All content in this area was uploaded by Thane S. Pittman on Jul 08, 2014
Content may be subject to copyright.
All content in this area was uploaded by Thane S. Pittman on Dec 18, 2013
Content may be subject to copyright.
The Psychology of Compensatory and Retributive Justice
John M. Darley
Department of Psychology
Thane S. Pittman
Department of Psychology
How do observers respond when the actions of one individual inflict harm on another?
The primary reaction to carelessly inflicted harm is to seek restitution; the offender is
judgedto owe compensation tothe harmed individual. Theprimaryreaction to harm in
flicted intentionally is moral outrage producing a desire for retribution; the harm-doer
must be punished. Reckless conduct, an intermediate case, provokes reactions that in
volve elements of both careless and intentional harm. The moral outrage felt by those
who witness transgressions is a product of both cognitive interpretations of the event
and emotional reactions to it. Theory about the exact nature of the emotional reactions
is considered, along with suggestions for directions for future research.
Where, in the varieties of justice, does the justice that
calls for compensation and punishment arise? In distrib-
utive justice, a group has received a benefit and the task
is to determine a just way to distributesharesof the bene-
fit among the group members. Distributive justice has
received a great deal of research attention, butpsycholo-
gists have done less with the set of fundamentally impor-
tant questions about what cognitions and emotions are
brought about when one person does harm to another,
and what actions toward the victim and against the
harm-doer are necessary to remove or otherwise deal
with this injustice. The problem here, as Shaver (1985)
aptly named it, is “the attribution of blame” (p. vii).
In this review we make a set of critical theoretical
distinctions. We suggest connections between the con
structs we propose and when possible cite evidence
that confirms the postulated theoretical linkages. We
then turn to the psychological purposes people have in
mind when they punish offenders and make the case
that people are motivated to punish by a just deserts
perspective, and that this is so because of the moral
outrage that is generated when a person intentionally
wrongs an innocent other. Finally, we argue that moral
outrage has an emotional as well as a reasoning com
ponent, draw on some recent literature to suggest what
the moral emotions might be, and cite some Functional
Magnetic Resonance Imaging (fMRI) research which
supports the notion that emotions are engaged when
moral reasoning is carried out.
Let us mention a few unusual aspects of our plan,
and some of its limitations. We primarily draw on
psychological theory and research in this review, but
occasionally enlist help from an unusual source, the
civil and criminal justice systems of our society. In
any society, the jurisprudential system must face
questions of blame, accountability, and punishment,
and the principles developed to do so are a possible
source of psychological insights (Darley, 2001). This
is so because for some hundreds of years the legal
system of our culture has faced the necessity to hear
real cases of harm and to decide what punishments
the perpetrators may deserve and what compensations
are required of them for the damages they caused.
One of the major influences on these decisions was
the morality of the community members as they func
tioned as actors in the court system. So, the rules that
courts developed are at least a starting point for clues
to the decision rules that ordinary people use to deter
mine their judgments about compensatory and puni
The jurisprudential system is not, however, an
ending point for our task, which is to articulate the
moral intuitions of the ordinary run of people; for a
number of reasons, laws and courts make decisions
that are not in accord with those that citizens would
make. Therefore, we cannot assume that the legal
codes are always consistent with community intu
itions about justice. Thus, when possible, we cite psy
chological research which demonstrates that the gen
eral principles we present do describe how people
think about the morality of punishment.
Personality and Social Psychology Review
2003, Vol. 7, No. 4, 324–336
Copyright © 2003 by
Lawrence Erlbaum Associates, Inc.
Requests for reprints should be sent to John M. Darley, Depart
ment of Psychology, Green Hall, Princeton University, Princeton,
NJ 08544–1010, email@example.com or to Thane S. Pittman, De
partment of Psychology, Gettysburg College, Gettysburg, PA 17325,
American social psychologists have learned to be
cautious about claiming that their discoveries transcend
their culture, and we need to be cautious about that also.
In this review, we claim only to be articulating how
harm-doing events are thought about in our culture.
Once the basic principles are articulated, it will be possi
ble to test for their trans-cultural generality. Eventually,
it should also be possible to use the legal codes of other
culturesas a windowintotheiruniquemoral intuitions.
Although we are cautious about generalizing across
cultures, we are bold elsewhere. We do think the broad
moral principles that we articulate are characteristic of
citizens in our culture; moral reasoning is a product of
socialization and the culture needs to have broad agree
ment on morality to bring about the internalization of
these understandings that normatively shape the con
duct of its citizens.
The Nature of Compensatory
and Retributive Reactions
Our firsttaskis to suggest some concepts that are nec
essary to model this area. There are several distinctive
interpretations of interpersonal behavior which trigger
the reaction that “justice needs to be done,” and there are
two distinctive blame-resolving actions that psycholog-
ically “do justice.” In this section, we first address the
two different forms of justice-resolving actions.
Justice-providing reactions fall into two general
classes, one directed toward the victim, the other to-
ward the perpetrator of the harm. The first class of jus-
tice-providing transactions is focused on compensa-
tion for the victim. Here the desire is to come as close
as is possible to restoring victims to the state in which
they were before the harm was inflicted. The second
class of justice-providing transactions is more con
cerned with punishment, with inflicting the appropri
ate punishment on the perpetrators for the wrongness
of their actions in inflicting the original harm.
The Impulse to Compensate
Consider a substantial harm that is inflicted uninten
tionally by the actions of one person on another. A’s car
rolls down the hill and crashes into B’s garage or C’s
prize horse. Certainly, we will want to know more about
the circumstances that led the car to break away from
whatever restraints the owner placed on it. Was the
owner’s actions in securing the car so inadequate as to
constitute reckless conduct, or was he or she negligent?
Or did the owner takemanyprecautionsthat nonetheless
failed due to some horrible set of accidental coinci
dences? In the latter case, observers are unlikely to see
this as a case requiring anything of the owner, but attri
butions of negligent or reckless conduct in the former
case will invoke a compensatory justice reaction
(Pittman & Darley, 2003). Because someone’s breach of
required conduct caused damages to another, people be
lieve that the person causing the damages must redress
this breach. As the landmark hornbook
on torts com
ments, “a tort is not crime,” (Keeton, Dobbs, Keeton, &
Owen, 1984, p. 2) but it does involve the creation of dam
ages to others by conduct that is socially unreasonable:
The civil action for a tort … is commenced and main
tained by the injured person, and its primary purpose is to
compensate for the damage suffered, at the expense of the
wrongdoers. If successful, the plaintiff (the damaged per
son) receives a judgment for a sum of money, enforceable
against the defendant. (Keeton et al., 1984, p. 7)
The goal, it seems, is to restore the victim’s life as
closely as possible to its preharm level.So, the task is for
A to repair the damage to B’s garage, if A has the skills to
do so, and the repair job needs to be of a quality equal to
the quality of B’s original garage. If by some wonderful
coincidence, A owns a prize horse that is the equivalent
of C’s prize horse, an intuitively perfect act of compen
sation would be for A to give C his horse to replace the
horse slaughtered by A’s runaway car. Now a further is-
sue arises, which is that C may have loved the horse that
was killed. A justice-doer may feel that C is owed some
compensation for the loss of the loved object, but it is
puzzling as to what form that compensation should take.
Here we have discovered something that is often re-
ferred to as pain and suffering, and sometimes persons
deciding on what compensation is due are asked to di-
vide their awards into amounts awarded for damages
and for pain and suffering.
In some cultures that do not have a pronounced spe-
cialization of labor, it may be possible that the compen
sation A gives to the victim is the restoration, with A’s
own labor, of the victim’s dwelling, or personal belong
ings; or by a fairly direct extension, A may replace C’s
destroyed crops or livestock with some that A had culti
vated or bred. In cultures in which labor is specialized,
and monetary economies have developed, the compen
sation may be more indirect, involving an assessment of
the monetary value of the damages, and a transfer of an
equivalent monetary sum from the accident-causer to
the injured party. As members of a monetized culture be
come familiar with the principles of valuation that cur
rencyallows,it becomes increasingly sensible for mem
bers of that culture to converge on what sorts of
monetary compensations are just. The notion of replace
ment costs is the one generally used.
