Aristotle, Moral Particularism, and the Indeterminacy of Principles

Rex Gilliland, Birmingham-Southern College

In a recent article, Terrence Irwin takes issue with the particularist interpretation of Aristotle’s ethics. Although he acknowledges that there are significant similarities between Aristotle and moral particularism, Irwin argues that Aristotle intends the virtues of character and the doctrine of the mean to provide normative guidance for moral agents as general rules. Irwin’s response to the particularist interpretation of Aristotle raises interesting questions about the nature of Aristotle’s ethics: Are the virtues merely character traits, as they are often understood, or do they guide moral judgment as principles? How might this affect our understanding of the relationship of Aristotle’s ethics to modern moral theories? In addition, Irwin’s article provides an opportunity to reexamine moral particularism and its theoretical commitments. If Irwin is right that the virtues of character and the doctrine of the mean function as moral principles, is this sufficient to undermine the particularist interpretation of Aristotle? Moral particularism is associated with a variety of intuitionism that rejects all general moral principles and portrays proper moral judgment as a pure ‘perceiving’ of particular moral characteristics. But is moral particularism committed to such a view?

In this paper, I agree with Irwin that Aristotle is not a moral particularist, though I find Irwin’s supporting argument to be inadequate. Irwin’s claim that the virtues of character and the doctrine of the mean provide normative guidance appears to be correct. However, I argue that moral particularists, contrary to their explicit assertions, do not reject all general moral principles. If this is true, then Aristotle’s relation to moral particularism is not decided by the presence of general moral principles in Aristotle’s ethics, but instead, I will argue, by the degree of indeterminacy found in these principles. Examining the characteristic of indeterminacy will allow us to redraw the boundary between moral particularism and competing views and to provide an alternative to the model of moral judgment defended by moral intuitionists. By way of conclusion, we will consider how Aristotle’s ethics should be classified in relation to modern ethical theories. If Aristotle is not a moral particularist, does he resemble moral universalists like Kant and Mill, generalists like Ross, or is his moral thought fundamentally different than any of these views?

A. Irwin’s Critique of the Particularist Interpretation of Aristotle

Moral particularism has emerged relatively recently as a position in ethics and its exponents include Jonathan Dancy, John McDowell, Martha Nussbaum, and David McNaughton. Moral particularists maintain that moral normativity is so complex that it cannot be adequately captured in a manageable set of moral principles.[1] Consequently, they argue, proper moral judgment consists not in the application of moral principles, but in a refined ability to ‘perceive’ the moral relevance of the particular features of an individual case. On their view, the moral wrongness of a particular act of lying is discovered through an intuition about this individual case and not through the application of a moral principle forbidding lying. In the debate about the nature of moral judgment, the main rivals of moral particularists are universalists such as Kant and Mill, and generalists like Ross.[2] On the opposite pole of the debate, universalists maintain, in contrast to particularists, that proper moral judgment consists in the application of moral principles to individual cases. Universalists claim that the salient characteristics of moral normativity can be captured in a manageable set of moral principles; as a result, principles provide an unequivocal, if not always specific, guide to moral judgment.[3] Generalists also emphasize the application of principles in moral judgment, though they are more moderate than universalists because, in contrast to the latter, they defend the possibility of moral dilemmas. According to generalists, when principles lead to conflicting moral determinations, the agent must independently assess which moral principle is most appropriate to the case at hand.

In the debate with moral universalists and generalists, particularists attempt to substantiate their claim about the complexity of moral normativity. Against universalists and generalists, they argue that the moral relevance of a particular feature (such as lying or the experience of pain) is holistic rather than atomistic.[4] Universalists and generalists maintain that the moral relevance of a particular feature will be identical in every case, even if, according to generalists, it can be overridden by another feature. For Ross, breaking a promise always counts against an action, even if it is overridden by another feature, such as bringing about a greater amount of good. For universalists and generalists, the features highlighted in moral principles are always morally relevant, and are always relevant in the same way (i.e., required, forbidden, or permissible). Particularists argue, in contrast, that the moral relevance of a particular feature is potentially unstable because it can be affected in unforeseen ways by the presence of other features. For moral particularists, there is no guarantee that a particular feature is always morally relevant, or always relevant in the same way when it is relevant.[5] Consequently, moral principles may be useful as rules of thumb that attune our sensitivity to features that are potentially relevant, but not for anything more than this.[6] Particularists support these claims by providing counterexamples against the moral principles defended by universalists and generalists, to show, for example, that lying is sometimes morally required, such as in the case of white lies, or morally irrelevant, as in the case of games that require the players to lie.[7] If the particularists’ claim that moral relevance is holistic is correct, then moral normativity would indeed be more complex than is usually assumed.

Moral particularists frequently defend their intuitionist model of moral judgment by turning to Aristotle. The particularists’ attraction to Aristotle is understandable because, like them, Aristotle emphasizes the complexity of moral normativity and the role of perception (aisthesis) in moral judgment. According to Aristotle, prudence (phronesis) consists in the ability to correctly ascertain the mean appropriate to a particular situation. Due to the complexity of each situation, correctly judging the mean is an extremely difficult task only few can accomplish. In “Ethics as an Inexact Science: Aristotle’s Ambitions for Moral Theory,” Irwin takes issue with the attempt by moral particularists to claim Aristotle’s ethics as a historical precedent for their position.[8] According to Irwin, although Aristotle argues that moral perception is necessary because ethics cannot provide exact guidance for moral agents, he still maintains that general rules provide normative guidance. Irwin supports his interpretation of Aristotle’s ethics with a detailed examination of the comments on the role of particulars and generalizations in Aristotle’s discussions of the inexactness of ethics, moral perception, and prudence. According to Irwin, Aristotle posits both universals (such as the doctrine of the mean and the virtues of character) and generalizations that admit exceptions (like the claim that virtue leads to eudaimonia) in order to provide normative guidance for moral agents (see 109, 111, 121-122, 129). If this is correct, then, for Aristotle, moral perception does not function in the absence of generalizations, but is primarily “a means of applying general rules to particular cases” (122).

