As part of my editing of Responsibility & Luck: A Defense of Praise and Blame, I decided to omit the following discussion of Nicholas Rescher’s “identity solution” from Chapter Two. Basically, his proposal was just too silly — particularly in its implications for personal identity — to be worthy of inclusion in my book. However, it’s not too silly to blog! I thought that it might of interest, particularly as I’ll discuss Chapter Two on Thursday’s Philosophy in Action Radio. Without further ado…

The Identity Solution

The epistemic and equalization solutions are not the only means of denying the dependence of moral desert on luck. Another prominent alternative, most clearly advocated by Nicholas Rescher, is the “identity solution.”[1] On this approach, a person is not judged for his actions and their outcomes but only for his character. Then, the very idea of constitutive moral luck is rejected as logically incoherent on the grounds that a person’s character constitutes part of his identity. As we shall see, this proposed solution to the problem of moral luck is subject to serious objections, particularly for its treatment of constitutive moral luck.

The basic starting point of the identity solution is that moral evaluations of persons should concern only qualities of character.[2] Moral character is “the prime consideration from a moral point of view,” so that “the moral significance of acts lies in their serving as evidence” of character of a particular kind.[3] This limitation on the objects of moral judgment is a response to the problem of moral luck itself. Rescher writes, “it is precisely because both one’s opportunities for morally relevant action and … the actual consequences of one’s acts lie beyond one’s control that they are not determinants of one’s position in the eyes of morality.”[4] The only way to eliminate the effects of luck on judgments of actions and their outcomes is to refrain from those judgments entirely. By doing that, the identity solution promises to eliminate resultant moral luck and circumstantial moral luck in two easy strokes. Let us see how.

To eliminate resultant moral luck, the identity solution ignores the actual outcome of an action as morally irrelevant due to the potential influence of luck on it.[5] Instead, it holds that moral judgments must be based on the character underlying the action that produced the outcome. So the moral objection to a drunk driver is not that he killed a pedestrian, but that he displayed his reckless disregard for the welfare of others in his choice to drive drunk. Thus Rescher writes, “people who drive their cars home from an office party in a thoroughly intoxicated condition, indifferent to the danger to themselves and heedless of the risks they are creating for others, are equally guilty in the eyes of morality (as opposed to legality) whether they kill someone along the way or not.”[6] By ignoring the actual effects of a person’s actions in any moral judgments of him, the problem of resultant moral luck never arises.

The identity solution disposes of circumstantial moral luck by a similar line of argument. A person’s true deserts cannot depend on his chosen actions, as any action may be influenced by circumstances beyond his control.[7] It claims that moral judgments must concern a person’s qualities of character instead; they must identify what a person would do if faced with various opportunities for doing good or evil. Consequently, Rescher states, “from the moral point of view, how people think and how they are decided and determined to act counts every bit as much as what they actually manage to do.”[8] So the person who would have driven drunk if he had not been working that night deserves as much blame as the person who actually drove drunk. However, absent the action of driving drunk, we might not realize that the would-be drunk driver is worthy of that blame. Yet in fact, whether recognized or not, the only difference between the potential and actual drunk drivers is one of “image” or “reputation,” not “moral condition.”[9] So in this approach, the problem of circumstantial moral luck disappears because a person’s deserts depend on what he would do in a variety of circumstances, not just what he actually does in his actual circumstances.

Undoubtedly, circumstantial and resultant moral luck cannot taint moral judgments of a person if he is never judged for his actions and their outcomes. However, we have reason to worry that this solution is ad hoc. If not for these troublesome cases of resultant and circumstantial moral luck, would any moral philosopher advocate limiting moral judgments to character? Likely not, as shown by the fact that ordinary moral judgments of persons routinely concern actions and outcomes. Moreover, to solve that problem by wholly rejecting all such judgments seems like overkill — perhaps with costs too high to bear.