The translation of damages into monetary amounts
becomes particularly psychologically puzzling when
COMPENSATORY AND RETRIBUTIVE JUSTICE
A hornbook is a text written by legal scholars that attempts to
summarize thebroad underlying principles that determine the disposi
tion of the cases that various courts have reached, in any number of
specific areas of law. Thus, there will be a hornbook on torts, in which
negligencerulesare summarized, one on the criminal law, and so on.
the goal is to assign a dollar value to, for instance, the
pain inflicted on the victim by the accident the perpetra
tor caused, or the related concept of the suffering caused
during the time it takes the victim to heal physically or
psychologically from the accident. Often it is difficult to
translate harm into a specific monetary equivalent. And
there are a number of other cases in which the concep
tual and psychological inadequacies of the monetized
compensation system are quite striking, for instance,
when a dollar value is assigned to the lost life of a young
husband and father to compensate the family for all they
will suffer from his death. We can expect some cases to
generate a good deal of argument about what amount
and kind of compensation is adequate. Nonetheless, the
concept of compensation, at least in its most
prototypical form of restoring the concrete condition of
the victim to what it was prior to the accident, is clear.
“The function of tort law is to compensate someone who
is injured for the harm he has suffered”(LaFave, 2000, p.
13). We next turn to the second prong of the justice-re
The Impulse to Punish
There are many cases in which the dominant im-
pulse is not simply to compensate. For example,
crimes against a person or a person’s property, such
as when a thief steals the property of another, or a
mugger violently robs another person, or a book-
keeper embezzles money from an employer, lead to a
visceral reaction that the criminal must be punished.
Intuitively, the following example will clarify the dif-
ference between punishment and compensation. If a
thief, eventually caught, were to compensate the vic-
tim immediately for the full monetary costs of the
theft, people would not consider that this was a suffi
cient infliction on the perpetrator to close the matter.
Instead people feel that suitable punishment must be
inflicted on the perpetrator by the judicial system, a
punishment in proportion to the moral gravity of the
offense committed. This is not accomplished simply
by requiring the perpetrator to compensate the victim
for damages. As the leading text on criminal law
noted (LaFave, 2000)
it has been argued with some force that the only real
basis for distinction between crimes and civil wrongs
lies in the moral condemnation which the community
visits upon the criminal but not (at least not so power
fully) upon his civil wrongdoer counterpart. (p. 12)
The expression of this moral condemnation in our
culture is done via the assignment of punishment.
When inflicted wrongs call for punishment, the
naïve perceiver also thinks the victim should be com
pensated by the perpetrator for the damages suffered.
(See LaFave, 2000, p. 13, Footnote 5, for a discussion
of the complicated ways this can be achieved in crimi
nal courts.) Compensation, however, is a desired but
not a sufficient condition for justice.
The Concept of State of Mind
What distinguishes between those violations of jus
tice that provoke a desire for compensation and those
that evoke a desire to punish? Research suggests that the
primary determinant lies in judgments about the perpe
trator’s state of mind. If perpetrators are judged to have
committed the harm unintentionally, naïve psychologi
cal thinking focuses on compensating the victim for the
damage done to his or her property or person, presuming
that restoration is judged to be appropriate. However,
when harm is thought to have been committed intention
ally, people see punishment as necessary.
Foreseeable and Unforeseeable
Legal distinctions provide some guidance to how
people think about foreseeable and unforeseeable harm.
Adapting Piaget’s (1932/1965b) famous case, consider
the person who broke 15 glasses, but did so under two
different circumstances. In both cases, the glasses were
on a tray on a table temporarily placed within reach of a
swinging door. In the first case, the person pushes open
the swinging door as she has done many times in the
past, but this time knocks over the tray, and all of the
glasses fall to the floor and shatter. Should the person
“have known” the tray was there? If the circumstances
were such that she could have had no inkling of this, then
she is not to blame. Therefore, she has not acted negli
gently and is not obligated to compensate the owner of
the glasses for their destruction. (Whomeverwas foolish
enough to leave them in the precarious position behind
the door might be, but not the person who came through
the door in the normal way.) On the other hand, if the per
son had been told that today, perhaps in preparation for a
party, in the room behind the door breakable things were
likely to be within reach of the door’s swing, the person
becomes responsible for the damages and must com
pensate the person who owns the glasses for their re
In other words, the person in the second sce
nario would be perceived as negligent (in ordinary
DARLEY AND PITTMAN
People do tend to give some punishment mitigation if they read
the paying back of the theft as an expression of remorse or renuncia
tion on the part of the thief. For an examination of the complexities of
this issue, see Pittman, Darley, and Robinson (2003).
It is worth noting that certain other compensatory moves are re
quired; for instance, the person better find a broom and dustpan and
sweep up the broken glass, find other glasses that can be used, and so
on. These are all encompassed by the requirement of putting the situ
ation back, as closely as possible, to the way it was before the dam
age was inflicted.
speech, careless). The standard legal test for negligence
(LaFave, 2000, pp. 248–249) requires that an ordinarily
prudent or reasonable person would have been aware of
the risk and would not have run it.
There is evidence that perceptions of negligence
affect moral judgments about offenders. Enzle and
Hawkins (1992) demonstrated that if “a prior negli
gence could either be inferred implicitly from con
temporaneous information … or was explicitly pro
vided,” then the actor was judged to have more moral
responsibility for the outcome (p. 184). Robinson and
Darley (1995, pp. 83–96, Study 8) presented descrip
tions of six different instances of harmful actions that
could be committed either without fault or negli
gently, and found increases in punishment assigned as
the offense moved from faultless to negligent com
mission (see also Pittman & Darley, 2003).
In contrast to harm that is the result of negligence, in
tentional harms trigger the motive to assign a punish
ment and secondarily require compensation. As LaFave
(2000, p. 229) commented, “the meaning of the word
‘intent’ in the criminal law has always been rather ob-
scure … .” But it is not obscure in naïve psychology if a
person believesthat certain actions on his or her part will
bring about an outcome, and freely takes those actions,
then he or she can be said to intend to bring about the out-
come. If the person also knew that the outcome was mor-
ally wrong, then he or she intended to bring about an un-
justifiable wrong outcome and deserves punishment. A
number of studies already reviewed contained condi-
tions that described intentional wrongdoing, and dem-
onstrated that more severe punishments were assigned
to the perpetrators of those actions.
As many have noticed, a harm done intentionally
does not always demand the punishment that is ordi
narily given for that harm (Alicke, 2000; Schlenker,
Britt, Pennington, & Murphy, 1994; Shaver, 1985;
Snyder, Higgins, & Stukey, 1983; Weiner, 1995).
There is a vast literature on the effects of justifications,
excuses, and mitigations for harm-doing on the punish
ments assigned for that harm. A justification serves to
render a harmful act as not only not wrong but as mor
ally right, as for instance is the case when a school
teacher shoots dead the madman holding her and her
children hostage and threatening to kill them. The bank
manager who opens the bank vault to robbers because
they threaten to kill the bank tellers unless he does so is
excused for his normally wrong action and receives no
punishment. Mitigations for the wrong done, such as
when the wrong was done under provocation, reduce
the punishments assigned. In developmental studies
(Darley, Klosson, & Zanna, l978), even very young
children recognized standard claims of mitigations and
justifications, and accordingly reduced the punishment
assignments they gave.