Irwin defines particularism as the claim that “perceptual judgements about particular situations are normatively prior to general rules.” In other words, general rules are “true to the extent that they summarize the particular perceptual judgements of virtuous agents,” and not because “they conform to true general principles” (102, 103). Placing priority on particular perceptual judgments is necessitated by the chief difficulty facing the particularist interpreters of Aristotle: What should particularists make of moral generalizations such as the doctrine of the mean and the virtues of character? Are they not principles intended to guide moral judgments? As Irwin notes, the particularist response depends heavily on Aristotle’s comments on the inexactness of ethics, since it centers on the claim that these generalizations “are simply incomplete summaries of the considerations that the virtuous person recognizes” (101) – a partial description of the reasons behind virtuous acts.

In his response to the particularists, Irwin provides an impressive interpretation of Aristotle’s ethics, one that is much more painstaking and detailed than any of the corresponding particularist readings. Fortunately for the moral particularists, however, the received view is on their side. The prevalent interpretation of the virtues is that they are merely character traits found in the morally wise person and not a set of principles that are capable of guiding moral judgment. Consequently, the burden of proof is on Irwin to show that, contrary to the particularist interpretation, the virtues function as more than a limited description “of the considerations that the virtuous person recognizes.” Irwin argues on the basis of a detailed textual analysis, and the main steps in his argument are as follows: First, Irwin discusses how Aristotle’s comments about the inexactness of ethics pertain to moral generalizations. At the beginning of the Nicomachean Ethics, Aristotle claims that, due to the nature of the subject matter, ethics can be only be stated “roughly and in outline,” in principles that hold true “for the most part” (1094b11-27). According to Irwin, this does not mean that the generalizations that Aristotle proceeds to introduce are incomplete or deficient. Some of these generalizations, such as the doctrine of the mean and the virtues of character, are in fact exceptionless, and those which are not, such as the claims that virtue leads to eudaimonia and wealth is a good, are sufficiently exact to guide moral judgment (109-111). For Aristotle, it is not necessary to provide a greater degree of precision in moral principles, due to the practical purpose of ethics (114). In addition, Irwin argues that there is no indication that the generalizations found in the Nicomachean Ethics should be taken less seriously than those introduced in Aristotle’s theoretical works, or should be denied normative priority over particular moral judgments (112-113).

Next, Irwin examines Aristotle’s comments on perception. Perception, for Aristotle, concerns particulars, and particularists maintain that Aristotle’s concept of moral perception has nothing to do with generalizations. In contrast, Irwin argues that, “Aristotle thinks of perception primarily as a means of applying general rules to particular cases” (122; see 119). As Irwin notes, if the qualifications to a general rule are complicated or numerous, an agent may judge more effectively with the unqualified rule and a trained capacity to perceive the relevance of particular cases (119). He argues that, for Aristotle, the agent identifies the exceptional cases and ‘fills in the blanks’ by perceiving the moral status of particulars where rules fail to provide explicit guidance. Finally, Irwin turns to Aristotle’s discussion of prudence. In this section, Irwin considers the particularist counterclaim, based on a reading of Aristotle’s notion of prudence, that perception is not a means of applying general rules. Instead, particularists argue, perception is the capacity for ‘situational appreciation’, the trained ability to identify the morally relevant features of a situation (122). Aristotle’s moral generalizations, according to particularist interpreters, are simply reminders of certain features that may be morally relevant, rather than general rules capable of guiding moral judgment (123). In response, Irwin argues that in order to show that this is the case, it would be necessary to demonstrate that Aristotle “treats all non-trivial generalizations as reminders that we are morally permitted to disregard if our perception of particular situations prompts us to act differently” (124). Irwin considers a related claim that might commit Aristotle to this view. In his distinction between prudence (phronesis) and understanding (nous), Aristotle highlights the role of perception in each of these virtues of thought. Most importantly, he states that understanding is a type of perception in which universals are derived from particulars. Perhaps the claim that universals are derived from particulars is evidence that, for Aristotle, particular judgments have normative priority over generalizations. Against such an interpretation, Irwin argues that although particular judgments have some sort of priority ‘to us’ in the process of discovery, universal judgments are, for Aristotle, prior ‘by nature’ (124-125). Consequently, Irwin concludes, there is no evidence to support the claim that Aristotle’s moral generalizations are not intended to provide normative guidance.

To Irwin’s critique of the particularist interpretation of Aristotle we can add the following point: Aristotle’s comments on the relationship of prudence and deliberation suggests rather strongly that, for him, moral perception is normatively guided by moral generalizations. In Book VI of the Nicomachean Ethics, Aristotle indicates the significant role that deliberation plays in prudence when he states that the prudent person is one who deliberates finely on the topic of living well (1140a26-32). According to Aristotle, deliberation guides virtuous action through reflection on issues such as the mean and the good, and requires knowledge of both universals and particulars (1141b9-23, 1142a14-24, 1142b28). The use of universals and particulars in deliberation is described in most detail in the passage from Book VII commonly known as the practical syllogism. There, Aristotle gives an example of a deductive syllogism, with a universal major premise and a particular minor premise, the latter acquired through perception (1147a25-31).[9] If Aristotle understands deliberation as a deduction from universals, as this passage suggests, then, for him, moral generalizations guide particular perceptual judgments in a manner incompatible with particularism.

B. Moral Particularism and Exceptionless Moral Principles

Although Irwin’s textual analysis is very thorough, there is a significant flaw in his argument against the particularist interpretation of Aristotle. Irwin, however, is not to blame for this flaw. Ironically, it is the moral particularists who are responsible because the flaw arises from an inadequacy in the way that they conceive their own position. According to its proponents, moral particularism rejects all general moral principles. However, this claim is false. I will argue that there is a type of exceptionless moral principle that moral particularism is committed to, as is any view that defends the possibility of moral judgment.[10] Consequently, even if it is correct to say that general moral principles such as the virtues of character and the doctrine of the mean provide normative guidance, this is insufficient to prove that Aristotle is not a moral particularist. Whether or not our reconsideration of moral particularism succeeds in rescuing the particularist interpretation of Aristotle, it will remedy a misconception found in model of moral judgment defended by particularists.

