Printer Friendly

Punishment and moral sentiments.

In 1883, the Victorian Judge James Fitzjames Stephen famously argued that:
   The sentence of the law is to the moral sentiments of the public in
   relation to any offence what a seal is to hot wax. It converts into
   a permanent final judgment what might otherwise be a transient
   sentiment ... the infliction of punishment by law gives definite
   expression and solemn justification to the hatred which is excited
   by the commission of the offence. (1)

It is often the case that our passions become unnerved upon learning that a heinous crime has taken place. However, what is the relationship between our moral sentiments and the justification of punishment? One position echoed above is that our moral sentiments provide for punishment's justification. My focus here will be on Adam Smith's theory of punishment and the role that moral sentiments play in this theory. I will argue that interpreters, such as Knud Haakonssen, Eric Miller, and John Salter, are mistaken to view Smith's position as essentially retributivist. (2) In fact, Smith defends a unified theory where punishment serves retributivist, deterrent, and rehabilitative goals together. I will conclude with some critical remarks on how well this account may speak to contemporary concerns.


Smith's theory of punishment. Smith begins by asking us to assess the merit or demerit of actions from the standpoint of an impartial spectator. (3) This device offers us what I will refer to as "the impartiality standard." (4) This standard is used to determine whether a given act deserves punishment. Therefore, we are not to focus on our own particular judgments, but judge instead from the perspective of an impartial observer. (5) We assess actions how we imagine the impartial spectator should assess them if called upon. For Smith, satisfaction of this impartiality standard is a guarantee of impartial justice.

Determining whether or not an act requires punishment in fulfillment of the impartiality standard requires that we examine individual actions in light of three factors. The first pertains to the intentions behind an action. (6) Our intentions may be met with approval or disapproval as determined by an impartial spectator. If our intentions do not merit disapproval, then we cannot be susceptible to punishment. However, we may become susceptible to punishment when the opposite is true. Nevertheless, the mere possession of a censurable intention is insufficient to merit punishment. For Smith, punishing someone solely on account of their thoughts where no crime has occurred is "the most insolent and barbarous tyranny." (7)

The second factor relates to the external action performed. Thus, the impartial spectator begins by asking what was intended and then proceeds to inquire about what action took place. These external actions do not always admit of approval or disapproval independently of other factors. For example, an external action may consist of pulling the trigger on a gun. This action may be morally problematic, but it may not. Perhaps I was simply practicing my marksmanship skills at a shooting range. Of course, this action may become morally problematic in different contexts, such as if I was shooting my gun with an intent to kill an innocent person. External actions become morally problematic in light of the intentions that give rise to them.

A third factor concerns the consequences of an action: we ask what happened as a result of an action. Not unlike external actions, consequences also do not readily admit of approval or disapproval taken separately. For example, a consequence of an action may be the death of another person at my hands. Whether or not this consequence is morally problematic will depend on my relevant intentions. Therefore, if the consequence arose from my acting in self-defense, then the consequence is not morally problematic. The situation would be rather different if I was acting murderously. We may only be held to account for consequences that we have intended. (8)

We assess actions in light of these factors when we consider whether an action may deserve punishment. (9) Punishable actions are not merely actions that would be met with disapproval by the impartiality standard, but something more. Instead, punishable acts are actions that "must appear to deserve punishment." (10) We may satisfy this standard when an action possesses both an intention meriting disapproval and harmful consequences. For Smith, these often run together. He says: "There can be no proper motive for hurting our neighbour, there can be no incitement to do evil to another, which mankind will go along with, except just indignation for evil which that other has done to us." (11) The intentional infliction of harm to others is punishable. This position shares some similarity with John Stuart Mill's harm principle which states that "the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others." (12) Harm to others should be always legally forbidden.