Unintentional and Intentional Harms,
Compensation, and Punishment
Reactions: A Summary
It will be useful if we organize what we said ear
lier into a set of more concise statements tracing the
thought processes of a person who perceives a harm-
The first question is whether the harm is perceived as
intentional or unintentional. If the harm is perceived as
unintentionally caused, then the next question the
perceiver must solve is whether the harm-doer has some
responsibility to compensate the victim of the harm. If a
harm-doer could not foresee that his actions would have
caused harm, and if he exercised the normal standards of
care expected of people in our society, then he or she will
be judged not responsible for the harm and will not owe
compensation for it. However, if the harm-doer could
have foreseen the harm of his or her actions, or alter-
nately, if his or her actions fell below the standards of
care we expect people to take to avoid harming others,
then he or she owes compensation to the victim. The cost
of the physical damages primarily determines the
amount of compensation required, but compensation
for pain, suffering, and psychic disruption suffered by
the victim sometimes also influences how much com
pensation is perceived as necessary.
On the other hand, when harm is perceived as inten
tionally caused, the observer first considers what pun
ishment is due to the harm-doer. We suggest that the first
reaction the observer has is to punish as if the harm was
inflicted in a completely unwarranted way. Next,the ob
server must determine whether there was any justifica
tion or excuse for the harm being done (e.g., consider
any possible mitigating causes). If any of these condi
tions exist, then the punishment is eliminated or ad
justed downward. (The distinctions among justifica
tions, excuses, and mitigations, and their psychological
mechanisms, are discussed in more detail in Darley et
al., 1978.) In comparison to unintentional harm, we may
note an asymmetry here: the observer to intentional
harms will, in addition to determining punishment, also
assess compensation against the harm-doer using the
same considerations as are used to determine compen
sation for unintended harms. However, in general, the
observer to negligent harm will assign compensation
COMPENSATORY AND RETRIBUTIVE JUSTICE
We should acknowledge a complication here. In our current
thinking, punishments are not generally assigned to negligent causa
tion. That assumes that the respondents could assign compensation,
but in the Robinson and Darley (1995) study, a compensation option
was not provided. Thus, we suggest that the assignment of mild pun
ishment was the only way that these respondents could signal the de
sire for the assignment of a duty to compensate. In the Pittman,
Darley and Robinson (2002) study in which it was possible to assign
compensation as well as punishment, compensation was assigned
and punishment was generally not assigned, as expected.
but not punishment to the negligent action, although we
next discuss a specific complex case in which an unin
tentional action is sometimes punished.
The Hybrid State
of Reckless Causation
The previous section summarizes how perceivers
analyze prototypical cases of harms done intention
ally and unintentionally. It is tempting to quit the
analysis with those clear prototypes, but there is an
important intermediate case to analyze. That is, there
is one other state of mind customarily distinguished
in which a person can cause a bad outcome in a way
that psychologically seems intermediate between in
tentional and negligent causation: A person can act
and perhaps cause an accident because his or her ac
tion was reckless, which is worse than simply being
Legal definitions about recklessness differ in differ
ent jurisdictions, and psychological reasoning about it
also seems confused. Nonetheless, the difference be
tween negligent and reckless conduct does matter to
observers. In the Robinson and Darley (1995) study
mentioned earlier, actions committed recklessly re-
ceived considerably more punishment that those com-
mitted negligently but less than punishments given to
the same actions committed intentionally. In fact, the
data patterns suggested that the biggest increment in
punishment occurred between negligent and reckless
commission. Clearly, naïve observers sometimes react
much more extremely to reckless commissions than to
The differentiation between negligent and reckless
causation has been suggested to rest on two issues. If
an actor negligently causes an action, the actor did not
consider the risks of the harmful outcome but should
have done so. The reckless person, by contrast, is either
(a) conscious of the risks he is running but chooses to
run them, or (b) runs the risk of causing greater harms
than are risked in cases of ordinary negligence, or both.
At the moment, we know of no research that disam
biguates how ordinary people think about these two
factors and combines their effects. The psychological
question is what circumstances provoke in ordinary
people the desire to punish reckless individuals. This is
a topic that calls for further research.
The civil legal system’s treatment of recklessness is
revealing, involving the concept of punitive damages,
and foreshadows our discussion of the reasons under
lying the impulse to punish. Punitive damages can be
assessed from the perpetrator of a harm expressly to
punish him for his poor conduct. As Keeton et al. com
mented, the concept of gross negligence
applies to conduct which is still, at essence, negligent
rather than actually intended to do harm, but which is so
far from a proper state of mind that it is treated in many
respects as if it were so intended. Thus it is held to jus
tify an award of punitive damages.” (italics added,
1984, p. 213)
As Keeton et al. (1984, p. 9) also remarked, the
harm in instances in which punitive damages are at is
sue “has the character of outrage frequently associated
with crime.” The notion of outrage, with its connota
tions of a moral and emotional reaction, is one to which
we will return.
The punitive impulse is a seemingly alien intru
sion into the civil justice system that occurs just when
we would expect moral outrage reactions to be
aroused. Within the civil justice system, these puni
tive damages are conceived of as serving deterrent
purposes (Keeton et al., 1984, p. 9). However, this is
not why ordinary people assign punitive damages. As
the word “outrage” suggests, people punish for rea
sons of just deserts.
Of course, punitive damages are extracted as a
monetary fine, and not as a prison sentence. However,
we suggest that ordinary people think some reckless
actions morally require the imposition of society’s
major sanction of a prison term. We further suggest
that this will occur when harm is the result of reck-
less rather than simply negligent acts, and the dangers
courted were severe.
In support of this hypothesis, one study (Karlovac &
Darley, 1988, Experiment 4), found that about 50% of
perceivers assigned some prison time as well as com-
pensation to the victim when the harm-doer parked a
truck on a hill above a children’s playground, and took
what perceivers saw as too few precautions against the
truck rolling downhill. (The truck was described as get
ting loose, rolling down the hill, but harming only a
piece of property.) When only property damage was
risked by parking the truck, the few precautions condi
tion triggered only an assignment of compensation and a
fine. In another scenario (Pittman & Darley, 2003), in
which a person droverecklessly through aschool zone at
a time of day when children were traveling to school, as
the consequences of this reckless act became more se
vere, so did the desire to impose a jail sentence in addi
tion to punitive damages.
The Spectrum of Events That Trigger
the Compensatory or Punitive Blaming
So far, we have written as if it were only serious
harms that cause observers to engage in blaming
analyses. But perhaps the infliction of more ordinary
interpersonal harms triggers at least some of the same
negative reactions. If we pause to think, we realize
that we do not consciously engage in the elaborate
blaming analyses we have outlined here in all of the
DARLEY AND PITTMAN
instances in which a person commits some untoward
act; instead, many reactions are the result of auto
matic thinking that only sometimes leads to checks
on the validity of these more automatic conclusions.
Generally, we engage in these analyses only when
some reasonably grave damage has been inflicted.
Less grave offenses, like harsh words delivered inten
tionally or failing to pick up your spouse on a cold
winter day, are left to more informal, conceptual pro
cesses, perhaps followed by interpersonal processes
in which blame, compensation, and punishment are
negotiated. Two interesting psychological questions
arise here. First is the question of what other ordinary
events between people also trigger questions of com
pensation and blame. In an article that is remarkable
in its psychological interest (and in the degree to
which it has not been noticed within the psychologi
cal justice community), Lempert (1972) suggested
that the judicial analysis of contract violations is a
useful model for ordinary persons thinking about in
formal commitment violations and the sense of out
rage and betrayal they provoke. The second question
is about procedural justice. To what degree do the in
formal sanctioning inquiries between ordinary dispu-
tants take on the form of the more elaborate inquiries
characteristic of the criminal justice system that lead
to justice system-sanctioned punishments and justice
system-enforced compensatory actions? This is a
question about how the forms and processes of for-
mal inquiry shape more informal processes. Or alter-
natively, it is an inquiry about to what degree our
naïve sense of fair procedures have shaped the prac-
tices of the criminal justice system.