As we have seen, moral particularists maintain that proper moral judgment is not guided by moral generalizations. They frequently argue for moral particularism on the basis of Wittgenstein’s account of rule-following, and associate their view with moral intuitionism and the anti-theory position in ethics.[11] The link between particularism and intuitionism should not be surprising, for an intuitionist model of moral judgment seems to be the only viable option left if one rejects moral generalizations. How else can we discover the moral status of particulars except via intuition? If particularism is committed to moral intuitionism, it is necessary to specify which version since some versions of moral intuitionism are generalist in nature. According to Dancy, particularism advocates a model of moral judgment in which the moral relevance of features of cases is ‘perceived’ intuitively on a case by case basis.[12] This stands in contrast to moral intuitionists like G.E. Moore, who maintain that we intuit moral generalizations.[13] In addition to claiming that we only intuit particulars, Dancy maintains that proper moral intuition is not innate, and that moral intuitions are not immediately justified. Following Aristotle, Dancy describes moral perception as a trained capacity and places great emphasis on moral education.[14] In addition, Dancy claims that particular moral judgments, like particular aesthetic judgments, can be justified by explaining the way one ‘sees’ the relationship between the salient features of the case at hand.[15] This opens up the possibility of detailed deliberation and moral debate about particulars, though it is hard to tell how far Dancy would extend this possibility because he does not indicate where, on his view, discussion ends and intuition begins.

One consequence of the intuitionist model of moral judgment advocated by virtually all moral particularists is that particularists often exhibit anti-theoretical tendencies, maintaining that normative theories and claims about such things as the nature of basic moral categories distort our understanding of moral judgment and interfere with its exercise. On this view, proper moral judgment is intuitive and there is little we can contribute to clarify how it functions in general besides examples from individual cases. The result of this anti-theoretical tendency is a quietism about the nature of moral categories, which can be seen in Dancy’s claim that, in moral judgment, “there is nothing that one brings to the new situation other than a contentless ability to discern what matters where it matters.”[16] Moral judgment is a contentless ability, according to Dancy, because it receives no normative guidance from moral generalizations or from knowledge about the nature of moral categories such as right and wrong, good and bad, etc. Although moral theory may be useful in other ways, it in no way improves one’s ability to judge the moral status of individual cases or to justify one’s judgments. Moral judgment is a capacity for intuition that arises from experience with particulars instead of the application of a moral theories. But is moral judgment really a contentless ability? Is it really possible for proper moral determinations to consistently arise from mere ‘perception’ or ‘intuition’, to arise without the guidance of one or more general principles? Is it possible to make moral judgments without some knowledge of moral categories, without some theoretical commitments in ethics?

In conflict with the anti-theoretical tendency found in most proponents of moral particularism, there is a type of exceptionless moral principle that moral particularists are committed to. I will refer to this principle as the principle of moral force. The principle of moral force is a higher-level moral principle that is implicit in the concept of a moral category. Moral categories are often distinguished from purely theoretical categories by pointing to the fact that there is an implicit recommendation found in the distinction between various moral categories. The categories moral and immoral, right and wrong, good and bad, virtue and vice, and between forbidden, permissible, and required are not equally acceptable options that imply a certain neutrality or indifference about which ones are chosen. Rather, they contain an intrinsic priority of some options over others, a teleology that posits norms.[17] The teleology implicit in moral categories provides normative guidance for all moral judgments, and can be stated in the exceptionless principle “Do what is moral.” This principle contains a recommendation that one act morally instead of immorally. The principle of moral force, as I have formulated it, may sound extremely general. This is intentional: One might reformulate the principle to say “Do what is moral because it is required by duty,” “Do what is moral because it is good,” or “Do what is moral because it is virtuous.” I am attempting to avoid debates about the priority of the right and the good, and my contention is that the principle of moral force, and the simple model of normativity found in it, underlie all of these more specific reformulations. Without the normative distinction between moral categories, moral claims would lack normative force. The principle of moral force is what provides normative force to moral claims, whether one explains the nature of this normative force in a manner that is deontological, consequentialist, or aretetic. My contention, then, is that there is a normative priority implicit in every set of moral categories, a priory that provides normative guidance for every instance of moral judgment.[18]

If it is true that the principle of moral force provides normative guidance for all moral judgments, then, despite what its advocates claim, particularism does not reject all general moral principles. Particularists affirm the possibility of moral judgment, and all instances of moral judgment are normatively guided by the priority of morality over immorality. Moral judgment is neither purely intuitive nor purely a matter of particulars because moral normativity is based to some extent on the distinction between general moral categories. Consequently, there is an implicit reference to the moral principle “Do what is moral” in every instance of moral judgment, whether the agent is aware of it or not. How might particularists respond to these claims about the principle of moral force? For one, particularists might attempt to defend their version of moral intuitionism by arguing that the distinction between moral categories is not based on a general principle, but rather is intuited on a case by case basis. In addition, particularists might argue that the principle of moral force is not really a moral principle or is, at best, a trivial one since it does not refer to features of cases.