One significant problem with limiting moral judgments to character is that a person who deliberately chooses to act contrary to his well-established dispositions or acts in the absence of any settled dispositions could not be subject to any kind of moral judgment, no matter how noble or depraved the action. So when a habitually cautious driver impulsively decides to drive home one night despite somewhat too much to drink, he cannot be blamed either for his drunk driving or for the ensuing accident, since he does not possess the reckless character of a habitual drunk driver. Any attempt to argue, as a defender of the identity solution would likely do, that the driver should be blamed for the recklessness motivating that particular act reintroduces the problem of moral luck, as such fleeting impulses may be influenced by luck. In this case, perhaps the driver was diverted from his cautious habits by the death of his overbearing father in a freakish vending machine accident that afternoon. Consistency would require the identity solution to give this drunk driver a free pass by refraining from any blame of him for his short-lived reckless drunk driving.

This objection to the identity solution’s approach to moral judgment, while worrisome, is not decisive. The most compelling arguments against this proposed solution concern its approach to claims about luck in character. By its insistence that a person’s moral standing depends solely on his character, the identity solution compresses all moral luck into constitutive moral luck. So its case against moral luck as a whole depends on its arguments against constitutive moral luck. If its distinctive analysis of constitutive moral luck fails, as we will see it does, then the problems of resultant and circumstantial moral luck return in full force.

The identity solution’s core case against constitutive moral luck consists of a denial of the logical coherence of luck in character on the grounds that character constitutes a person’s identity. Rescher’s general principle of luck and identity is that a person “cannot meaningfully be said to be lucky in regard to who [he] is, but only with respect to what happens to [him].”[10] Presumably, that is because a person must exist as someone definite in order to be lucky or unlucky. So it is nonsensical, for example, to ask what kind of person you would be if unlucky enough to be born to starving Somali refugees, Russian peasants, or movie stars. In such cases, you would not exist at all; someone else would. Next, the identity solution claims that a person’s moral psychology — his motives, intentions, commitments, inclinations, virtues, and vices — constitutes part of his identity as a person. Consequently, Rescher observes, “it makes no sense to say things like, ‘Wasn’t it just a matter of luck for X to have been born an honest (trustworthy, etc.) person, and for Y to have been born mendacious (avaricious, etc.)?’”[11] That is because “it is just exactly those dispositions, character traits, and inclinations that constitute these individuals as the people they are.”[12] The identity solution’s argument against constitutive moral luck is quite simple:

Premise 1: Character is an aspect of a person’s identity.
Premise 2: A person’s identity cannot be subject to luck.
Conclusion: Character cannot be subject to luck.

So the very idea of constitutive moral luck is incoherent because it supposes the impossible, namely that a person’s character can be subject to luck.

The identity solution also explicitly denies the relevance of a person’s control (or lack thereof) over his character to moral responsibility for that character on similar grounds.[13] Rescher argues that the attempt to refrain from moral praise for virtues and moral blame for vices based on lack of control “involves a category mistake because the whole control issue is irrelevant here from the angle of moral concern.”[14] Again, that is because “one’s inclinations, disposition, and character … are the crucial part of what constitutes oneself as such.”[15] All that matters from a moral perspective is that a person has a certain virtue or vice. So “the immoralist cannot … plead her natural inclinations and tendencies, and expect her innate cupidity, avarice, or lecherousness, or the like to get her off the moral hook” because “it is exactly her disposition that condemns her.”[16] Merely possessing some virtue or vice is sufficient for moral praise or blame.

This identity-based argument against constitutive moral luck suffers from fatal defects from beginning to end. First, the argument wrongly assumes that all of a person’s moral qualities are essential to his personal identity. Second, the argument ignores the fact that luck shapes a person’s moral development. Third, the argument implies that a person can be morally blamed for any deficiency in his nature, including mental and physical birth defects. Let us consider each objection in turn.