The Psychological Nature
of the Reaction to Intentional Injustice
It is instructive to ask why people seek to punish
wrongdoers. In our research and in writing this review,
we may have stumbled toward the answer. The reaction
is one of moral outrage leading to a desire to inflict a
just deserts punishment on the offender. To explicate
our point of view, we first review the evidence suggest
ing that people assign punishments for intentional
transgressions to inflict a just deserts punishment on
the offender. Next, we conceptualize the notion of
moral outrage that we think motivates the desire for a
just deserts punishment.
The Naïve Psychological
Motive for Punishment
The literature on the moral philosophical under
pinnings of criminal justice offers several reasons
why people might seek to punish wrongdoers. First,
there is a set of utilitarian or consequentialist reasons
for punishment; for example, that punishment is
given to modify future behavior, specifically, to make
the commissions of crimes less likely. The most
prominent of these, called to our attention by
Bentham (1970), is the deterrence of future crimes,
both by the punished criminal (specific deterrence)
who learns that the pleasure gained from the crime is
insufficient to outweigh the pain inflicted by the pun
ishment; and by the general populace who is made
aware of the punishments they risk by observing the
punishment inflicted on the person convicted of the
crime. Lately, the incapacitative perspective has cap
tured our national attention. This pessimistic stance is
that we prevent future crime by sending people to
prison who are likely to commit future crimes—that
is, those who have committed one or more crimes in
The rehabilitationist justification is also a utilitarian
perspective, but one that is most socially optimistic and
tenuously connected to the ordinary concept of punish
ment. It attempts to subject criminals to a set of training
and attitudinal adjustment procedures that will reedu
cate them so that they have the resources and motiva
tions to become productive members of society. This
was a stance that actually was used to determine sen-
tences in the United States some years ago, and various
educational and vocational training programs were put
in place in prisons to implement it. The programs were
underfunded, and the philosophy of punishment shifted
to more harsh treatments, so rehabilitation has largely
disappeared from the set of possibilities that can now be
The one non-utilitarian motive for punishment is
generated by the desire to give culprits their just deserts
for committing crimes. The purpose of punishment is
not to alter behavior in the future, but to punish perpe
trators for their past behavior. Assigning a just deserts
punishment to a crime may produce a sentence that de
ters many individuals from committing the crime, but
this effect is merely a happy byproduct of a just deserts
orientation rather than a purpose of it.
This perspective, often called a retributionist
stance, is often thought of as a vindictive and retro
grade motive for punishment, stemming from the old
and barbaric lex talionus rule of “an eye for an eye.”
However, there are also a number of philosophical
justifications of it, many of which are covered in Pe
ter French’s recent book, The Virtues of Vengeance
(2001). Nozick (1981, pp. 366–370) is particularly ef
fective in differentiating between the possibly mor
ally justifiable infliction of retribution from the
less-justifiable infliction of revenge. Elster (1990) ex
amined the set of norms that will govern retaliatory
justice in any society.
The psychological underpin
COMPENSATORY AND RETRIBUTIVE JUSTICE
Actually, Elster (1990) examined the norms governing acts of
revenge, but his conditions apply to retribution.
nings of this stance have been articulated by Darley
(2002), Hogan and Emler (1981), Vidmar (2001,
2002), and Vidmar and Miller (1980).
If we broaden the scope of our analysis, we realize
that the intentional infliction of a wrong injures not
just the victim but what might be called the “fabric of
society.” As Tyler and Boeckmann (1997) pointed
out, “when considering the problem of criminal devi
ance from a societal perspective, people may wish to
punish severely in order to symbolically reassert the
status of the violated rules.” One sees the force of this
argument. Many have emphasized the importance of
a stable worldview in rendering one’s environment
predictable and potentially controllable (e.g., Jones &
Davis, 1965; Kelley, 1967). Threats to such world
views are not accepted happily, and indeed the re
search on control motivation (e.g., Pittman, 1993;
Pittman & Pittman, 1980), learned helplessness (e.g.,
Seligman, 1975), cognitive dissonance (e.g.,
Festinger, 1957), terror management (e.g., Soloman,
Greenberg, & Pyszczynski, 1991), the just world hy
pothesis (Lerner, 1980), and the Piagetian notion of
disequilibration (Piaget, 1932/1965b) all point to the
investment individuals have in their personal stable
interpretations of their world and to the costs of shak-
ing the foundations of those interpretations (see
Pittman, 1998, for a more comprehensive review).
Societal norms and worldviews are typically ab-
sorbed early in development, tend to be subjectively
accepted as reality rather than simply one of many
cultural worldviews, and are likely to be fundamental
aspects of a person’s way of comprehending reality.
They are thus likely to be defended with real vigor
when threatened, and this reaction must be a large
part of the underpinnings of the motivation to give an
offender his or her just deserts. Furthermore, criminal
violations of social norms constitute a rather direct
threat to an individual’s well-being, to the extent that
the possibility of being the victim of such transgres
sions seems more possible and arouses personal fear
and anxiety. An illustration of the important role of
culturally accepted values and attitudes under duress
is provided in a study on the effects of mortality sa
lience on reactions to deviations from accepted cul
tural norms. Rosenblatt, Greenberg, Solomon,
Pyszczyski, et al. (1989) showed that following a re
minder of mortality, their participants reacted more
negatively to a transgression of societal norms and
conversely more positively to behavior reinforcing
such norms. In other words, when under threat, peo
ple revert to the norms of their culture and enforce
them more strongly. A recent study by Oswald,
Hupfeld, Klug, and Gabriel (2002) found additional
support for the role of societal threat in affecting pun
ishments. In a multidimensional scaling of a
large-scale survey, perceived societal threat grouped
closely with the desire to punish and the desire to so
cially exclude the offender.
There are several other converging lines of evidence
which suggest that it is the just deserts impulse that
motivates the assignment of punishment, at least by
citizens of our culture The first line of evidence comes
from many studies (e.g., Klein, Newman, Weis, &
Bobner, 1982) that demonstrate a high correlation be
tween the ratings people give of the compurative “seri
ousness” of various categories of crimes, such as mur
der or theft, and the length of prison sentences that the
raters later assign to persons who commit instances at
those crimes. In addition, there is a good deal of con
sensus among citizens about the comparative serious
ness of crimes. Pontell, Granite, Keenan, and Geis
(1985) demonstrated that police chiefs’ ratings of
crime seriousness show considerable consensus
among chiefs, and with ratings taken from citizens.
Velez Diaz and Megargee (1970) showed that young
criminal offenders rate the comparative seriousness of
categories of crimes in ways very similar to the ratings
of ordinary citizens.
Seriousness, we suggest, measures the moral
weight of the transgression named in the crime, and
the generally high correlation found between the
moral offensiveness of the crime and the sentence it
receives suggests that a just deserts perspective drives
sentencing. For instance, Blum West (1985) found
that the determinants of the perceived seriousness of
an offense are not solely a function of the severity of
the harms done or of the economic losses that the
harm caused, but also of the judgments concerning
the wrongfulness of the motive for the crime and the
criminal’s intent. Rossi, Simpson, and Miller (1985)
reported a similar finding, in that the relation between
perceived crime seriousness and the desired sanctions
was modified by, among other things, characteristics
of the offender. A number of other studies have simi
lar findings (Darley, Carlsmith, & Robinson, 2000;
Finkel, Hurabiell, Hughes, Crystal, & Watanabe,
1993; Harlow, Darley, & Robinson, 1995; Pontell,
Granite, Keenan, & Geis, 1985; Robinson & Darley,
1995; Rossi, Berk, & Campbell, 1997; Rossi, Waite,
Bose, & Berk, 1974; Sanderson, Zanna, & Darley,
2000). Warr, Meier, and Erickson (1983, p. 75) sug
gested that it is the seriousness of offenses rather than
their frequency that is the criterion used to determine
preferred punishments, and commented that their re
DARLEY AND PITTMAN
When the impulse to punish an offender of social norms is
thwarted in some fashion, the impulse to return to an orderly
worldview may be satisfied by turning on the victim. As the victim
is derogated, the world is again seen as an orderly place in which
people get what they deserve (Lerner, 1980). It can be seen, then,
that derogation of the victim also stems from the desire to mete out
sults “cast doubts on recent claims that publicly pre
ferred punishments are based on utilitarian motives.”