When another is responsible for harmful consequences that affect others, punishment is deserved. (13) On this view, punishable acts are other-regarding harms: we punish harm inflicted on others alone. (14) We assess the severity of intentional harms in proportion to an impartial spectator's resentment over what befell the victim. (15) Smith argues:
   in all cases the measure of the punishment to be inflicted on the
   delinquent is the concurrence of the impartial spectator with the
   resentment of the injured. If the injury is so great as that the
   spectator can go along with the injured person in revenging himself
   by the death of the offender, this is the proper punishment. (16)

Punishment is, in part, backward looking, returning "evil for evil that has been done." (17) We look back to the intention that was formed in support of an action and the effect the harmful consequences had on the injured. Our concern is not whether the injured actually does feel justified resentment on account of the harm he has endured. Instead, our perspective remains that of the impartial spectator and of the impartiality standard: punishment is deserved where a person becomes "the natural object of a resentment which the breast of every reasonable man is ready to adopt and sympathize with." (18) Thus, our concern is whether an impartial spectator would feel justified resentment upon considering an individual case. This resentment finds satisfaction only where punishment is inflicted on the persons responsible for causing justified resentment. (19)

Punishment must be proportionate. Smith says:
   As the greater and more irreparable the evil that is done, the
   resentment of the sufferer runs naturally the higher; so does
   likewise the sympathetic indignation of the spectator, as well as
   the sense of guilt in the agent. (20)

The proportionality of punishment mirrors the degree of the impartial spectator's resentment in considering the criminal's responsibility for causing harm and the suffering of the criminal's victim after being subjected to harm: the greater a criminal's guilt and the victim's suffering, the more severe the appropriate punishment. We ought not punish anyone more severely than would be agreeable to any indifferent person lest our action become most detestable. (21) This follows from the impartiality standard. Furthermore, our punishments should not degrade criminals. Thus, Smith argues that "[t]he judge who orders a criminal to be set in the pillory, dishonours him more than if he had condemned him in the scaffold." (22) The problem is that the criminal will likely feel irrevocably degraded by the punishment when he ought to feel most degraded by his criminal act. (23) Therefore, shaming criminals should be avoided. (24)

The punishment of criminals satisfies our impartial resentment. Punishment is deserved where the impartial spectator should rightly feel resentment on account of wrongful actions. However, our resentment is not satisfied when we have merely identified who should be punished, but when they are punished. Recall that our consideration of resentment pertains both to the criminal and to the public. We are attentive both to the criminal's responsibility for a harmful action and the suffering of those at his hands.

Punishment satisfies resentment with respect to both criminals and the public. First, punishment aims to enable the criminal to grieve and repent for his action. For example, Smith argues that our impartial resentment will remain unsatisfied until the criminal is "made to repent and be sorry for this very action." (25) It is not enough that a criminal is sorrowful in being punished for his crime, but that he feels remorse because he has committed a crime. (26) A criminal must both accept responsibility for his crime and express regret. Smith's theory of punishment is then more than retributivist alone. Not only must punishment be deserved and proportionate as retributivists would maintain, but criminals should become remorseful for the crimes they have committed in being subjected to punishment. This rehabilitative feature of Smith's theory further demands that our punishments not be degrading, or they risk being counterproductive.

Secondly, punishment has the further aim to serve as an example to the public in order to promote general deterrence. (27) Smith says, "that others, through fear of the like punishment, may be terrified from being guilty of the like offence." (28) Of course, this is not a theory that would permit any penal practice that might yield a deterrent effect. Nor would it ever entail the punishment of innocent persons, if this could have a deterrent effect. This is because this view of deterrence is predicated, again, on a view of deserved and proportionate punishment. When we choose which form this punishment might take, we must then take care to help enable a criminal to be sorry for his crime and also give reason for others to avoid performing crimes for fear of being punished. Therefore, deserved and proportionate punishments must "produce all the political ends of punishment; the correction of the criminal, and the example to the public." (29)

Punishment, thus, has a hybrid character where retribution, deterrence, and rehabilitation are brought together in an interesting way. (30) For Smith, punishment is given only to the deserving and in proportion. This is its retributive feature. Furthermore, punishment aspires to reform criminals by making their repentance possible. This is its rehabilitative feature. Finally, punishment is meted out in order to dissuade members of the public from criminal activity. This is its deterrent feature. Taken together, Smith offers a unified theory of punishment bringing together retributive, deterrent, and rehabilitative features. It is important to recognize this multidimensional character of Smith's theory. He rejects punishment as the imposition of suffering on criminals for its own sake. Instead, punishment is both a reaction to harm, but also a defensive mechanism. This defense takes the shape of trying to help bring about a change of heart in a criminal by helping him recognize his wrongdoing as wrongdoing so that his criminality will end. (31) Smith's theory of punishment is retributivist to the extent that desert and proportionality play a central role. However, his theory of punishment is also not merely retributivist in its attempt to incorporate both rehabilitative and deterrent features. (32)


