Stealing Guilt: Freud, Twain, Augustine and the Question of Moral Luck

by Derek Stanovsky

Originally published in American Imago, vol. 63, no. 4: 445-461, Winter 2006. Copyright © 2006 The Johns Hopkins Univeersity Press. Used by permission.

This paper re-examines the philosophical debate surrounding the issue of moral luck through the lens of Freudian psychoanalytic theory. The author argues that Freud's writings on moral luck, which have not previously been discussed in this context, provide not only a cogent explanation for the reasons behind the existence of moral luck but also a compelling argument for the constitutive role played by moral luck in the formation of moral agency and moral identity. Freud's own example of a story by Mark Twain, "The First Melon I Ever Stole," is explored in this context and compared to other more typical examples of moral luck as is another example of moral luck having to do with the theft of pears drawn from the writings of St. Augustine.


Moral luck is the strange phenomenon by which twists of fate sometimes work to change or amplify our moral judgments of actions taken by ourselves or others. When things turn out well, we tend to be lenient towards even grave transgressions. When things go wrong, however, and tragedy strikes, guilt and blame may know no bounds and may attach themselves to even our smallest mistakes. It is this power of good and bad luck to alter and magnify moral blame in cases where the degree of responsibility remains a constant that has intrigued and puzzled ethicists in recent years.

The term "moral luck" was coined by Bernard Williams and given currency in an initial exchange with Thomas Nagel on the subject. Although Williams and Nagel differ over exactly what conclusions ought to be drawn from occurrences of moral luck, both philosophers see the phenomenon as a problem for moral theory generally. The issue is how blame or praise can be apportioned when luck is involved and individuals are not believed to be primarily responsible for their actions and their outcomes. It seems wrong to hold someone accountable for something truly beyond his control, yet once the existence of moral luck is acknowledged, it becomes difficult to find areas of life immune to the vagaries of fortune such that moral judgment is still possible and appropriate.

Thus, when moral luck is discussed, it is most often raised as a difficulty for a particular ethical theory, or for ethics generally. The particular defenses that have been mounted by philosophers in response to the challenge posed by moral luck have been extremely diverse, but there is a shared recognition of the problem, and the project has usually been to formulate ways in which ethical theories might surmount, or at least circumnavigate, the difficulties raised by the occurrence of moral luck. [End Page 445]

In this paper, I wish to call attention to one theory that explicitly raises the issue of moral luck not as a problem but as a support for its own claims regarding ethics and ethical theory. This theory is Freudian psychoanalysis. Freudian theory has yet to be mined for insights into the issues raised by moral luck, even though Freud is one of the few authors to have recognized and addressed the issue directly in his writings. What distinguishes Freud's treatment of moral luck from more recent discussions is that he approaches it as something other than an interesting stumbling block on the path to an otherwise rational and coherent system of normative ethics. For Freud, it is the psychological phenomenon of moral luck that is of primary interest, and questions of normative ethics arise only as a secondary concern. While this shift in emphasis may not initially seem very promising for ethicists, there are substantial insights to be gained from Freud's analysis of moral luck both from the standpoint of moral psychology as well as from that of normative ethics.

In the conclusion to his article responding to Bernard Williams, Thomas Nagel writes: "The problem of moral luck cannot be understood without an account of the internal conception of agency and its special connection with the moral attitudes." He then concedes: "I do not have such an account" (1993, 69). If one were to propose a one-line definition of psychoanalysis, one could do much worse than describe it as "an account of the internal conception of agency and its special connection with the moral attitudes." What Freud's analysis offers is an account of why moral luck occurs, its connection with the moral sentiments generally, and its role in the production of moral agency and moral identity. In short, Freudian theory can provide the link between moral luck and moral agency that Nagel notes is lacking. Such a link is necessary if one is to address adequately the issues for normative ethics raised by these questions from moral psychology, and escape the pitfall of attempting to detach the rational moral agent from every unintentional outcome of his or her actions, a project that may simultaneously run the risk of destroying the very moral identity of the agent it seeks to explain.