Using some recently developed experimental meth
ods, we sought to provide further support for the con
clusion that just deserts concerns motivated sentence
assignments (Jacoby, Jaccard, Kuss, Troutman, &
Mazursky, 1987). The first set of experiments used a
“policy capturing” approach, in which the researchers
created variations in vignettes about crimes, and asked
people to assign the appropriate length of sentence to
each offender about whom they read. Variations in the
story that were relevant to a just deserts or
incapacitative perspective could therefore be examined
for their effect on sentencing. In one study (Darley et
al., 2000), just deserts motives were pitted against
incapacitative motives and in a second (Carlsmith,
Darley, & Robinson, 2002), against deterrence mo
tives. In both studies, the results supported the just
deserts motive for sentencing goals.
A third study (Carlsmith, 2001) used the process
tracing methods of Jacoby et al. (1987). Participants
serially acquired information that they wished to see
to determine what they thought was the appropriate
punishment for a criminal incident. For instance, if
they chose to acquire information about the embez-
zler’s motive, they learned that the embezzlement was
done to get funds to continue to lead a dissolute life.
After they acquired each item of new information,
participants rated their tentative sentence and their
confidence in the correctness of that sentence.
Carlsmith found that people tended to first seek out
information about just deserts and then to later seek
out incapacitative information. Information about de-
terrence, however, was rarely examined. Sequential
judgments of confidence were also affected more by
the just deserts information, and less so by the inca
pacitation information, even correcting for order ef
These results, although they require confirmation
on other respondent populations, suggest that people,
at least in our culture, are motivated by just deserts
concerns when they assign penalties for crimes. Other
cultures may place different emphases on other mo
tives. According to Hamilton and Sanders (1988, p.
183), “Americans were more favorably disposed than
Japanese to retribution, incapacitation and general de
terrence, while Japanese were more favorably disposed
to specific deterrence, rehabilitation, and restitution.”
Lueng and Morris (2001) pointed out that this may be a
general difference found between collectivistic and in
dividualistic cultures, which may in turn be a differ
ence in social-attributional tendencies. Individualistic
cultures image an autonomous, agentic individual, and
if an individual offends, see his internal traits and in
tentions as the cause of the offense. As Na and Loftus
(1998) demonstrated, Korean law experts and college
students attributed criminal behaviors more to situa
tional pressures or social forces than did Americans.
Not surprisingly, therefore, they would put more reli
ance on sanctions that had some promise for reintegrat
ing the individual into the group. However, Carroll,
Perkowitz, Lurigio, and Weaver (1987) pointed out
that within our culture, different individuals may take
different stances toward the causes of crime. Those that
hold a more conservative stance feel that crime is
largely a product of forces within the criminal individ
ual, whereas liberals see crime as the result of social
and institutional forces. As this suggests, within-cul
ture variation in responses to harm and crime are likely
to be as important to consider as cross-cultural varia
tion in these reactions.
Moral Outrage as the Producer of Just
In previous research, we originally used a question
on the degree of “moral outrage” that the witness to an
injustice was feeling as a check on the intensity of the
witness’s reaction to various transgressions. We found
first that this measure correlated quite well with the se-
verity of the punishment the respondent assigned to the
transgressor, and second, that partial correlation tech-
niques demonstrated that moral outrage was a major
mediator between, for instance, perceptions of the spe-
cifics of the crime (such as the motive for which it was
done or the intensity of the harm the harm-doer intended
to do), and the eventual sentence. This caused us to think
further about the concept of moral outrage, its connec-
tions to just deserts reactions to transgressions, and its psy-
chological basis. We first summarize what we have said
about compensatory and punitive justice reactions. Then,
and rather tentatively, we advance some further thoughts
about the centrality of the moral outrage reaction.
Summary—A Model of Compensatory
and Punitive Justice Reactions
Figure 1 depicts our argument thus far. Depending
on the perceiver’s judgment about a perpetrator’s
state of mind, different moral reactions are generated
(high moral outrage in the case of intentional harm,
low moral outrage in the case of negligence, and no
moral outrage in the case of a pure accident). As we
have argued, reckless conduct falls somewhere in be
tween intentional and negligent states of mind, but
we have not represented that path in the model for the
sake of clarity and simplicity. High moral outrage,
coupled with a focus on the perpetrator, leads to re
tributive impulses and to a desire to punish. Low
moral outrage, with a focus on the victim, leads to re
storative impulses and to the imposition of compensa
tion in the case of negligence.
COMPENSATORY AND RETRIBUTIVE JUSTICE
and Retribution in Social Dilemmas
In this issue, Schroeder and his colleagues
(Schroeder, Steel, Woodell, & Bembenek, this issue)
make a convincing case for the analytic utility of jus
tice conceptualizations in understanding behavior in
social dilemmas. Interestingly, they find they need the
concepts of compensation for harms and punishment
for intentional rule-violating harms to complete their
justice-based analysis. We would conceptualize the sit-
uation as follows. If the participants in a social di-
lemma have not come to explicit understandings about
the rules of the game, then a defecting action by one
player may trigger only rule-setting behavior for future
trials. (This might well take the form of an “educa-
tional discussion,” about what all stand to lose if defec-
tions become common.) On our terms, the other play
ers are accepting the claim that the defector did not
understand or perhaps did not intend to inflict dispro
portionate loss on the other players. But once the trial
results are in and the defector sees that (we assume) the
majority sought cooperation, the defector should be
able to realize the undue advantage he has gained. At
that point, a spontaneous offer of compensation made
by the defector can be a very wise move, in that it sig
nals that he made an inadvertent mistake and now rec
ognizes it. This should avoid the negative attributions
that the others would make about a deliberate defec
tion. The reason that defectors would want to avoid
those attributions is that the “psycho logic” of justice is
such that deliberate defections require retributive pun
ishment, and the other group members, if the structure
of the dilemma permits them to do so, are eager to ad
minister that punishment because they are very angry.
Anger aside, there are rational reasons for retribu
tion. As Schroeder’s group (this issue) comments, “ret
ribution serves as a vital group maintenance function,
reasserting the integrity of the group and the values that
it holds” (p. 383). Non-incidentally, it also reminds all
that the rule set, particularly if it was previously made
explicit and agreed on, must not be violated. However,
they also remind us that there is “a retribution-driven
escalatory danger” in situations such as social dilem
mas, in which there is often not an agreed-on sanction
ing system at least initially in place. This is so because
the anger of those transgressed against, and the cer
tainty with which they hold the perception that the of-
fender did it deliberately, may cause them to retaliate
so severely that the transgressor then feels the punish-
ment was excessive, causing him or her to retaliate in
turn when he or she has the next opportunity to do so.
Those of us who run classroom exercises involving the
prisoner’s dilemma know we can count on this dy-
namic occurring in a good many of the teams.
Next, we turn to a consideration of the nature of
moral outrage. What specific emotions might be impli-
cated in the moral outrage reaction?
As the word“outrage” suggests, we think that there is
an emotional component to the reaction to wrongdoing.
Some havesuggested the same idea and developed theo
ries that enable us to be more precise about what emo
tions are at issue. For instance, Rozin, Lowery, Imada,
and Haidt (1999) have advanced the “CAD hypothesis”
suggesting that across cultures, three emotions—con
tempt, anger, and disgust—become “moralized” during
development and become the emotions that form the
emotional components of moral reactions. Rozin et al.
further adopted Shweder’s suggestion that each of these
emotions is uniquely tied to violations in a particular
sphere of human actions that are regarded as moral
transgressions if one observes moral discourse
cross-culturally (Shweder, Mahapatra, & Miller, 1987;
Shweder, Much, Mahapatra, & Park, 1997). Spe
cifically, the moral emotions are contempt, anger, and
disgust (thus CAD). Anger, they suggest, is associated
with violations of a person’s autonomy. These auton
omy violations, in which another individual’s person or
DARLEY AND PITTMAN
Figure 1. Justice assignment following a case of harm-doing as a function of perpetrator’s judged state of mind, perceiver’s moral reac
tions, and perceiver’s retributive or restorative impulses.
property is harmed, trigger an angry emotional reaction
that is specifically moral in tone. This moral anger reac
tion is probably closest to the one we indexed when we
asked respondents about the degree of moral outrage
that they felt in response to reports of transgressions.