Moral particularists might react to the prospect of being committed to the principle of moral force by responding in much the same manner as the particularist interpreters of Aristotle do to the claim that Aristotle posits moral generalizations. As we saw above, their strategy is to argue that moral principles are merely incomplete summaries of the considerations the virtuous person recognizes and have no role in moral judgment. On this view, moral principles are merely incomplete descriptions of the reasons behind past actions, and there is no guarantee that they provide proper normative guidance. In the present context, particularists could argue that the principle of moral force describes how the proper moral agent acts without itself providing any normative guidance. If this was the case, then the distinction between moral categories would not be based on general discursive knowledge. Rather the distinction would be made on a case by case basis in particular intuitions. On this view, the proper moral agent does not distinguish between moral categories as a result of ethical theorizing or through an explicit or implicit reference to her moral education, but through the purely intuitive recognition that one should act morally instead of immorally in this case. This would explain why, for Dancy, we can say little to nothing about the nature of moral categories.

There are two difficulties that arise for this view of how agents distinguish between moral categories. The first is that particularists have not shown why the intuitionist model is preferable to general discursive knowledge. If there is a set of generalizations useful for moral judgment, what reason could we have not to use them to guide our actions besides an ascetic desire to make things more difficult for us than necessary? Would not restricting oneself to intuitions when useful generalizations are available also increase the likelihood of error in moral judgments? Apparently, particularists can only show that the intuitionist model of moral judgment is preferable by demonstrating that the principle of moral force is a questionable generalization or that intuition is noticeably more effective – quicker and better able to avoid error – than the application of the principle of moral force. The second appears to be the more promising option, though it is itself quite challenging, given our ability to use conceptual generalizations both efficiently and accurately.

The second difficulty facing the intuitionist model is that it appears more plausible to explain the distinction between moral categories through the principle of moral force. The particularist-intuitionist can provide no explanation of the great consistency with which we make this distinction, and seems to posit one judgment too many in the case of particular moral judgments. According to the particularist-intuitionist, it just happens to be the case that morality has priority over immorality in all previous instances of moral judgment. On this view, the consistency found in the distinction between moral categories is purely accidental, and the possibility is left open that the priority will be reversed in future cases of moral judgment. Of course, there are moral skeptics and nihilists who entertain the possibility that immorality is normatively prior to morality. But by doing so they undermine the possibility of moral judgment, a consequence that the moral particularist cannot accept. In contrast, the principle of moral force makes it possible to explain the stability of our moral categories: The normative structure found in our moral categories remains consistent in our moral judgments because it is built into our discursive knowledge of moral categories.

In addition, the intuitionist account of moral categories has peculiar implications for moral judgment since it requires a great number of extra judgments that would otherwise be unnecessary. Irwin makes this point in his criticism of the particularist interpretation of Aristotle when he writes: “Aristotle does not suggest … that we need any special exercise of perception to see that we must act bravely in this particular case where brave action will involve significant harm” (115; see 123-124). It seems strange that, in every instance of moral judgment, the proper moral agent would need to consider whether she should act morally. On the traditional view of morality, it is precisely the proper moral agent – in contrast to continent and incontinent moral agents – who would not need to consider whether to act morally.

Against the intuitionist model of moral judgment, I am defending the claim that moral judgment is always guided by at least one moral generalization. Moral judgment is not purely intuitive because it requires the normative guidance of the principle of moral force. The priority of morality over immorality that lies at the heart of moral judgment is grounded in our discursive knowledge of general moral categories, not in particular intuitions. To modify a famous quotation from Kant’s theoretical philosophy: Intuitions without moral principles are blind, just as moral principles without intuitions are empty.[19] In moral judgment, generalizations and particular intuitions work hand in hand.

In the following section, we will discuss the relationship between moral generalizations and particular intuitions in more detail. Presently, however, there is another possible particularist response to consider. Particularists might accept the claim that they are committed to the principle of moral force, but argue either that it is not in fact a moral principle or that it is trivial. If the principle of moral force is a principle that every moral agent accepts, a principle that guides every instance of moral judgment, it is quite different from moral principles such as “Never lie” or “Assist others in need,” which are only applicable to certain cases. The principle of moral force is applicable to every case of moral judgment, whereas the latter principles, like other traditional moral principles, are applicable only to cases in which deceit is possible or there are others in need whom one can assist. In addition, the principle of moral force, due to its generality, does not provide specific guidance. The prohibition against lying and the duty of beneficence, in contrast, give the agent some degree of determinate guidance in the individual cases in which they are applicable.

In the debate between moral particularism and universalism, moral principles are typically understood as statements that connect particular features of an individual case with moral categories, connecting the feature of lying, for example, with the moral category of forbidden actions, and the feature of assisting others in need with the category of required actions. These are particular features that are not necessarily present in every individual case. But in the principle of moral force, there is no mention of particular features of cases. Rather, this principle only refers to the general categories moral and immoral (the latter only implicitly). The principle of moral force, “Do what is moral,” is extremely general and indeterminate since it is applicable to every case and does not provide specific guidance for moral judgment. The particularist might argue that by classifying “Do what is moral” as a moral principle we are illegitimately stretching the concept of a moral principle. But are we really broadening the category of moral principle when we include moral recommendations that make no reference to particular features of cases? There is precedence for this type of moral principle in Kant’s concept of the categorical imperative. I am referring here to only to the general definition of the categorical imperative – a command that applies to everyone, all the time – and not to the specific formulations of the categorical imperative (universal law, end in itself, etc.) or the duties derived from them. It is widely recognized that the categorical imperative, the claim that we should always do what is morally required, is a moral principle even though in its general form it contains no reference to particular features of cases. Like the principle of moral force, it is an exceptionless moral rule that is intended to guide individual instances of moral judgment. As this suggests, it is not so strange to refer to the recommendation that we “Do what is moral” as a moral principle.

But even if the principle of moral force can legitimately be called a moral principle, it might still be trivial. Given its level of generality and indeterminacy, perhaps it is trivially true or morally redundant. Perhaps moral particularists can accept it without significantly altering their position. Unfortunately for particularists, this is not the case. As I have argued, the principle of moral force stands in conflict with the anti-theoretical tendency of many particularists, and the version of moral intuitionism they defend. Particularists would have to significantly qualify both of these aspects of their view before they could accept this principle.