There are three main points I will argue in this paper. First, Freud's account of moral luck differs substantially from any other [End Page 446] discussion of this topic in the philosophical literature and, for this reason alone, is worth closer consideration. Second, psychoanalysis offers a non-trivial explanation of the reason moral luck occurs without having to minimize either its extent or its importance within moral life. Third, Freud's analysis illuminates the link between moral luck and moral agency. This is an argument with practical implications for normative ethical theories, many of which attempt to expunge moral luck entirely. Drawing on Freud's account, I shall argue that moral luck is more likely to be a constitutive component of moral character and moral agency than a merely contingent one, and that psychoanalysis can provide a cogent explanation of this connection between moral luck and moral agency.

Ethics and Its Discontents

Freud's most detailed and sustained analysis of morality, as well as his most famous and influential, can be found in Civilization and Its Discontents (1930). Among the many facets of moral life that Freud addresses in this work, there appears the following passage:

The field of ethics, which is so full of problems, presents us with another fact: namely that ill-luck - that is, external frustration - so greatly enhances the power of the conscience in the superego. As long as things go well with a man, his conscience is lenient and lets the ego do all sorts of things; but when misfortune befalls him, he searches his soul, acknowledges his sinfulness, heightens the demands of his conscience, imposes abstinences on himself and punishes himself with penances. (126)
Freud here explicitly addresses the phenomenon recently denominated "moral luck" by Williams. After beginning with an observation with which most ethicists would wholeheartedly agree, namely, that "the field of ethics . . . is so full of problems," he then goes on to discuss the "fact" of moral luck.

It is noteworthy that what Freud takes to be the problem is not how the pangs of guilt that so often accompany instances [End Page 447] of (bad) moral luck can be reconciled with the demands of a coherent moral theory. Instead, it is the phenomenon itself that interests him. That is, what Freud believes to be in need of explanation is why feelings of guilt arise in the first place, not whether or not these feelings are morally justified. Freud takes the existence of moral luck as a given, and from this starting point he deduces some interesting conclusions about the nature of morality. This is the exact opposite of the way many ethicists have proceeded. Often an ethical theory of one stripe or another is assumed (either explicitly or implicitly), and from this starting point conclusions are drawn about the nature of moral luck. This difference in approach is striking and significant for the kinds of conclusions one is able to draw from cases involving moral luck.

The specific kind of "ill-luck" Freud considers is that arising from "external frustration" or, as it is more often named in the current debates, "extrinsic luck" (Williams 1993a, 41). Extrinsic luck concerns factors outside of the agent's control that affect the outcome of his projects, such as natural disasters, accidents of all types, and other unforeseeable events. This is opposed to "intrinsic" moral luck, which deals with factors beyond the agent's control but that nonetheless contribute to the agent's character and abilities. These include such things as inclinations and talents. As Daniel Statman (1993, 27) has noted, there are gray areas where it is difficult to distinguish extrinsic from intrinsic factors. For instance, are accidents of birth that affect the character of an individual cases of extrinsic or intrinsic luck? In an effort to clarify these points, Nagel (1993) further subdivides these categories into the four cases: constitutive, circumstantial, causal, and resultant luck. Constitutive luck includes a person's "inclinations, capacities, and temperament," circumstantial luck concerns "the kind of problems and situations one faces," whereas causal and resultant luck refer to those factors beyond the agent's control that affect either what actions are taken or the outcome of those actions (60). Martha Nussbaum (1986) and Claudia Card (1996) have each explored issues surrounding constitutive and circumstantial moral luck in their works. However, most discussions have centered on cases involving causal and resultant luck. [End Page 448]

Freud's discussion of "external frustration" also fits most easily under the heading of resultant luck: the way our actions and plans actually turn out, as opposed to the way we intend them to turn out. This is the kind of luck Kant considers as well in the famous opening passages from the Grounding for the Metaphysics of Morals (1785), where he writes of the jewel-like quality of a good will apart from any results it may achieve in the world. This locus classicus in Kant is the starting point for Nagel's reflections on moral luck. In general, cases of resultant luck tend to be the clearest and least ambiguous examples in the literature on moral luck.