A second moral emotion in their system is disgust,
which has an interesting role in our culture.
“Sociomoral disgust is triggered by a variety of situa
tions in which people behave without dignity or strip
others of their dignity” (Rozin et al., 1999, p. 576).
Racist actions and child abuse frequently provoke dis
gust reactions among Americans, although a wider
range of violations of purity and divinity provoke dis
gust in other cultures.
The third moral emotion is contempt, generated to
ward those we regard as lower in the social hierarchy.
This is the sort of emotion that racists feel for the tar
gets of their prejudice. It might be termed a moral reac
tion of one who is high on the social dominance scale,
and feels that an assertion of social dominance is ap
propriate (Sidanias & Pratto, 1999).
Anthropologists, particularly the psychologically
minded anthropologist Rick Shweder (Shweder et al.,
1987), have suggested that our culture’s equation of
moral wrong solely with the act of harm-doing to an-
other individual is a product of a culture that has the im-
age of the individual as an autonomous, independently
functioning, agentic individual, whose actions should
not be limited unless they damage the well-being of oth-
ers. Certainly this is the image that, for instance, arises
from Mill’s (1993) discussions on liberty and morality.
We are, in other words, a culture in which we would ex-
pect anger to be the most frequent response to things we
regard as moral violations because those violations are
ones in which other people’s autonomy of action has
been violated by harms to their person or property.
of the Moral Emotion Analysis
Researchers may be right or wrong in their specifi
cations of the emotions that are the ones that become
part of the moral judgmental system, or even about
whether emotions are part of the moral judgmental
process. Still, a number of interesting hypotheses arise
from combining the work on moral emotions with
work on retributive justice.
First, in all cultures, it is those violations arousing
moral emotions that trigger the retributive reaction,
rather than violations that produce nonmoral emotions
or no emotions at all. Violations that trigger moral emo
tions are those that threaten or undermine the person’s
fundamental worldview in ways that require a strong re
tributive response. The arousal of nonmoral emotions
may be triggered by other sorts of violations and may
produce negative reactions to the person arousing the
emotion, but not retributive ones. What those reactions
are, and what sorts of violations produce them, remain to
be specified(and it wouldbe an interesting task to do so),
but they do not provide the license to inflict punishment
on or require compensation from the other.
Second, different cultures may weight the three
moral emotions differently. For instance, we would
predict that our culture generates retributive reactions
largely to autonomy violations and there is evidence
that this is so. “The divinity code is so foreign to the
moral system used by educated Americans that most of
these participants assigned what we took to be such vi
olations (ones generating disgust) to the non-moral
category” (Haidt, Koller, & Dias, 1993).
Recently, Tetlock has been elaborating a related no-
tion of sacred values, values that create categories and
events that should not be traded off for more secular
commodities and benefits. Adding certain specific con-
cepts from his general thinking (Tetlock, Kristel, Elson,
Green, & Lerner, 2000) about people’s reactions to what
he calls taboo trade-offs and forbidden base rates to the
present discussion generates some further questions for
research. Is it the case that a person who seriously pro-
poses a taboo trade-off is seen to have violated a funda-
mental social truth in a way that provokes one of the
moral emotions and generates a retributive response? If
a rich person offers to buy a kidney from your daughter
to implant in her daughter, is she transgressing sacred
valuesin ways that trigger the moral emotional reactions
that arise when right-thinking people feel it appropriate
to mete out punishment? Accessing a forbidden base
rate may also generate moral-emotional reactions and
punitive desires. If the use of racial statistics for profil
ing to determine whom to stop and frisk for drugs or
weapons offends many of us, is it because it violates
some fundamental value in a way that arouses emotional
reactions of anger, or contempt, or disgust in us directed
toward the profiling individual?
One theoretically “clean” pattern would arise if it
were the intentional wrongdoing actions that generated
strong moral-emotional reactions of all three sorts, be
cause we have demonstrated that it is those actions that
bring forth in respondents the assignment of
retributional punishment in response. This hypothesis
holds that the hostility triad (anger, contempt and dis
gust) is uniquely associated with the retributive reaction
to intentional wrongs. The unintentional case of negli
gently not foreseeing a harm that one’s actions caused
COMPENSATORY AND RETRIBUTIVE JUSTICE
Briefly, two other clusters of emotions have morally relevant
properties in the Rozin et al. (1999) system. Shame, embarrass
ment, and guilt “involve ongoing assessments of the moral worth
and fit of the individual self within the community. They are
internalizations of the social order within the individual” (p.
574–575). Thus, these would be the emotions a person might feel
when he or she has transgressed against others in the social system.
Other emotions seem to pick up on the suffering of others, are la
beled other-suffering emotions, and may motivate acts of positive
justice toward those individuals.
does not generate the moral emotions, and the in-be
tween case of reckless action generates moral emotions
of a lower intensity than the intentional ones, along with
the occasional assignment of a retributional punishment
over and above the compensatory one that is usually in
voked. Alternately, a differential moral-emotional
cueing system is possible; in it, intentional wrongdoing
might invokeanger,strongly connected with the retribu
tive impulse resulting from violation of the norm of per
sonal autonomy. Negligently causing accidents may
cause the negligent individual to be regarded with con
tempt, an emotion “often felt by members of one group
for the members of other groups regarded as inferior”
(Rozin et al., 1999, p. 575). It would then be the emotion
of contempt that is associated with the judgment that the
harm-doer must compensate those who were negli
gently harmed. Cultural variants are possible; for in
stance, in a culture that recognizes violations of purity
and divinity as moral offenses, these offenses might
generate retribution and the purity-violation-marking
emotion of disgust.
The previous discussion suggests that a strong con
comitant of the desire to retributively punish an inten
tional transgressor is the moral emotion(s) that the
transgression provokes. The suggestion that emotion is
involved in moral reactions is not novel, but it is in de-
tail certainly rather new. The contrast suggestion, of
course, is that moral reactions are cognitive in nature, a
stance that is characteristic of much of Piaget’s (1965a,
1932/1965b) work and certainly of Kohlberg’s (1969)
thinking. Lately, however, scholars are asserting the
basic role of emotions in moral judgmental processes
(Kagan & Lamb, 1987, Shweder & Haidt, 1993). As
Rozin et al. pointed out “cross-cultural work has begun
to demonstrate the cognitive-developmental theories
work less well outside of Western middle-class popula
tions and that emotional reactions are often the best
predictors of moral judgments” (1999, p. 574; they
cited Haidt, Kollar, & Dias, 1993, and Shweder et al.,
1987, to support this conclusion.)
The exact nature of the causal connections among
motivational, cognitive, and emotional reactions to
harm-doing remains to be explored. For example, emo
tional reactions may be a by-product of moral viola
tions, or they may play a central causal role in the de
termination of retributional responses. There are some
initial indications that the retributive impulse gets
some of its energy from emotional reactions. First,
Goldberg, Lerner, and Tetlock (2000) have conducted
several studies which demonstrate that there may be a
carryover of anger that is not resolved (because one
harm-doer escapes punishment) to a second offender in
a second story who then draws a more severe sentence.
This is reminiscent of the “excitation transfer effect”
studies of Zillman, Katcher, and Milavsky (1972) who
demonstrated that emotional arousal could persist and
“leak” into the intensity of emotions felt in later situa
tions, or of studies of the misattribution of arousal, in
which reactions such as attitude change following dis
sonance induction were magnified or short-circuited
by the interpretations placed on accompanying arousal
states (e.g., Pittman, 1975; Zanna & Cooper, 1974).