Our discussion of the principle of moral force reveals that moral particularists go too far when they claim to reject all moral generalizations. Since they defend the possibility of moral judgment, particularists are committed to the principle of moral force. Moral judgment would not be possible without the normative guidance that this principle provides. But is moral particularism still a coherent position if it is committed to an exceptionless moral principle? Is it possible to distinguish the principle of moral force from the moral generalizations that particularists would still reject, and thus preserve the border dividing particularism from generalism and universalism? As we have seen, the principle of moral force is distinguished from other moral principles by the fact that it does not refer to particular features of individual cases, such as lying, benefiting others, or developing one’s talents.[20] Unlike these other principles, the principle of moral force has no bearing on the moral relevance of particular features of individual cases. Its indeterminacy about the moral relevance of particulars distinguishes it from other moral principles, and also indicates why it is compatible with moral particularism. Since it is indeterminate about the moral relevance of particulars, the principle of moral force does not predetermine the moral status of any given situation. Hence, moral particularism is still a coherent position because the fact that it is committed to a moral principle that is both exceptionless and indeterminate does not preclude it from rejecting determinate moral principles and thereby leaving the moral relevance of particular features to be determined by moral perception. However, moral particularists are forced to qualify their rejection moral generalizations and to rephrase their criticisms of moral universalism and generalism. As we will see in the following section, the distinction emerging from our discussion of the principle of moral force – the distinction between determinate and indeterminate moral principles – leads to a clearer understanding of the relationship of moral particularism to universalism and generalism.

The non-triviality of moral particularism’s commitment to the principle of morality is suggested not only by the fact that it conflicts with the anti-theoretical tendency of many proponents of moral particularism, but also because opens up additional possibilities for moral justification. As we have seen, particularists such as Dancy maintain that moral judgment is a contentless ability and argue for this claim on the basis of moral intuitionism and Wittgenstein’s account of following rules. However, moral judgment cannot be contentless. As I have argued, moral judgment is only possible on the basis of moral categories and these categories must have enough conceptual content to allow us to keep them distinct from one another. Moral judgment requires at least some discursive knowledge of moral categories. And, as this suggests, moral judgment cannot be purely intuitive, though the content of the categories that guide moral judgment might be indeterminate in regard to particulars.

Though moral particularism is often criticized for eliminating the possibility of moral justification, this criticism is misdirected. As we saw above, Dancy argues that it possible to justify moral judgments on a case by case basis, discussing and debating the moral relevance of particular features in this case without claiming that the same features are morally relevant in every case or relevant in the same way when they are relevant. He illustrates this process by drawing an analogy to the justification of aesthetic judgments.[21] Dancy’s insight is an important contribution to moral particularism and increases its plausibility. But there is no reason why moral justification must stop here. Dancy argues against the use of moral generalizations in moral justification and suggests that we should avoid discussion of the nature of basic moral categories.[22] But if the particularists’ commitment to the principle of moral force reveals additional possibilities for moral justification, is it not the case that Dancy is drawing the limit to moral justification too narrowly due to his anti-theoretical inclinations? It is true that moral particularists will never be able to elucidate the nature of basic moral categories in as much detail as universalists and generalists because they cannot define moral categories in terms of particulars features of cases, such as pleasure and pain or the universalizability of moral maxims. However, particularists can elucidate the nature of basic moral categories to some degree by comparing and contrasting these concepts with one another and related notions. We can think of this along the lines of coherentist models of epistemology: Concepts such as rightness, goodness, virtue, morality, legality, and justice are related to one another, and can improve our understanding of the individual concepts by comparing and contrasting them with one another. Such a discussion allows us to better explain why an act is moral and may also sharpen moral perception. There is no reason internal to particularism not to examine the nature of moral categories in this manner, and, considering what it adds to the particularist attempt to justify moral judgments, every reason to follow through with it.

C. Aristotle: Moral Particularist, Universalist, Moderate, or None of the Above?

As I have argued, moral particularism is committed to a type of exceptionless moral principle, i.e., the indeterminate principle of moral force. This principle normatively guides every act of moral judgment because it is implicit in moral categories. If moral particularists are in fact committed to the normative guidance of an exceptionless moral principle, it is necessary to reconsider whether Aristotle is a particularist. Irwin’s critique of the particularist interpretation of Aristotle undermined by this alteration of the definition of particularism because what is at issue is no longer whether Aristotelean moral concepts such as the doctrine of the mean and the virtues of character merely provide normative guidance, or stated differently, whether they are merely moral generalizations. Rather, it is also a question of the relative indeterminacy of these moral principles. Are these principles as indeterminate as the principle of moral force and thereby compatible with particularism? In this section, we will conclude our discussion by reconsidering the particularist interpretation of Aristotle and the possibility of locating Aristotle in the debate between universalism and particularism.

If Irwin’s claim about the normative guidance of moral generalizations in Aristotle’s thought is insufficient to prove that Aristotle is not a particularist, the issue can be resolved by considering the status of these generalizations. Are the doctrine of the mean and the virtues of character indeterminate or determinate principles? In other words, are they claims about the moral relevance of particular features of individual cases? There should be no doubt that Aristotle’s moral principles have some degree of indeterminacy: As Irwin points out, Aristotle states that these principles do not provide precise guidance for moral agents (115-118, 121). This indeterminacy can be seen, e.g., in the doctrine of the mean, in which Aristotle maintains that virtue concerns having certain feelings or doing certain actions at the right time, toward the right people, in the right way, etc., but leaves the meaning of the concept of rightness open so that rightness will be determined by moral perception within the context of each individual case (1106b20-24).