Williams's own example (1993a) of a truck driver who, through no fault of his own, runs over a child can be taken as an illustration of an external frustration that leads to bad results. In this case, the driver's intentions are only to get from here to there, and his actions are consistent with this goal, yet despite his lack of malice, and absent even any negligence on his part, his actions bring about the tragic death of a child. Here the external event of a child running in front of the truck drastically alters the outcome of the driver's actions. Nagel (1993, 61) argues that this is simply a case of plain old bad luck and not yet a case of specifically bad moral luck. This is so because the driver is entirely without fault and thus has no reason to reproach himself. The driver may in fact feel badly, but he bears no moral blame for the death. For Nagel, it would only be a case of bad moral luck if the driver had been in some way negligent, even to a small degree. If this were true, then the driver would reproach himself much more harshly for this negligence in the case where a child had been killed than in the case where the drive had been uneventful. The actions would be the same, but the feelings about those actions would be incommensurable. The reason for this different, and harsher, evaluation of his actions would be due to bad moral luck, that is, the chance event that a child had darted into the road.

Williams, however, points out, contra Nagel, that such an analysis presupposes a great deal: "what is the point of insisting that a certain reaction or attitude is or is not a moral one? What is it that this category is supposed to deliver? . . . I still cannot see what comfort it is supposed to give me, or what instruction it offers to other people, if I am shunned, hated, unloved, and [End Page 449] despised, not least by myself, but am told that these reactions are at any rate not moral" (1993b, 254). The fact that a truck driver is likely to feel guilt and reproach himself for the death of the child even if he was not negligent in the least is something that needs explanation. To say that this instance of guilt is not moral, but that the same feelings in different circumstances would be moral, simply begs the question. Why are people prone to such feelings to begin with? If they are unwarranted, then why do they occur? It is here that Freud's analysis of moral luck can shed some needed light.

The Anatomy of Guilt

Freud's analysis begins with the guilt generated by situations of bad moral luck, like that of the truck driver, and works from there to uncover the source of those feelings. Briefly, the explanation he puts forward is the following:

Fate is regarded as a substitute for the parental agency. If a man is unfortunate it means that he is no longer loved by this highest power; and, threatened by such loss of love, he once more bows to the parental representative in his superego - a representative whom, in his days of good fortune, he was ready to neglect. (1930, 126-27)
For Freud, situations involving bad luck simply provide the superego with opportunities to vent its pent-up aggression on the ego. The superego is in a position of practical omniscience with regard to the agent's actions and desires, and so knows of his many forbidden wishes and repressed desires. Bad luck is construed as a punishment and as a sign of disapproval and loss of the parents' love. This situation results in feelings of guilt out of proportion with either the offense or the agent's degree of responsibility. This is only to be expected since, for Freud, the guilt is not over the event, it is over all those other things for which one ought to have been punished but escaped. Ultimately, it is over fear of the loss of love by one's parents for all the unworthy things one has done or imagined. Happenings in the world for which one is not responsible can, legitimately, [End Page 450] be a matter of indifference. But disapproval by one's parents, regardless of the reasons, is inherently something of enormous concern, capable of inciting strong emotions. In the case of bad moral luck, events in the world become linked with that fear of being unloved in a direct and powerful way. The bad luck becomes a sign of that loss of love; it becomes itself the punishment for one's transgressions, and the resulting guilt becomes all too easy to explain.

Now, the particular instance of the ill-luck of the driver killing the child may not initially seem to be a very good example of the kind of ill-luck Freud is discussing. After all, it is the child who has the worst luck, not the driver. In what way is the death of the child to be construed as punishment by Fate directed at the driver? Better examples would perhaps be ones where the misfortune befalls the agent and not some third party. There is no insurmountable difficulty in using the truck driver example even here, however. All that needs to be present is a tragic event for which the agent bears no real moral responsibility and yet over which he suffers real guilt. The truck driver incident could very well fulfill these requirements, and we can easily imagine such a driver reproaching himself with harsh recriminations, swearing never to drive again, and suffering severely from guilt and pangs of conscience.

Different drivers, of course, may respond differently to such tragedies. Feelings of guilt are not something that can be predicted simply from the events themselves. For psychoanalysis, certainly, the relations between events in the external world and events inside the psyche are never simple or straightforward. However, the fact that individual experiences of guilt may vary widely even when their moral culpability remains the same once again suggests that something other than moral judgment may lie behind these feelings. The response described here, of self-reproach and guilt, may sound extreme (and may in fact be extreme), but it certainly falls within the range of recognizable human reactions to such a tragedy. The peculiarity that such responses are not morally warranted by the situation is precisely what stands in need of explanation. However, there are ways to strengthen this example still further.