The first objection, in essence, is that the identity solution presupposes that all of a person’s moral qualities are essential to his personal identity in its analysis of constitutive moral luck. That supposition is necessary to the argument: it is why even imagining a person’s moral qualities to be otherwise would be to imagine a whole new person. Rescher implies it in describing “one’s inclinations, dispositions, and character,” whether within one’s control or not, as “a crucial part of what constitutes oneself as such” — without any qualification or distinction between essential and nonessential traits.[17] Even more telling is his objection to the mere idea of luck in character on the grounds that “it makes no sense to envision a prior feature-less precursor [of a person] who then has the good or bad luck to be fitted out with one particular group of character traits rather than another.”[18] While that is certainly true, it is not the model at work in constitutive moral luck. Constitutive moral luck only requires far more modest imaginations, such as, “How would my life be different if I’d been born with more athletic ability, like my sister?” or “What if I were raised to be more even-keeled, like my cousin?” or “Would I trust people more if my mother hadn’t abandoned me when I was twelve years old?” Such questions are perfectly sensible: they do not require imagining a “feature-less precursor” of oneself, nor imagining oneself to be a whole new person. That is because those qualities — and many (if not most) others — are not essential to a person’s identity.[19] So if Joe loses a finger in a boating accident, suddenly develops a taste for scallops, or decides that white lies are morally wrong, he does not become a new person. If all qualities of a person were essential to his identity, then Joe would be a new person in those ordinary cases of personal change. That is absurd.

In fact, the inessential qualities of a person can be (and often are) shaped by luck; they can change based on the influence of forces beyond the person’s control without undermining his personal identity. Constitutive moral luck concerns those kinds of qualities, i.e., a person’s potentially variable character traits, not any qualities essential to his personal identity.[20] Consequently, the identity solution’s appeal to identity does not eliminate the problem of constitutive moral luck, not even partially.[21] It only makes the perfectly ordinary phenomena of personal change impossible to fathom.

The fact that a person’s moral character can change over time sheds light on the identity solution’s simple deductive argument against constitutive moral luck. In Premise 1 (“character is an aspect of a person’s identity”), “identity” refers broadly to all the various qualities of a person, changeable or not. Yet Premise 2 (“a person’s identity cannot be subject to luck”) only makes sense if “identity” refers to the essential aspects of a person’s identity, such that any change would create a new and different person. Hence, the basic argument of the identity solution is fallacious: it depends on an equivocation in the meaning of “identity.”

The second objection is that the identity solution implicitly depends on a highly implausible view of the origin of moral character, namely moral nativism. It ignores the myriad influences of luck in the cultivation of character by presuming (implausibly) that a person’s whole moral psychology is set at birth. Thus Rescher focuses solely on the incoherence of claiming luck in being “born honest” or “born mendacious” without ever considering the far more plausible claim that a person cultivates an honest or mendacious character over the course of his life, partly influenced by luck in his upbringing.[22] That implicit nativism also makes sense of Rescher’s claim that control is wholly irrelevant to character, so that to excuse some vice on the grounds of lack of control “involves a category mistake.”[23] On his view, character is not an aspect of life over which a person might exert control or not, as he does in attending college or bearing children. Instead, character is beyond a person’s control in a deeper sense: when it is created, the person does not exist yet to exert any control over it.

In fact, however, most (if not all) of a person’s moral character is gradually cultivated over the course of his life by his thinking, his choices, and his actions.[24] Clearly, that process of moral development can be influenced by luck. It is a matter of luck, for example, whether the class bully happened to pick on you in sixth grade or not, whether your parents emphasized education or not, and whether your kindly grandmother recovered from her heart attack or not. The identity solution accepts that luck can influence a person’s experiences because that is “what happens to one.”[25] The simple fact is that such experiences shape a person’s moral character, meaning that they affect “who one is,” sometimes profoundly.[26] Yet the identity solution’s whole case against constitutive moral luck depends on the claim that such identity is immune to luck, as if the events in a person’s life never affect his moral character. That is clearly false. Even more strangely, the identity solution would require us to suppose that a person’s character is static from birth to death. After all, if character were essential to a person’s identity, then any change to it, whatever the cause, would make that person into someone new. In sum, the identity solution’s denial of constitutive moral luck based on moral nativism crumbles when faced with the obvious fact that a person’s character changes based on his experiences.