However, if we examine Aristotle’s definitions of the particular virtues of character, it becomes clear that these moral principles, though they are less determinate than many traditional moral principles, are not as indeterminate as the principle of moral force. Aristotle defines each virtue of character in terms of a particular feature of a case, i.e., the degree to which one experiences of a certain feeling or does a certain action. For example, Aristotle defines bravery as experiencing the right amount of fear, and generosity as giving and taking the right amount of money (1106b16-20, 1107b9-15). These moral principles are quite indeterminate, for Aristotle leaves the proper amount of these feelings or actions to be determined by moral perception on a case by case basis. However, Aristotle still takes a definite position on their moral relevance. The virtues of character are too determinate for moral particularism because particular features such as fear and giving and taking money, according to Aristotle, are always relevant to moral judgment.[23] For moral particularists, there is no guarantee that any particular feature is always morally relevant, or always morally relevant in the same way when it is relevant.

In addition, Aristotle makes a stronger claim about the moral relevance of other features, a point that commentators often overlook. Aristotle claims that some actions and feelings that do not have a mean are always vicious, and lists spite, shamelessness, envy, adultery, theft, and murder as examples (1107a9-13). These particular features are always morally relevant and always relevant in the same way. The moral prescriptions implicit in these vices of character are very specific. Unlike the virtues of character, there is no leeway for moral judgment to determine the right amount of these particular features because, for Aristotle, any presence of these features is morally forbidden.[24] Although Aristotle’s emphasis on the consideration of particularity in his account of moral judgment is similar to moral particularism and many of his moral principles have a great deal of indeterminacy, these principles are too determinate for Aristotle to be characterized as a moral particularist: All of the virtues and vices of character contain claims about the moral relevance of particular features of cases, which, for the moral particularists, must be left indeterminate.

If Aristotle is not a particularist, can we still classify him in relation to the contemporary debate between moral universalism and particularism? This is an issue that Irwin does not explore. We will take a look at it by way of conclusion because it raises interesting questions about the nature of Greek virtue ethics, as well as the role of intuition in modern moral theories.

Particularist interpreters of Aristotle and defenders of virtue ethics may cringe at the suggestion that Aristotle is a moral universalist like Kant or Mill, or even a generalist like Ross. However, this revulsion is based on a misunderstanding of the generalist and universalist positions as well as the difference between virtue ethics and modern ethical theories. If Irwin’s reading of Aristotle is correct, virtue ethics is not distinguished from modern ethical theories like Kant’s ethics and utilitarianism by the fact that the latter posit moral rules whereas virtue ethics emphasizes character traits. As Irwin argues, claims about character traits can lead to generalizations that provide normative guidance in the same way as claims about the consequences of actions and the formal structure of intentions. In addition, recent scholarship has shown that modern philosophers such as Kant also place significant emphasis on character traits and habituation.[25] The compatibility of the concept of virtue with modern ethical theories blurs the common distinction between these theories and virtue ethics. In my view, the characteristic that distinguishes virtue ethics from modern ethical theories is not that one emphasizes character traits and the other posit rules, but rather that the rules posited by virtue ethics are usually more indeterminate than those posited by modern ethical theories, requiring greater input from the deliberation about and intuition of particulars.[26]

We have seen that moral particularism is not merely a facile rejection of moral principles, though this is the way it is often understood. Moral particularism is more subtle than it is usually given credit for, as can be seen in the fact that it preserves the possibility of moral justification and debate. However, particularists are rarely appreciative of the nuances found in the views with which they are competing. Particularists often claim that the universalist model of moral judgment, given its emphasis on the application of moral principles, is a crude mechanism for churning out moral determinations. They claim that universalism is insensitive to context and eliminates any significant input from the moral agent.[27] This characterization is inaccurate, for it overlooks the role of non-rule based judgment in these theories. Non-rule based judgment, the deliberation about and intuition of particulars, is found in one form or another in every model of moral judgment. As we will see, moral universalism and generalism are more flexible and nuanced than their critics usually realize. Non-rule based judgment fills in the gap or indeterminacy of the rules guiding moral judgment, and, as the name suggests, it is a judgment made about particulars without any specific guidance from moral principles. We have seen that moral particularism leaves a broad scope for non-rule based judgment: Moral perception operates within the indeterminacy of the principle of moral force, and determines the moral relevance of every particular feature of a case. Since moral perception requires both a general rule and intuition, the contrast between particularism, generalism, and universalism is a difference of degree rather than kind. Intuition has a role in universalism and generalism as well, though admittedly a more limited one. In what follows, we will take a closer look at the difference between moral universalism and generalism and the types of non-rule based judgment operative in each of them. Only after doing this will we be prepared to return to the issue of classifying Aristotle’s ethics.

As we saw above, both moral universalists and generalists maintain that proper moral judgment consists mainly in the application of principles. What distinguishes these positions, however, is that generalists defend the possibility of moral dilemmas, a possibility that is ruled out by moral universalism. Since universalists and generalists hold that the application of principles is a central element of moral judgment, they are committed to a type of non-rule based judgment I will refer to as application judgment.[28] This is due to the fact, pointed out by Kant and others, that rules, even simple mathematical rules, never provide complete guidance concerning their proper application. Wittgenstein’s comments about the act of following a rule are pertinent here: A rule never provides us with enough information to eliminate the possibility of misunderstanding the rule and misapplying it.[29] Since rules inevitably contain some degree of indeterminacy regarding their application, error is always possible. Application judgment is the form of non-rule based judgment that determines how to apply general principles to particular cases through the use of intuition and, in some instances, deliberation. Adjudicative judgment is a necessary component of any model of moral judgment based on the application of principles.[30]

Since generalists like Ross assert the possibility of moral dilemmas, they are committed to another type of non-rule based judgment I will refer to as adjudicative judgment. Moderates are ethical pluralists and maintain that there are a number of irreducibly basic moral principles that have the potential of conflicting with one another. For example, Ross claims that the basic moral duties include the duty to keep one’s promises, the duty to make up for past wrongs, and the duty not to harm others.[31] According to Ross, when these principles conflict, such as in the case of a promise whose fulfillment would lead to harm, the conflict is adjudicated by the moral agent without the specific guidance of any moral principle. In adjudicative judgment, the moral agent determines which duty or group of duties is most appropriate to the case at hand, using intuition and sometimes deliberation. In the example just mentioned, the moral agent must decide whether in this case the duty to keep a promise overrides the duty to avoid harm to others, or vice versa.[32] Although conflicting principles are reconciled by non-rule based judgment, on the moderate model of proper moral judgment the agent always follows (i.e., sides with) at least one moral principle when judging an individual case. Adjudicative judgment never overrides all of the moral principles that one is considering. Consequently, moral agents are guided by moral principles to the extent that they choose between the moral determinations generated by the application of principles.[33]