Consider what happens if we assume that the child killed in the accident was the truck driver's own. This reverse oedipal [End Page 451] twist creates more interest for the psychoanalyst. That the driver might feel overwhelming guilt, and that the source of those feelings might be sought in the unconscious death wishes that haunt even the most loving parent-child relationship, would not be hard to infer. In this case, though, the moral responsibility remains exactly the same and so would not alter any of the discussions above, but the guilt generated by the incident would be multiplied many times over. All that has been changed is that a case of bad moral luck has been transformed into one of very bad moral luck.

The point in fabricating this example is simply to demonstrate that moral responsibility does not provide an exhaustive explanation for feelings of guilt. In accidentally killing one's own child, the presence of guilt and the suffering this would entail are readily apparent. It is also perhaps easier to see how the agent might construe the incident as a punishment directed at himself for his own faults and shortcomings and reproach himself accordingly. It is interesting that none of the examples in the literature on moral luck has been elaborated in this way. It is typically assumed that the reason one feels guilty is for moral reasons, and the hunt is then on for various permutations to change the level of moral responsibility involved. But why assume this? All we know for sure is that in these kinds of situations people often do feel guilt. Why not elaborate them in such a way so as to increase the feelings of guilt? From this we might more readily see what there is about the situation that incites these feelings and what is really operating at the root of instances of so-called moral luck. Williams was right to note that it is not comforting to tell a person that some feelings of guilt are not moral while otherwise identical feelings in slightly different circumstances are moral. What this highlights is simply that morality and guilt have a less straightforward relationship than many ethicists would like to admit, a state of affairs that psychoanalysis has no trouble accommodating.

In a footnote immediately following the passage from Civilization and Its Discontents where Freud first broaches the subject of moral luck, he discusses just such an example where the level of moral culpability remains constant, yet the level of guilt rises dramatically: [End Page 452]

This enhancing of morality as a consequence of ill-luck has been illustrated by Mark Twain in a delightful little story, The First Melon I Ever Stole. This first melon happened to be unripe. I heard Mark Twain tell the story himself in one of his public readings. After he had given out the title, he stopped and asked himself as though he was in doubt: "Was it the first?" With this, everything had been said.1 (126)
The doubt introduced by Twain when he asks whether it was really the first melon he ever stole underscores that it is only the fact that things went wrong that made this theft memorable and an occasion for self-reproach. All the other ripe, sweet, and juicy melons he had previously stolen were erased from memory. However, Twain's newly awakened conscience does not trouble him overmuch. Instead of feeling guilt over his theft or seeking to provide restitution, Twain becomes angry with the unscrupulous farmer who was trying to sell that sour melon to an unsuspecting customer.

For Freud, Twain's satirical morality tale highlights that pangs of conscience are often felt most strongly when things go wrong. Although the theft of an unripe melon occasions only a cursory reflection on the propriety of stealing, the sour taste nonetheless serves as a proxy for the absent parental chastisement. The oedipal sources of Twain's youthful guilt and need for punishment are not difficult to spot. In this connection, one might recall all the other masochistic boys in Twain's semi-autobiographical fiction, such as Tom Sawyer, who habitually misbehave to gain the attention they crave from the ones they love, even if that attention brings with it a beating (Avalos 2005, 50-52).

According to Freud, what induces guilt, illness, and neurosis is a frustration of one sort or another. In Twain's story, the frustration is an external one in the form of the sour melon itself. However, in a dialectical twist, Freud (1916) also explores cases where individuals are "wrecked by success" and fall ill when their dreams come true and their repressed wishes are fulfilled. Freud resolves this apparent contradiction by noting that the unconscious is happy to spare the individual the internal sufferings of guilt so long as there is sufficient suffering imposed [End Page 453] externally through being deprived of his wish. Once the wish is granted, though, the internal sense of guilt may be unleashed by the superego as a compensating punishment directed at the always unworthy ego. Thus, the punishment is constant and unrelenting. Either one is deprived of what he truly desires or one is punished by feelings of guilt if he should succeed in attaining his goal. In either case, there will be suffering.