Smith on the sentinel. I have argued that Smith's theory of punishment is not merely retributivist in character, but that it also incorporates rehabilitative and deterrent elements within a multidimensional position. In both his Lectures on Jurisprudence and The Theory of Moral Sentiments, Smith offers us the sentinel case, and this example highlights well the way in which Smith's theory of punishment is multidimensional.

The sentinel case is as follows. Smith says:
   the military laws punish a sentinell who falls asleep upon guard
   with death. This is intirely founded on the consideration of the
   publick good; and tho we may perhaps approve of the sacrificing one
   person for the safety of a few, yet such a punishment when it is
   inflicted affects us in a very different manner from that of a
   cruel murtherer or other atrocious criminall. (33)

Here the sentinel is on watch, but fails to complete his shift and falls asleep at his post. There is no suggestion that the sentinel intends to do harm nor need harm actually be a result. Nonetheless, Smith is clear that the sentinel would require punishment in this case, namely, death.

Smith is not arguing that sentinels are akin to murderers in making this claim. (34) While both may deserve the same punishment, their crimes are deserved for different reasons. The sentinel alone deserves capital punishment "on the consideration of the public good" rather than on any intention to harm. (35) Smith says:
   For tho in many cases the publick good may require the same degree
   of punishment as the just revenge of the injur'd, and such as the
   spectator would go along with, yet in those crimes which are
   punished chiefly from a view to the publick good the punishment
   enacted by law and that which we can readily enter into is very
   different. (36)

Thus, the maintenance of the public good may require punishment as severe as that meted to murderers. While an impartial spectator would agree with this situation, Smith says also elsewhere:
   [The impartial spectator] looks upon the sentinel as an unfortunate
   victim, who, indeed, must, and ought to be, devoted to the safety
   of numbers, but whom still, in his heart, he would be glad to save;
   and he is only sorry, that the interest of the many should oppose
   it. (37)

However unfortunate, the sentinel's punishment is just.

The sentinel case is instructive in several respects. First, Smith discusses the sentinel case in more than one work. Second, it is not the only case where Smith argues that the maintenance of public safety may be the primary factor when determining the severity of punishment. (38) Together, this helps offer us some further evidence that Smith's theory of punishment is more than merely retributive. If it was only retributivist as others have argued, then the sentinel case poses a real problem. The sentinel intends to harm no one and no one may be harmed. Nonetheless, the sentinel has, however innocently, endangered the public good beyond a tolerable extent. For this reason, he deserves capital punishment. In fact, I would argue that this case does not pose a real problem for Smith. The reason is that Smith's theory is not merely retributivist and, in fact, it is sensitive to additional features, such as deterrence and the need to promote public safety. Thus, Smith's discussion of the sentinel case is consistent with his multidimensional theory of punishment. (39)


Two possible concerns. I have argued that Smith has a multidimensional theory of punishment that incorporates retributive, deterrent, and rehabilitative elements. This hybrid interpretation is further able to account for certain examples, such as the sentinel case. Thus far, my aims have been to reconstruct a new understanding of this theory. In this section, I want to raise a few possible concerns before concluding in the next and final section.

One possible concern is that, for Smith, punishable crimes are other-regarding and not victimless. The worry is not that Smith's views are anachronistic. In fact, few were as sensitive as Smith to the historical character of punishment. (40) Nevertheless, punishment requires an action with harmful or likely to be harmful consequences for others?' Therefore, we would punish the thief, the murder, and even the sentinel. Others are harmed through loss of property, life, and general safety through these actions.

The situation is different with self-regarding harms that are victimless. Examples might include driving offenses, drug offenses, or possibly prostitution. If someone were to become addicted to methamphetamine, then they would perform an action causing harm although only to himself. The concern is not that Smith should have anticipated a need for drug offenses: again, my claim is not that Smith's views are anachronistic. Instead, the possible concern is that self-regarding offenses, such as drug offenses, as a category appear beyond the realm of punishable acts. (42) Of course, this is only a possible concern depending upon whether you believe victimless crimes should be recognized as punishable crimes: it would be no inconsistency for Smith to stand his ground and more libertarian-friendly readers might warmly endorse this position.