Since, for Ross, basic moral principles may conflict with and override one another, none of these principles are exceptionless. Instead of being an absolute requirement, for Ross, a moral principle, expresses a prima facie duty. In other words, it describes an act that “would be a duty proper if [there] were not at the same time of another kind [of duty] which is morally significant.”[34] Although generalists need not claim, as Ross does, that all basic moral principles admit of exceptions, at least two of these principles must be prima facie duties in order to preserve the possibility of moral dilemmas.

In contrast to generalists, moral universalists rule out the possibility of moral dilemmas. Since virtually every universalist posits a variety of moral principles,[35] some means of avoiding conflicts between these principles must be introduced. Universalists use two main strategies: Establishing an order of ranking, and formulating the principles in such a way that they do not apply to contrary actions. Universalists who follow the first strategy posit a hierarchy between principles so that the basic moral principles have the highest ranking and override lower-order principles when the two lead to conflicting moral determinations.[36] For example, Mill argues that the principle of utility has priority over other principles, such as the duty not to lie. Similarly, conflicts between lower-order principles of the same rank are resolved in many cases by referring to a higher order principle.[37] The name ‘moral universalism’ reflects the fact that the principles of the highest rank – the basic moral principles – are exceptionless (i.e., universal) since they are never overridden by other principles.

This strategy, however, is not completely effective if there is more than one principle of the highest rank. A second strategy, which can be used independently or in conjunction with the first, has been developed to eliminate the possibility of conflict between multiple principles of the highest rank. Principles are formulated in such a way that they do not require contrary actions; consequently, they cannot lead to conflicting moral determinations about the same action. Kant’s ethics is a good example of the latter strategy used in conjunction with the first. Like Mill, Kant posits a hierarchy of principles. Kant distinguishes between perfect and imperfect duties, and maintains the priority of the former. However, since there are numerous perfect duties, such as the duties not to lie, not to steal, not to commit suicide, conflict between perfect duties still appears to be a live possibility. This possibility is eliminated in the way that Kant formulates perfect duties. Kant differentiates perfect and imperfect duties by referring to them as narrow and wide duties, respectively.[38] As this second pair of names suggests, Kant defines the perfect duties narrowly enough that they do not apply to contrary actions. The features of cases cited in perfect duties are mutually independent: For example, Kant appears to believe that it would never be necessary to steal or commit suicide in order to avoid lying.[39]

A final form of non-rule based judgment can be most easily discussed in conjunction with the example of Kant’s imperfect duties. As we have seen, Kant defines perfect duties narrowly: They concern specific actions like lying, stealing, and committing suicide. Imperfect duties, on the other hand, are ‘wider’ or more indeterminate. Included among Kant’s imperfect duties are the duty to help others in need and the duty to develop one’s talents. What it means to satisfy the imperfect duties is less clear than is the case with the perfect duties because the actions referred to in the former (e.g. helping others in need and developing one’s talents) are much less determinate than specific actions such as lying, stealing, and committing suicide. With the exception of a few borderline cases, it is relatively clear what it means to lie or commit suicide. Since they are narrowly limited activities, it is comparatively easy to figure out how to satisfy the prescriptions against them. In contrast, developing one’s talents and helping others in need are more indeterminate types of actions and include a broader range of possible activities. This can be seen in the fact that one might help a person in need by lying or stealing or by doing a number of other things such as providing physical or material assistance. Since imperfect duties are broadly defined, it is more difficult to determine how best to satisfy them than perfect duties. There are a wide variety of ways in which one can help a person in need or develop one’s talents, and in many cases, it is far from obvious which way is most appropriate.

Another difficulty arises from the level of generality found in imperfect duties. Since they concern a broad range of actions, these duties often conflict with one another and with perfect duties: Should I stop and help the victims of an auto accident, or continue on to class in order to develop my intellectual abilities? Should I lie in order to help a friend out of a jam or to get into a better university? As we have seen, Kant rules out potential conflicts between perfect and imperfect duties by positing the priority of the former. According to Kant, one cannot adequately satisfy imperfect duties by lying, stealing, etc. However, given the narrowness of the perfect duties, they cannot adjudicate conflicts between imperfect duties. The prescription against lying, for example, provides no help in determining whether to assist a person in need or develop one’s talents. In order to avoid both of these potential problems – the indeterminacy of imperfect duties and potential for conflict between them – Kant grants the moral agent great latitude in applying the imperfect duties. It is up to the agent to determine how much emphasis to place on each imperfect duty in a particular case, and how to best satisfy these duties. In the previous example, one might determine that helping the accident victims takes precedence over attending class that evening, and that, lacking medical training, the best way to help the victims, is to find a phone and call for the paramedics.