However, it remains true that stealing a melon, finding it delicious, and suffering no bad consequences seems far less likely to give rise to guilt than when the stolen fruit is defective in some way. The pleasure from the ill-gotten gain may even tend to counteract and overshadow any pangs of guilt that do manage to slip in. Consider in this connection the case of St. Augustine and his famous theft of the pears. In Book 2 of the Confessions, Augustine laments an occasion in his adolescence when he and some of his friends stole pears out of a neighbor's tree. They did not need the pears and did not even really want them, eventually throwing most of their booty to the pigs. For Augustine, it is the pointlessness of his theft that makes it an act of evil. Judging from his later writings, the guilt over this episode stayed with him his whole life. However, the suspicion arises that the power and longevity of his guilt could not have been due simply to the theft of worthless pears. During puberty, Augustine had been sternly warned by his mother against adultery. As Peter Rudnytsky observes: "The disproportion between the triviality of Augustine's offense and the extreme importance he imputes to it may thus be explained psychoanalytically as a displacement of the sense of guilt properly attaching to the former events [his mother's warnings] onto the latter [his theft of the pears]" (1994, 140).2

Thus, the source of the guilt may not lie in the obvious transgression. Instead, the theft becomes the occasion for inflicting already existing oedipal guilt onto the unsuspecting ego. Freud writes: "Paradoxical as it may sound, I must maintain that the sense of guilt was present before the misdeed, that it did not arise from it, but conversely - the misdeed arose from the sense of guilt. These people might justly be described as criminals from a sense of guilt" (1916, 332). That is, the theft may have been caused by pre-existing feelings of guilt and used as an outlet for those feelings, rather than being itself the cause of the guilt. [End Page 454]

Augustine also notes, however, that the pears he stole in his youth and over which he suffered such guilt were "attractive in neither color nor taste" (1991, 28?29). Here we may find a small opening for moral luck to intrude into the story. Had the pears been sweeter, one wonders whether Augustine's conscience would have been more lenient. Like Twain after him, it may have been the bad luck of a particular outcome, the stealing of such unappealing fruit, that occasioned his reflections on morality and opened the floodgates of his guilt. Had the theft served any useful function, it might have been possible to justify it, or at least to enjoy it. If these speculations are correct at all, the real reason for the unlikely theft was to provide an opportunity for self-punishment. Unlike Twain, who vents his newfound moral outrage on the turpitude of the farmer who was trying to sell the green melon, Augustine reserves his opprobrium for himself.

The stories of Twain and Augustine both make interesting examples of cases of moral luck, in many ways better than the more frequently used examples. They do so because they highlight the agent-specific nature of moral luck. To a third party, it makes very little difference whether the melon or the pear was sweet or not in judging the morality of the action. Stealing fruit is simply wrong. But to the agent, and for the agent's feelings towards those actions, the taste of the fruit might be extremely relevant. It is only the intervention of an external event, the bad luck of stealing an unripe piece of fruit, that introduces a perceived punishment. What Freud is able to explain is why instances of bad moral luck can exacerbate feelings of guilt. What is at work is not morality. What is at work is the unconscious. These stories are the flip side of cases that are more typically considered in discussions of moral luck, such as the truck driver story. Instead of a person with no responsibility who suffers pangs of guilt when misfortune occurs, in these instances a person who bears full moral responsibility feels guilt, or feels it more intensely, when bad luck intervenes. Freud's account is able to explain both sorts of cases.

The second case may not seem to merit attention because it does not directly cast doubt on ethical theory as such, but only on the agent's own morals. The usual view is that a person ought to feel guilt over theft, and the lack of such feelings is taken to show a lack of moral character, not to provide evidence that in [End Page 455] some instances stealing is morally acceptable. For this reason, such cases are usually of little interest to ethicists. However, they do need to be considered if one wants to explain the origin of guilt, not to justify a particular ethical theory. If guilt both occurs and fails to occur when an agent both is and is not responsible, then it may be that guilt and morality have a very different connection from the one that is often supposed. If cases of the first type throw suspicion on the adequacy of ethical theory, as both Williams and Nagel have argued, then perhaps cases of the second type should be examined as well. Freud's account of moral luck has the merit of taking seriously both sorts of situations and not artificially truncating his range of examples. It now remains to be seen what difference such an account of moral luck might make for ethical theory generally.

Character, Culture, and Moral Agency

Despite their other differences, Williams and Nagel agree that moral luck influences our conception of what it means to be a moral agent. Nagel writes:

it is not enough to say merely that our basic moral attitudes toward ourselves and others are determined by what is actual; for they are also threatened by the sources of that actuality, and by the external view of action which forces itself on us when we see how everything we do belongs to a world that we have not created. (1993, 69)
Here, Nagel concludes that moral luck must play a role in any adequate account of moral agency since such luck is endemic to every moral undertaking. Williams makes a similar point, arguing further that conceptions of moral agency that do not take account of moral luck and that attempt to exclude and eliminate all traces of the accidental and unintentional will do fatal damage to their own conceptions of moral agency.