A second possible concern is that Smith's theory of punishment is mistaken to understand punishable actions as moral wrongs. Crimes are punishable actions. We determine punishable actions with respect to an impartial spectator's consideration of the intention behind an action, the external action, and the presence of harmful consequences. Actions that intentionally harm others are punishable and its proportionality is determined with reference to "the concurrence of the impartial spectator with the resentment of the injured." (43) Punishable actions must produce an evil and, thus, crimes are acts of moral wrongdoing. (44) Therefore, crimes are both other-regarding harms and mala in se, or wrong-in-itself, as evil actions. Examples might include murder, theft, and rape. These crimes do not admit of exceptions: while it may be permissible to kill another perhaps in self-defense, it is never permissible to murder the innocent.

Again, those who are favorable to natural law and legal moralist jurisprudence will be untroubled by this account. My claim is not that Smith's position is inconsistent nor, once again, am I claiming his position is anachronistic. Instead, I merely raise the possible concern that not all crimes may be evil actions. The sentinel may perhaps inadvertently perform an act of evil in falling asleep at his post, as this may endanger the public safety of his community. It is less clear how other criminal acts are evil, especially self-regarding and victimless crimes often regarded as mala prohibita, or wrong-as-legally-prohibited. (45) For example, some US highways reduced maximum speed limits to 55 miles per hour, not because of safety concerns, but in order to better facilitate oil conservation in the early 1970s.

Another example is treason. Smith does argue in favor of punishing treason. He states that traitors are "the most detestable" of villains and that "treason is perhaps the only exception" to his rule that we always punish criminal acts more than attempts: even "a treasonable conversation" may be punished as severely as an act of treason. (46) Treason is then a punishable act, and even its mere contemplation may arouse the resentment of the impartial spectator. Nevertheless, the traitor is not necessarily the sentinel, and a traitor to a tyrannical regime may be deserving of praise instead of condemnation. In any event, I wish only to raise the possible concern about Smith's theory depending more generally upon how persuaded we are by the need to accommodate mala prohibita crimes, as it is at least unclear how they might fit.

These two possible concerns are related. Often mala prohibita crimes are understood as self-regarding and victimless. They are also controversial and we may believe it best to criminalize more narrowly. (47) Perhaps these two worries are strengths of his account. Nevertheless, I raise these two matters as possible concerns. Whether they are concerns may be more a matter of which theory of law one finds most appealing and neither are necessarily a deep problem for Smith.


Conclusion. My primary aim has been to challenge the commonplace understanding of Smith's theory of punishment. Most understand this theory to be retributivist and there is certainly a retributivist element to Smith's theory. However, this common understanding has neglected the multidimensional character of Smith's theory, and I have argued that it is not merely retributivist, but more than this. Instead, Smith offers us a hybrid account that brings together retributivist, deterrent, and rehabilitative elements. I have further argued that this new hybrid understanding of Smith's theory of punishment is helpful when grappling with passages, such as the sentinel case, which might trouble retributivist accounts. I then concluded with a brief examination of two possible concerns, although I argued neither clearly threatens Smith's account.

Whatever its possible limitations, Smith's theory of punishment is more complex than has often been thought, and it has a clearly multidimensional character. My hope is that the interpretation offered here will encourage others to shed further new light on Smith's sophisticated account. (48)

(1) James Fitzjames Stephen, A History of the Criminal Law of England (London, 1883): 81-2.

(2) For example, see Chad Flanders, "Retribution and Reform," Maryland Law Review 70 (2010): 128-32, 134; Knud Haakonssen, The Science of a Legislator: The Natural Jurisprudence of David Hume and Adam Smith (Cambridge: Cambridge University Press, 1989), 116; Eric Miller, "'Sympathetic Exchange,' Adam Smith, and Punishment," Ratio Juris 9 (1996): 196; Paul Russell, Freedom and Moral Sentiment: Hume's Way of Naturalizing Responsibility (Oxford: Oxford University Press, 1995), 143-4; and John Salter, "Sympathy with the Poor: Theories of Punishment in Hugo Grotius and Adam Smith," History of Political Thought 20 (1999): 207. Each argues that Smith's theory of punishment should be understood as primarily a retributivist theory of punishment.