When weighing competing duties against one another, the moral agent uses adjudicative judgment. The agent does not have to choose between competing duties, but could weigh them equally or place more emphasis on one than the other.[40] Universalists and generalists share another form of non-rule based judgment in common as well. As our discussion of imperfect duties indicates, some moral principles are more indeterminate than others. The more indeterminate a moral principle is, the more leeway the agent has in determining how to best satisfy the principle.[41] The application of indeterminate moral principles such as Kant’s imperfect duties involves a type of non-rule based judgment with which we are already familiar: moral perception, the perceptual form of judgment associated with Aristotle and moral particularism. Though initially it may seem inappropriate to claim that perceptual judgment is found in some versions of moral universalism and generalism – in theories based on the application of moral generalizations, it is helpful to recall that, as we argued above, perceptual judgment is always guided by at least one moral generalization, i.e., the principle of moral force. Although it is guided by additional moral generalizations when used by universalists or generalists, perceptual judgment retains the same basic structure. Determining how best to apply indeterminate moral principles such as the principle of moral force or imperfect duties is a task that depends quite heavily on one’s imagination and experience. The greater ability one has to invent novel options and to recall a wide range of relevant information, the better one will be able to satisfy these principles. After gathering together the possible options, one makes an intuitive choice, which may be guided by prior deliberation. Perceptual judgment functions in the same manner in universalism and generalism as in particularlism, though the environment is more restrictive, since there are other more determinate general principles that guide it in addition to the principle of moral force. Though universalism and generalism are more restrictive than particularism, they can still grant a significant amount of leeway to the moral agent in judging particulars, allowing her to be sensitive to context and to respond in a creative manner.

Now that we have a better understanding of the nature of moral universalism and generalism, can we place Aristotle’s ethics under one of these headings? Does it resemble either of these positions or is his ethics significantly different from both? If Irwin is correct when he argues that the moral generalizations in Aristotle’s thought are intended to provide normative guidance, I see no prima facie reason not to compare Aristotle’s ethics to these other two theories. Like universalism and generalism, Aristotle’s model of moral judgment requires both the guidance of general moral principles that pertain to the moral relevance of particular features, and multiple forms of non-rule based judgment, which include intuition. There may be a difference of degree in the determinacy of the moral principles posited in each theory, but a similar difference of degree is also found between competing universalist theories and between competing moderate theories. The primary question, then, is whether Aristotle’s ethics more closely resembles moral universalism or generalism: Does Aristotle assert or rule out the possibility of moral dilemmas? There is no suggestion in Aristotle’s writings of irreconcilable conflicts among the virtues and vices of character. In fact, Aristotle maintains that the practically wise person satisfies all of these moral principles simultaneously (1144b33-1145a2; see 1103a25-26, 1129b26-35). What makes this possible is the degree of indeterminacy found in the virtues of character: Aristotle defines them so broadly (as ‘the proper amount of a feeling or action’ while leaving the proper amount to be determined on a case by case basis by perceptual judgment) that the moral agent has enough leeway to satisfy each of the virtues in a way that is compatible with the others.[42] Consequently, Aristotle’s ethics resembles moral universalism more closely than generalism.

A closer examination of the similarities between Aristotle’s ethics and universalism suggests that Aristotle should be placed in this category. And – though this claim might meet a great deal of initial resistance – that Kant is the universalist that Aristotle most closely resembles. Consider the following points: Despite the fact that particularists take Aristotle’s notion of moral perception as their model for moral judgment, Aristotle’s moral perception is too determinate for the particularists because it operates within the constraints of the virtues and vices of character. A similarly constrained version of perceptual judgment has a major role in Kant’s model of moral judgment, given the indeterminacy of the imperfect duties. A second similarity between Aristotle and Kant is that Aristotle’s distinction between the virtues and vices of character parallels Kant’s distinction between the perfect and imperfect duties. Certain vices of character, such as spite, adultery, and theft, are narrowly defined like the perfect duties.[43] On the other hand, the virtues are character, such as bravery and moderation, are defined broadly like the imperfect duties.[44] According to Aristotle, the virtues of character admit of a mean but these vices do not. The virtue of bravery is indeterminate because there is a broad range of possible states it could refer to, i.e., any point in the continuum between the feelings of fear and confidence, and we cannot determine, outside of a given context, where the mean in fact lies. In contrast, the vice of spite is relatively specific because it is clear which feeling is being referred to in this vice. Given these similarities, Aristotle appears to fall into the category of moral universalism.

However, there is an objection we need to consider that may upset this classification of Aristotle’s ethics. One might ask whether the list of virtues and vices of character that Aristotle provides is comprehensive. If not, does this indicate that there are feelings and actions not covered by the virtues and vices of character, opening the door, in certain cases, to a less constrained type of perceptual judgment? Responding to the first question, it is not clear whether Aristotle thought he provided an adequate list of the vices of character. In relation to the virtues of character, Aristotle seems to provide a comprehensive list of them because he apparently attempts to consider every possible feeling and action that may be virtuous (1107a32-1108b10). However, this is not the case with the vices of character that do not admit of a mean, of which Aristotle only provides a few examples. On the other hand, Aristotle implies that it would be relatively easy for his readers to identify the remaining vices.[45] As it turns out, Aristotle generates his list of the particular virtues and vices of character from the distinctions available to him in the Greek language of his time. As such, these principles reflect the views of 4^th century B.C. Greek aristocracy. Given Aristotle’s comments, he seems to think that an educated Greek aristocrat would not have any difficulty identifying other vices of character.

However, even if Aristotle’s list of virtues and vices of character is not comprehensive, this has no effect on the classification of Aristotle as a moral universalist. One need not provide a comprehensive list of moral principles in order to be a universalist, or a generalist. Instead, one need only assert that all morally relevant features could be captured in a manageable set of moral principles. Ross acknowledges the possibility of discovering new moral principles, and it is seems clear that Kant and Mill could add additional principles without upsetting the universalist model of moral judgment. In addition, even if some virtues and vices of character are still undiscovered, this does not open the door to a more indeterminate form of perceptual judgment. Every use of moral judgment is already constrained by the virtues and vices of character that have been discovered thus far. Consequently, even in cases that are especially indeterminate, perceptual judgment is still significantly constrained.

In conclusion, the fact that the virtues and vices of character provide guidance concerning particulars without leading to moral dilemmas suggests that Aristotle should be classified as a moral universalist. The virtues of character are, admittedly, more indeterminate than the moral principles defended by universalists such as Kant and Mill. However, this only goes to show that moral universalism and generalism are more flexible and versatile than is usually thought.

Notes