There may also, of course, be more complex and pri
marily cognitive explanations for the Tetlock et al. re
sult, but it does fit well with a hypothesis that holds that
the magnitude of the retributional response is to some
extent tied to the level of moral-emotional arousal pro
voked by the transgression.
Finally, some recent neural imaging research sup
ports the broad claim that emotional reactions, as well
as cognitive ones, are involved in moral judgments. In
one study (Greene, Somerville, Nystrom, Darley, &
Cohen, 2001), respondents chose between two alter
nate responses to various dichotomous decision tasks
they were given. Some of the decisions were of the sort
the authors called “personal moral” decisions: whether
for instance to push a specifically imagined other per
son off a bridge to block the path of a runaway trolley
car that would otherwise strike and kill five people, or
to smother a crying baby whose cries might give away
the location of a whole group hiding from those who
sought to find and kill them. During the contemplation
of this set of problems, many brain areas were subject
to functional magnetic resonance imaging to deter-
mine which of them were differentially active. The re-
sults clearly indicated that the areas found to be differ-
entially active in the moral-personal condition were
first, reasoning areas, and second, ones that recent
functional imaging studies have found to be associated
There are a number of ways of describing the interac
tion of cognition and emotion that we are suggesting is
involved in moral judgments of the sort described here.
One accords generally with a number of statements of
the emotional appraisal perspective. Various complex
stimuli require quite elaborate cognitive appraisal. For
instance, what counts as a threatening gesture differs
considerably among cultures, as does what counts as a
demeaning or a provoking remark. Obviously to “de
code” the meaning of these remarks is a complex task
which involves considerable cognitive work. (This is
not to deny that this work, on the part of the normally so
cialized individualin a culture, proceeds in ways that are
well practiced and mostly automatic.) However, such
gestures or remarks are also experienced as instantly
generating various emotions and the generally auto
DARLEY AND PITTMAN
To complete the story, nonmoral scenarios were ones involving
the solution of problems rather like arithmetic ones. The
“moral-nonpersonal” problems were ones that people would tend to
say involved moral issues (in fact, they are ones that philosophers
have tended to regard as revealing people’s moral thinking), but did
not seem to the researchers to have the “involvement” of problems
we would say were prototypic moral ones. In the research they
tended to have the brain signature of the nonmoral scenarios.
matic recruitment of a ready set of responses. The
neuroscientist Damasio (1994) talked about the arousal
of certain emotions as ones that carry motivational sig
nal value about states of affairs to seek out, or avoid, or
perhaps to condemn, and this seems to be a perspective
that is compatible with what we are suggesting about the
role of emotions in moral judgments. Obviously, a great
dealof research is needed to sortall of this out, and it may
be prudent to consider whole patterns of reactions rather
than to single out one particular aspect of the complex
reactions to moral violations. In the end, it may be a bad
idea to make too definite a separation between “cogni
tion” and “motivation and emotion.”
Alicke, M. D. (2000). Culpable control and the psychology of blame.
Psychological Bulletin, 126, 556–574.
Bentham, J. (1970). An introduction to the principles of morals and
legislation. London: The Athone Press. (Original work pub
Blum West, S. R. (1985). The seriousness of crime: A study of popu
lar morality. Deviant Behavior, 6, 83–98.
Carlsmith, K. M. (2001). Why do we punish? Retribution, deter-
rence, and incapacitation as motives for punishment. Disserta-
tion Abstracts International, 62, 3B. (University Microfilms
Carlsmith, K. M., Darley, J. M., & Robinson, P. H. (2002). Why do
we punish? Deterrence and just deserts as motives for punish-
ment. Journal of Personality and Social Psychology, 83,1–16.
Carroll, J. S., Perkowitz, W. T., Lurigo, A. J., & Weaver, F. M. (1987).
Sentencing goals, causal attributions, ideology, and personality.
Journal of Personality and Social Psychology, 52, 107–118.
Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and the
human brain. New York: Putnam.
Darley, J. (2002). Just punishments: Research on retributional justice.
In M. Ross & D. T. Miller (Eds.), The justice motive in everyday
life (pp. 314–333). New York: Cambridge University Press.
Darley, J. M. (2001). Citizens’ sense of justice and the legal system.
Current Directions in Psychological Science, 10, 10–13.
Darley, J. M., Carlsmith, K. M., & Robinson, P. H. (2000). Incapaci
tation and just deserts as motives for punishment. Law and Hu
man Behavior, 24, 659–683.
Darley, J. M., Klosson, E. C., & Zanna, M. P. (1978). Intentions and
their contexts in the moral judgements of children and adults.
Child Development, 49, 66–74.
Elster, J. (1990). Norms of revenge. Ethics, 100, 862–885.
Enzle, M. E., & Hawkins, W. L. (1992). A priori actor negligence
mediates a posteriori outcome effects on moral judgment. Jour
nal of Experimental Social Psychology, 28, 169–185.
Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL:
Finkel, N. J., Hurabiell, M. L., Hughes, K. C., Crystal, D. S., &
Watanabe, H. (1993). Commonsense justice: Jurors’ notions of
the law. Law and Human Behavior, 17, 487–506.
French, P. (2001). The virtues of vengeance. Lawrence: University of
Goldberg, J. H., Lerner, J. S., & Tetlock, P. E. (1999). Rage and rea
son: The psychology of the intuitive prosecutor. European
Journal of Social Psychology, 29, 781–795
Greene, J. D., Somerville, R. B., Nystrom, L. E., Darley, J. M., & Co
hen, J. D. (2001). An fMRI investigation of emotional engage
ment in moral judgment. Science, 293, 2105–2108.
Haidt, J., Koller, S. H., & Dias, M. G. (1993). Affect, culture and mo
rality, or is it wrong to eat your dog? Journal of Personality and
Social Psychology, 65, 613–628.
Hamilton, V. L., & Sanders, J. (1988). Punishment and the individual
in the United States and Japan. Law and Society Review, 22,
Harlow, R. E., Darley, J. M., & Robinson, P. H. (1995). The severity
of intermediate penal sanctions: A psychophysical scaling ap
proach for obtaining community perceptions. Journal of Quan
titative Criminology, 11, 71–95.
Hogan, R., & Emler, N. P. (1981). Retributive justice. In S. C. Lerner
(Ed.), The justice motive in social behavior: Adapting to times
of scarcity and change (pp. 125–143). New York: Plenum.
Jacoby, J., Jaccard, J., Kuss, A., Troutman, T., & Mazursky, D.
(1987). New directions in behavioral process research: Implica
tions for social psychology. Journal of Experimental Social
Psychology. 23, 146–175.
Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The
attribution process in person perception. In L. Berkowitz
(Ed.), Advances in experimental social psychology (Vol. 2, pp.
220–266). New York: Academic.
Kagan, J., & Lamb, S. (1987). The emergence of morality in young
children. Chicago: University of Chicago Press.
Karlovac, M., & Darley, J. M. (1988). Attribution of responsibility
for accidents: A negligence law analogy. Social Cognition, 4,
Keeton, W., Dobbs, D., Keeton, R., & Owen, D. (1984). Prosser and
Keeton on the Law of Torts (5th ed.). St. Paul, MN: West.
Kelley, H. H. (1967). Attribution theory in social psychology. In D.
Levine (Ed.), Nebraska symposium on motivation (Vol. 15, pp.
192–238). Lincoln: University of Nebraska Press.
Klein, R. J., Newman, I., Weis, D. M., & Bobner, R. F. (1982). The
Continuum of Criminal Offenses instrument: Further develop-
ment and modification of Sellin and Wolfgang’s original crimi-
nal index. Journal of Offender Counseling, Services and Reha-
bilitation, 7, 33–53.
LaFave, W. (2000). Criminal law (3rd. ed.). St. Paul, MN: West.