(3) See Adam Smith, The Theory of Moral Sentiments, ed. Knud Haakonssen (Cambridge: Cambridge University Press, 2002):, 3.1.2. Citations to this work are to the section, chapter, and section numbers.

(4) "The impartiality standard" is my own terminology and it is not used by Smith, although I believe it satisfactorily captures what is of importance for our purposes concerning normative judgments relating to punishment considerations.

(5) By we, I refer to all of us, as we should all occupy the standpoint of the impartial spectator.

(6) I simply note that Smith appears to run two distinctions together in this part of The Theory of Moral Sentiments. He speaks only of whether an appropriate intentional disposition is present or absent: I am either responsible (and, hence, a possible candidate for punishment) or not (and innocent). He denies that we may be responsible, but not deserve punishment in certain circumstances.

(7) Smith, The Theory of Moral Sentiments, Herbert Hart notes that we might all become criminals if a bare intention to commit a crime was punishable. See H. L. A. Hart, Punishment and Responsibility: Essays in the Philosophy of Law (Oxford: Oxford University Press, 1968): 127.

(8) Smith, The Theory of Moral Sentiments, 2.3.2.

(9) Smith, The Theory of Moral Sentiments, 2.3.1.

(10) Smith, The Theory of Moral Sentiments,

(11) Smith, The Theory of Moral Sentiments,

(12) J. S. Mill, On Liberty, ed. Elizabeth Rapaport (Indianapolis: Hackett, 1978), 9. While I claim there are echoes of Mill's harm principle that resonate with this part of our discussion of Smith's theory of punishment, I am not arguing that Mill and Smith shared similar theories of punishment: they did not.

(13) Smith, The Theory of Moral Sentiments,,, 6.2.2.

(14) Smith, The Theory of Moral Sentiments,

(15) Smith, The Theory of Moral Sentiments,

(16) Smith, The Theory of Moral Sentiments, See also Adam Smith, Lectures on Jurisprudence, ed. R. L. Meek, D. D. Raphael, and P. G. Stein (Oxford: Clarendon, 1978), 2.90 (citing lecture and section numbers).

(17) Smith, The Theory of Moral Sentiments,

(18) Smith, The Theory of Moral Sentiments, For an opposing view, see John Gardner, Offences and Defences: Selected Essays in the Philosophy of Criminal Law (Oxford: Oxford University Press, 2007), 220.

(19) Smith, The Theory of Moral Sentiments, We even owe victims our tears in some circumstances (see ibid., I believe that Smith is mistaken to believe that everyone might be expected to share the same moral point of view with respect to all judgments of criminality. There may be broad agreement or conventional morality, but there will always be reasonable pluralism regarding comprehensive doctrines. For example, see Hart, Punishment and Responsibility, 171: "It is sociologically very naive to think that there is even in England a single homogeneous social morality whose mouthpiece the judge can be in fixing sentence ... Our society, whether we like it or not, is morally a plural society; and the judgments of the relative seriousness of different crimes vary within it far more than this simple theory recognizes." See also John Rawls, Political Liberalism (New York: Columbia University Press, 1996) and Thom Brooks and Martha C. Nussbaum, eds., Rawls's Political Liberalism (New York: Columbia University Press, 2013).

(20) Smith, The Theory of Moral Sentiments,

(21) Smith, The Theory of Moral Sentiments,

(22) Smith, The Theory of Moral Sentiments,

(23) Smith, The Theory of Moral Sentiments,

(24) This raises interesting questions about individual cases where the criminal is unrepentant and the use of modest shame punishment within specific safeguards may help instill a sense of justified guilt.

(25) Smith, The Theory of Moral Sentiments,

(26) For example, Miller argues that punishment is part of a sympathetic exchange where the punishment is proportionate to the victim's justified resentment and justified by moderating victim resentment through the position of the impartial spectator: this sympathetic exchange is a process of exchanging sympathies between victim and offender in agreement with the judgment of the impartial spectator. While a highly insightful analysis in many respects, this account focuses too narrowly on punishment as proportionate exchange and omits the additional functions that Smith attributes too punishment. See Miller, "'Sympathetic Exchange.'"