In discussing the appropriateness of feelings of guilt in an instance where the agent bears no moral responsibility for an unfortunate outcome affecting another person, Williams writes: [End Page 456]

it would be a kind of insanity never to experience sentiments of this kind towards anyone, and it would be an insane concept of rationality which insisted that a rational person never would. To insist on such a conception of rationality, moreover, would, apart from other kinds of absurdity, suggest a large falsehood: that we might, if we conducted ourselves clear-headedly enough, entirely detach ourselves from the unintentional aspects of our actions . . . and yet still retain our identity and character as agents. (1993a, 44)
It is this identity of the moral agent that is at stake in the issue of moral luck. The question becomes how constitutive experiences of moral luck are to the moral character of the agent, and whether or not the experiences of guilt over situations involving bad moral luck can be eliminated without fundamentally altering the identity of that moral agent.

Consider the truck driver example again. Could a driver who suffered no guilt over the accidental death of a child ever be the moral equivalent of one who did display such guilt? Is the presence or absence of feelings of guilt truly negligible to our account of the moral character of the agent? One could plausibly argue that feelings of guilt under such circumstances are so normal and unexceptional that the complete lack of such feelings would be sufficient grounds to doubt the moral character of any agent who did not possess them. It is at least difficult to see how the argument could be made that feelings of guilt not only are not required but are, in fact, completely irrelevant to the moral identity of the agent. To claim that an agent should not feel guilt in this situation, or that these feelings are morally irrelevant, or even supererogatory, is to come close to the "insane concept of rationality" against which Williams warns. It entails an almost inhuman level of detachment "from the unintentional aspects of our actions." In this light, the feelings of guilt experienced by the driver, no matter how seemingly irrational, are unlikely to be superfluous to his moral identity, and are more likely to be an important constitutive part of that identity. In the case of Augustine, we have an agent who feels what many would regard as a morally appropriate guilt over his theft, but taken to an extreme degree. However, this [End Page 457] excessive guilt over an otherwise minor transgression is clearly formative of Augustine's character. A St. Augustine who did not feel such guilt, for so long, or with such intensity would not be the same imposing figure we all know. He would be a substantially different moral agent with a substantially different identity. Drawing on Freud's analysis of Twain's anecdote, I argued above that it was possible that a similar "enhancing of morality as a consequence of ill-luck" may also have taken place with regard to Augustine's stealing of the unappealing pears. If such an accident could have exacerbated the guilt over the event, even to a small degree, then we have the possibility that increased guilt due to bad moral luck might be something that helps to create the moral character of the agent. In that case, moral luck would play a role in constituting exactly what is extraordinary about the moral character of St. Augustine.

A similar argument could be made about the extraordinary character met with in the satirical self-portrait by Mark Twain. It is the awakening of an otherwise dormant conscience by the bad luck of an unripe melon that constitutes what is unique, and uniquely funny, about this otherwise amoral agent. For Twain, this instance of bad moral luck may be constitutive of a curmudgeon rather than a saint, but it is still fundamental to the moral character and identity of that agent. To eliminate the role of moral luck in this example would transform the identity of the agent from a hapless moral bungler to a much less interesting cold-blooded sociopath who simply never feels guilt over his thefts, even those that go awry.

If what is remarkable about Augustine is his excess of guilt, then what is remarkable about Twain is his deficiency of guilt except under adverse conditions. In either case, it can be argued that even the small amount of bad moral luck present in these examples works to help constitute the moral identities of these otherwise very different agents. In the truck driver example, too, it seems implausible to argue that a driver who does not experience guilt over the death of the child is the moral equivalent of a driver who does experience such guilt. In each of these situations, the guilt generated by bad moral luck is not clearly ancillary to the moral identity of the agent but seems more likely to be constitutive of that identity. To attempt to expunge all traces of moral luck from these examples would be to erase [End Page 458] much of the moral character of each of these agents. It certainly would not leave the moral identities untouched.

