types of counseling and therapy, as Braithwaite plausibly argues.) Nor
is it clear why the fact that a certain sort of program is good for every-
one is taken to show that it is not the best way of addressing criminal
offenders. So Markel is just guessing when he asserts that the most effec-
tive way of reforming offenders is through punishments calibrated to
exhibit “proportional severity.” It seems implausible to me— although it
would certainly be an improvement if punishments really did this, rather
than exhibiting over- the- top severity for such victimless crimes as drug
offenses.
What is especially interesting about these two proposals, which pre-
sent themselves as nuanced forms of retributivism, is the inexorable pull
of the future. Rather than engage in the type of cosmic balance thinking
I have rejected, these two theorists turn to communication and reform—
all the while maintaining that they are still in the retributivist camp.
Much the same can be said of the family of views dubbed “expressive
retributivism”— of which Duff, indeed, is standardly classified as a lead-
ing exemplar, although, because of the complexity of his view, I have
given it separate treatment. These views combine the claim that certain
crimes are serious wrongs that ought to be publicly acknowledged as
such with the claim, so to speak, that “talk is cheap”— that only hard
treatment properly expresses society’s extremely negative evaluation of
these crimes and, by the same token, the value society attaches to human
life and safety.40 Now I have already agreed with the first claim, and my
own Millian form of consequentialism can admit it, whereas other types
of welfarism might not. But surely the first claim cannot be meant as
purely backward- looking: these are things that society commits itself to
protecting in an ongoing way. To say, “This thing that occurred is very
bad” has a point beyond cheap talk only to the extent that it reinforces
commitment and determination to protect human life and safety going
forward. So, the claim may be partly retrospective, but it is at least partly prospective.
But then the claim that only hard treatment properly expresses the
severity of society’s negative judgment needs prospective empirical eval-
uation. Very likely just letting violent offenders go free is not a good strategy for protecting human life; we can all agree on that. But surely using
the methods that empirical study shows to be best at protecting human
life would be the best way for society to show that it takes the problem of
Everyday Justice
191
crime seriously. It’s just not serious to indulge in fantasies of proportion-
ality: indeed that seems to me a lot of “cheap talk.” Instead, let’s really
study the problem and see what works. And of course, back to our ear-
lier point, this means thinking not just about the deterrent effect of pun-
ishment ex post; it also means spending money on education, nutrition, and other ex ante strategies. The idea that many people are hardwired to offend and are deterred only by fear of something extremely unpleasant
is common, but utterly implausible. Only serious social investment really
shows that a society takes crime seriously— seriously enough not just to
throw people into brutal and degrading conditions, but seriously enough
to spend money where it will really do good going forward. We might
call this the expressive theory of social welfare spending.41
Indeed, one of the best accounts of the expressive theory (which does
not present itself as a form of retributivism), namely Jean Hampton’s
“The Moral Education Theory of Punishment,” insists that the promo-
tion of social welfare is a substantial part of what justifies punishment.
She then adds that an important part of its function is the way in which
it can “tea
ch both wrongdoers and the public at large the moral reasons
for choosing not to perform an offense.”42 She thus combines my emphasis on public acknowledgment with an interest in reforming offenders going
forward. She appears to be open to empirical evidence concerning how
those goals might best be pursued.43
Public expression of political values is particularly important when
those values have not been generally acknowledged. Thus, in cases
ranging from sexual harassment in the workplace to fraud in banking,
the greater the disregard and arrogance the conduct exhibits, the more
important it becomes to make a public statement affirming the fact that
we take this conduct to be very wrongful. In such contexts attaching an
unpleasant penalty to the conduct may well be essential for both specific
and, especially, for general deterrence. The idea of “teaching someone
a lesson” gets real traction in such cases, and is a reasonable prospect.
More important, the penalty educates the society about the conduct,
which they might unreflectively have approved. However, one should
not assume even in such cases that people are eager to commit bad acts
the minute they get a chance, and are deterred only through fear. In the
case of sexual harassment, education and public discussion have done
a great deal to make the harmfulness of this conduct evident. To pursue
penalties without also pursuing such education would be as foolish as to
punish elevator builders who cause harm, while neglecting to regulate
elevators ex ante.
Let us now turn to the other side of the “punishment” debate. Socrates
insisted that the state must never do wrong. Do we have an account of
why punishments, even those that are both humane and forward- looking,
192
Anger and Forgiveness
are not wrongs inflicted on people? It’s clearly not enough to say, “Well,
X asked for it,” since we have rejected the payback idea. And it is also
not enough to say, “Social welfare requires that we use X for our own
ends,” since we have said that the rule of law requires equal respect for
human dignity, and using someone as a mere means runs afoul of that
idea. Reformists will surely press these objections.
Some of the familiar swipes against consequentialist theories of pun-
Anger and Forgiveness Page 34