The third objection is that the identity solution’s approach to moral judgments of character suggests that a person could be morally praised and blamed for any aspect of his identity whatsoever. For Rescher, a person need not control his character to be judged for it. Instead, all that matters is that the person possesses some positive or negative trait. That has unwelcome implications. Most obviously, a person might also be subject to moral judgment for a wide range of qualities beyond his control. For example, a person could be morally praised for a high IQ and morally blamed for a low IQ. Even worse, a person could be blamed for a genetic birth defect such as Down syndrome because it makes him a burden to his parents. Clearly, such judgments would be unjust. Yet the identity solution’s approach to moral judgments of character would sanction them.

Ultimately, despite some ingenious twists, the identity solution cannot solve the problem of moral luck. Its limiting of moral judgments to judgments of character in order to solve the problems of resultant and circumstantial moral luck demands a compelling solution to the problem of constitutive moral luck. Yet its proposed solution depends on clearly false assumptions about essential qualities and moral nativism. It also entails absurd views about personal identity and moral responsibility. Consequently, the whole problem of moral luck remains unsolved.[27]


[1]  Rescher 1993. Unfortunately, Rescher’s theory is not particularly clear. He does not adequately explain or justify key claims, and he seems to contradict himself at various points. My explication will attempt to offer the most plausible reconstruction of his views. Notably, many commenters on moral luck classify Rescher’s approach as a form of the epistemic solution. While they often make similar arguments, the overall differences between Rescher and Richards, for example, warrant separate consideration.
[2]  Rescher 1993, p. 156.
[3]  Rescher 1993, p. 157.
[4]  Rescher 1993, p. 156.
[5]  According to Rescher, the outcomes of actions are often relevant to the substance of proper moral principles, in that people should be praised or blamed based on the expected outcome of their actions in the “ordinary course” of events (Rescher 1993, p. 157).
[6]  Later, Rescher explains that the law may consider actual outcomes as “a matter of social policy” (Rescher 1993, p. 159).
[7]  Rescher 1993, p. 158. Rescher distinguishes between a person’s “moral record” and his “moral standing.” The former is determined by his actions, whereas the latter is a function of his character. A person’s moral standing is of overriding importance (Rescher 1993, p. 158).
[8]  Rescher 1993, p. 155.
[9]  Rescher 1993, pp. 154, 157.
[10]  Rescher 1993, p. 156. Rescher views luck in a somewhat different way than Nagel. Instead of identifying luck with a lack of control, Rescher understands luck an accidental event of some value significance to a person (Rescher 1993, p. 145). This difference does not affect the argument.
[11]  Rescher 1993, p. 155.
[12]  Rescher 1993, p. 155.
[13]  For a more detailed (but still wrong) defense of the view that control is not relevant to praise and blame for mental states like emotions, see Adams 1985. The next section considers a more plausible argument for the irrelevance of control to judgments of character with “character-based compatibilism.”
[14]  Rescher 1993, p. 157.
[15]  Rescher 1993, p. 157.
[16]  Rescher 1993, p. 155.
[17]  Rescher 1993, p. 157.
[18] Rescher 1993, p. 155.
[19] Latus 2003, p. 471.
[20] I am doubtful that any moral qualities could be essential to a person’s identity because such qualities must be cultivated as described in Chapter Nine.
[21]  In the attempt to find some aspect of a person wholly immune to luck, Greco (1995, p. 94-5) differentiates between a person’s actual moral worth, based on his actual character, and his essential moral worth, based on whatever his character might have been in various alternative circumstances. However, any such judgments of essential moral worth would be pure guesswork, even about oneself. Consequently, they would provide little if any practical guidance in our dealings with others. Moreover, even Greco admits that essential moral worth would not be wholly immune to luck.
[22]  Rescher 1993, p. 155.
[23]  Rescher 1993, p. 157.
[24]  I would argue that a person’s moral qualities must all be cultivated, even though some basic features of personality might well be innate. The relationship between innate temperament and moral character is discussed in Chapter Nine.
[25]  Rescher 1993, p. 156.
[26] Rescher 1993, p. 156.
[27]  Further criticisms of Rescher’s views can be found in Latus 2003, pp. 470-2 and Latus 2000, pp. 158-60.
Suffusion theme by Sayontan Sinha