Given these examples, it seems justified to conclude along with Williams and Nagel that moral luck plays an important and constitutive role in the formation of moral character and identity. In each of the situations discussed above, morality is enhanced by an element of moral luck in such a way that the very character of the moral agent becomes bound up with his reactions to these external frustrations. This is precisely the point being asserted by both Williams and Nagel in their initial exchange on the topic of moral luck.

What is new here is that psychoanalytic theory can supply a reason for this entangling of luck and morality, something that few other theories have been able to offer. For Freud, this connection between luck and morality comes about because morality itself arises out of the sublimation of human aggression, an aggression that always requires new outlets even if it means the agent himself or herself must become the target for his or her own aggression. In this way, unfortunate accidents are seized upon as occasions to justify the venting of pent-up aggression on oneself in the form of guilt. These reproaches do not seem justified on any clear moral grounds, yet they occur and recur with monotonous regularity. The incorporation of moral luck as a predictable and normal event to which all moral agents are prone is one of the principal advantages psychoanalysis has to recommend itself in this philosophical debate. That it also gives rise to other problems within ethical theory does not obviate this benefit. Thus, Freud can account for the constitutive nature of moral luck in the moral identities of Augustine, Twain, and the truck driver without permitting it to violate any requirements concerning the rationality necessary for moral agency. In fact, far from violating the prerequisites for moral agency, psychoanalysis lends credence to the view that moral luck is itself one of the mechanisms through which moral agency and identity are produced.

In his conclusion to Civilization and Its Discontents (1930), Freud notes that the ethical demands made by society on the individual seem to have little regard for "whether it is possible for people to obey" them (143). For Freud, unlike Kant, "ought" in no way implies "can." It is this imbalance between the ethical [End Page 459] demands and psychological realities faced by individuals that leads Freud to his less than optimistic conclusions about the "discontents" inherent in human life. It may be that instances of moral luck are best understood as yet another place where ethics makes demands that individuals cannot carry out. The injunction to dissociate oneself from the unintentional aspects of one's actions does not imply that can be done. Here the problem is not simply that the demand is very difficult, but that it may be impossible to fulfill and still retain one's moral identity. Highlighting this conflict and giving one possible explanation of its origin is one of the most valuable insights psychoanalysis has to offer ethicists. A perfectly rational and coherent system of ethics may be unattainable for the less than perfectly rational and coherent agents that are human beings. Viewed in this way, moral life may not be purely rational, but neither is it inexplicable.


1. Two versions of this speech by Twain have been collected and published under the title "Morals Lecture" (1968; 1976).

2. In a further twist, Rudnytsky argues that the warnings of Augustine's mother against adultery are themselves a displacement of the unconscious incestuous dynamics in their relationship.


Augustine. 1991. Confessions. Trans. Henry Chadwick. New York: Oxford University Press.
Avalos, Julio C. 2005. "An Agony of Pleasurable Suffering": Masochism and Maternal Deprivation in Mark Twain. American Imago, 62:35-58.
Card, Claudia. 1996. The Unnatural Lottery: Character and Moral Luck. Philadelphia: Temple University Press.
Freud, Sigmund. 1916. Some Character-Types Met with in Psychoanalytic Work. S.E., 14:309-33.
-----. 1930. Civilization and Its Discontents. S.E., 21: 57-145.
Kant, Immanuel. 1785. Grounding for the Metaphysics of Morals. Trans. James W. Ellington. Indianapolis: Hackett, 1981.
Nagel, Thomas. 1993. Moral Luck. In Statman 1993, pp. 57-71.
Nussbaum, Martha. 1986. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Cambridge: Cambridge University Press. [End Page 460]
Rudnytsky, Peter L. 1994. Freud and Augustine. In Peter L. Rudnytsky and Ellen Handler Spitz, eds., Freud and Forbidden Knowledge. New York: New York University Press, pp. 128-52.
Statman, Daniel, ed. 1993. Moral Luck. Albany: State University of New York Press.
Twain, Mark. 1968. Morals Lecture. In The Trouble Begins at Eight, ed. Fred W. Lorch. Ames: Iowa State University Press, pp. 321-32.
-----. 1976. Morals Lecture. In Mark Twain Speaking, ed. Paul Fatout. Iowa City: University of Iowa Press, pp. 279-87.
Williams, Bernard. 1993a. Moral Luck. In Statman 1993, pp. 35-55.
-----. 1993b. Postscript. In Statman 1993, pp. 251-58.