Lempert, R. (1972). Norm-making: A contract law model. Law and
Society Review, 7, 1–32.
Lerner, M. J. (1980). The belief in a just world: A fundamental delu
sion. New York: Plenum.
Lueng, K., & Morris, M. (2001). Justice through the lens of culture
and ethnicity. In J. Sanders & V. Hamilton (Eds.), Handbook of
justice research in law (pp. 343–378). New York: Kluwer.
Mill, J. S. (1993). On liberty and utilitarianism. New York: Bantam.
Na, E. Y., & Loftus, E. F. (1998). Attitudes toward law and prisoner,
conservative authoritarianisms, attribution, and internal–exter
nal locus of control: Korean and American law students and
undergraduates. Journal of Cross-Cultural Psychology, 29,
Nozick, R. (1981). Philosophical explanations. Cambridge, MA:
Harvard University Press.
Oswald, M. E., Hupfeld, J., Klug., S. C., & Gabriel, U. (2002). Lay
perspective on criminal deviance, goals of punishment, and
punitivity. Social Justice Research, 15, 85–98.
Piaget, J. (1965a). The child’s conception of the world. Paterson, NJ:
Piaget, J. (1965b). The moral judgment of the child (M. Gabain,
Trans.). New York: Free Press. (Original work published 1932)
Pittman, T. S. (1975). Attribution of arousal as a mediator in disso
nance reduction. Journal of Experimental Social Psychology,
Pittman, T. S. (1993). Control motivation and attitude change. In G.
Weary, F. Gleicher, & K. Marsh (Eds.), Control motivation and
social cognition (pp. 157–175). New York: Springer-Verlag.
Pittman, T. S. (1998). Motivation. In D. Gilbert, S. Fiske, & G.
Lindsay (Eds.), Handbook of social psychology (4th ed., pp.
549–590). Boston: McGraw-Hill.
COMPENSATORY AND RETRIBUTIVE JUSTICE
Pittman, T. S., & Darley, J. M. (2003). Intent, recklessness, negli
gence, and punishment: The role of state of mind in the as
sessment of restorative and retributional justice. Unpublished
manuscript, Princeton University, New Jersey.
Pittman, T. S., Darley, J. M., & Robinson, P. H. (2003). The effectof the
timing and nature of expressions of remorse on sentencing judg
ments: Timing and sincerity matter, victim forgiveness does not .
Unpublished manuscript, Princeton University, New Jersey.
Pittman, T. S., & Pittman, N. L. (1980). Deprivation of control and
the attribution process. Journal of Personality and Social Psy
chology, 39, 377–389.
Pontell, H. N., Granite, D., Keenan, C., & Geis, G. (1985). Serious
ness of crimes: A survey of the nation’s chiefs of police. Jour
nal of Criminal Justice, 13, 1–13.
Robinson, P. H., & Darley, J. M. (1995). Justice, liability, and
blame: Community views and the criminal law. Boulder, CO:
Rosenblatt, A., Greenberg, J., Solomon, S., Pyszynski, T., et al.
(1989). Evidence for terror management theory: I. The effects
of morality salience on reactions to those who violate or uphold
cultural values. Journal of Personality and Social Psychology,
Rossi, P. H., Berk, R. A., & Campbell, A. (1997). Just punish
ments: Guideline sentences and normative consensus. Journal
of Quantitative Criminology, 13, 267–290.
Rossi, P. H., Simpson, J. E., & Miller, J. L. (1985). Beyond crime
seriousness: Fitting the punishment to the crime. Journal of
Quantitative Criminology, 1, 59–90.
Rossi, P. H., Waite, E., Bose, C. E., & Berk, R. E. (1974). The seri-
ousness of crimes: Normative structure and individual differ-
ences. American Sociological Review, 39, 224–237.
Rozin, P., Lowery, L., Imada, S., & Haidt, J. (1999). The CAD Triad
Hypothesis: A mapping between three moral emotions (con-
tempt, anger, disgust) and three moral codes (community, au-
tonomy, divinity). Journal of Personality and Social Psychol-
ogy, 76, 574–586.
Sanderson, C. A., Zanna, A. S., & Darley, J. M. (2000). Making the
punishment fit the crime and the criminal: Attributions of dan-
gerousness as a mediator of liability. Journal of Applied Social
Psychology, 30, 1137–1159.
Schlenker, B. R., Britt, T. W., Pennington, J., & Murphy, R. (1994).
The triangle model of responsibility. Psychological Review,
Schroeder, D. A., Steel, J. E., Woodell, A. J., & Bembenek, A. F.
(2003). Justice within social dilemmas. Personality and Social
Psychology Review, 7, 374–387.
Seligman, M. E. P. (1975). Helplessness: On depression, develop
ment, and death. San Francisco: Freeman.
Shaver, K. (1985). The attribution of blame: Causality, responsibil
ity, and blameworthiness. New York: Springer-Verlag.
Shweder, R., & Haidt, J. (1993). The future of moral psychology.
Truth, intuition, and the pluralist way. Psychological Science, 4,
Shweder, R. A., Mahapatra, M., & Miller, J. (1987). Culture and
moral development. In J. Kagan & S. Lamb (Eds.), The emer
gence of morality in young children (pp. 1–83). Chicago: Uni
versity of Chicago Press.
Shweder, R. A., Much, N. C., Mahapatra, M., & Park, L. (1997). The
“big three” of morality (autonomy, community, divinity) and
the “big three” explanations of suffering. In A. Brandt & P.
Rozin (Eds.), Morality and health (pp. 119–169). New York:
Sidanias, J., & Pratto, F. (1999). Social dominance. New York: Cam
bridge University Press.
Snyder, C. R., Higgins, R. L., & Stuckey, R. J. (1983). Excuses: Mas
querades in search of grace. New York: Wiley.
Soloman, S., Greenberg, J., & Pyszczynski, T. (1991). A terror
management theory of social behavior: The psychological
functions of self-esteem and cultural worldviews. In M. Zanna
(Ed.), Advances in experimental social psychology (Vol. 24,
pp. 93–159). New York: Academic.
Tetlock, P. E., Kristel, O. V., Elson, S. B., Green, M. C., & Lerner. J.
S. (2000). The psychology of the unthinkable: Taboo trade-offs,
forbidden base rates, and heretical counterfactuals. Journal of
Personality and Social Psychology, 78, 853–870.
Tyler, T. R., & Boeckmann, R. (1997). Three strikes and you are out,
but why? Law and Society Review, 3, 237–265.
Velez Diaz, A., & Megargee, E. I. (1970). An investigation of dif
ferences in value judgments between youthful offenders and
nonoffenders in Puerto Rico. Journal of Criminal Law, Crimi-
nology and Police Science, 61, 549–553.
Vidmar, N. (2001). Retribution and revenge. In V. L. Hamilton (Ed.),
Handbook of justice research in law (pp. 31–63). New Haven,
CT: Yale University Press.
Vidmar, N. (2002). Retributive justice: Its social context. In M. Ross
& D. T. Miller (Ed.), The justice motive in everyday life (pp.
291–313). New York: Cambridge University Press.
Vidmar, N., & Miller, D. T. (1980). Social-psychological processes
underlying attitudes toward legal punishment. Law & Society
Review, 14, 565–602.
Warr, M., Meier, R. F., & Erickson, M. L. (1983). Norms, theories of
punishment, and publicly preferred penalties for crimes. Socio
logical Quarterly, 24, 75–91.
Weiner, B. (1995). Judgements of responsibility: A foundation for a
theory of social conduct. New York: Guilford.
Zanna, M., & Cooper, J. (1974). Dissonance and the pill: An attribu
tion approach to studying the arousal properties of dissonance.
Journal of Personality and Social Psychology, 29, 703–709.
Zillman, D., Katcher, A. H., & Milavsky, B. (1972). Excitation trans
fer from physical exercise to subsequent aggressive behavior.
Journal of Experimental Social Psychology, 8, 247–259.
DARLEY AND PITTMAN