(27) See Smith, The Theory of Moral Sentiments,

(28) Smith, The Theory of Moral Sentiments,

(29) Smith, The Theory of Moral Sentiments,

(30) Smith, The Theory of Moral Sentiments, On hybrid and unified theories of punishment, see Thom Brooks, Punishment (New York: Routledge, 2012).

(31) See also R. A. Duff, Trials and Punishments (Cambridge: Cambridge University Press, 1986), 20.

(32) In contrast, it may be helpful to highlight the following passage on these three elements of punishment offered by Hegel: "Punishment, for example, has various determinations: it is retribution, a deterrent example as well, a threat used by the law as a deterrent, and also it brings the criminal to his senses and reforms him. Each of these different determinations has been considered the ground of punishment, because each is an essential determination ... But the one which is taken as ground is still not the whole punishment itself." G. W. F. Hegel, Science of Logic, trans. A. V. Miller (Amherst: Humanity Books, 1999), 465. For Hegel, retribution may serve as the ground of punishment, but punishment is more than mere retribution.

(33) Smith, Lectures on Jurisprudence, 2.92. It is worth noting similar considerations regarding engine-drivers asleep on the job. See T. H. Green, Lectures on the Principles of Political Obligation and Other Writings, ed. Paul Harris and John Morrow (Cambridge: Cambridge University Press, 1986): [section] 198.

(34) On murderers, see Smith, Lectures on Jurisprudence, 2.94.

(35) Smith, Lectures on Jurisprudence, 2.92.

(36) Smith, Lectures on Jurisprudence, 2.91.

(37) Smith, The Theory of Moral Sentiments,

(38) For example, see Smith, Lectures on Jurisprudence, 2.176-7.

(39) My position differs from D. D. Raphael who argues that Smith's discussion of the sentinel is inconsistent with the rest of Smith's account of justice. See D. D. Raphael, "Hume and Smith on Justice and Utility," Proceedings of the Aristotelian Society 73 (1972/73): 87-103, at 96-7.

(40) For example, see Smith, Lectures on Jurisprudence, 1.33-5 and 2.95 104.

(41) For example, see Smith, The Theory of Moral Sentiments,,, 6.2.2.

(42) I note that much of what Smith has to say about punishment is found not only in his Lectures on Jurisprudence but in The Theory of Moral Sentiments also, where the main worry is not law but an analysis of the moral sentiments of merit and demerit with examples from criminal law. See D. D. Raphael, The Impartial Spectator: Adam Smith's Moral Philosophy (Oxford: Oxford University Press, 2007), 74.

(43) Smith, The Theory of Moral Sentiments,

(44) Smith, The Theory of Moral Sentiments, Of course, Smith recognizes the irregularity of our sentiments and the important role that actual consequences can play in our sentiments about others and their conduct. This link between actions and consequences must not lead us to overlook the fact that punishment, for Smith, is meant to serve a larger role than communicating that an offender may deserve punishment (which may appear retributivist), but other additional functions that punishment should play including rehabilitation amongst others. My thanks to Samuel Fleischacker.

(45) In highlighting this possible concern, I cannot hope to resolve the long standing debate between those who endorse the separability thesis. On this thesis, see Matthew H. Kramer, In Defense of Legal Positivism: Law Without Trimmings (Oxford: Oxford University Press, 1999).

(46) See Smith, The Theory of Moral Sentiments, 6.2.22,

(47) See Douglas Husak, Drugs and Rights (Cambridge: Cambridge University Press, 1992).

(48) I am most grateful to Fonna Forman-Barzilai, Christel Fricke, Sam Fleischacker, Duncan Kelly, Raino Malnes, and Eric Miller for insightful comments and helpful discussions on earlier drafts.
COPYRIGHT 2012 Philosophy Education Society, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Brooks, Thom
Publication:The Review of Metaphysics
Article Type:Essay
Geographic Code:4EUUK
Date:Dec 1, 2012
Previous Article:Rethinking honor with Aristotle and Confucius.
Next Article:Locating the lost island.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters