Making a necessity of Virtue (Book Review)

Reviews the book `Making a Necessity of Virtue,' by Nancy Sherman.

Copyright © 2000. All rights reserved. Sherman, Nancy. Making a Necessity of Virtue. Cambridge: Cambridge University Press, 1997. Pp. xvii + 387. $64.95 (cloth); $19.95 (paper). Making a Necessity of Virtue is about the ethics of Aristotle and Kant. "Specifically," according to the first chapter, it "is about the role of emotions and practical reason in each theorist's account of ,;rtue," though with "greater attention to the place of emotions in moral character" (p. 21). This description is a bit mis-· leading. Much of the book has nothing directly to do with emotions, and the­interrelations of emotion and practical reason are often left for the reader tc· workout. Nancy Sherman starts from the idea, fostered by Kant himself, that his ethiC!. and Aristotle's are poles apart. Some have tried, however, to find a "rapproche-· ment" between (p. 2). Sherman does so, too, especially to the extent that she see! Kant as in part an "exponent oh;rtue" (p. 1) and as an advocate of a somewhat Aristotelian notion of practical reason (chap. 7). Still, she recognizes that "the rush to mutual accommodation" can go too far, and therefore she aims "to re .. store the distinctive theoretical structures of each tradition within a dialogue that nonetheless acknowledges some shared terrain" (p. 2). Often it is difficult to tell when Sherman is discussing the actual ,;ews of Kant and Aristotle themselves, or their actual historical "traditions," or else various possible philosophical articulations or developments of their thinking. She notes that her "debate is a reconstructed one" (p. xi), but nonetheless she often ap­peals to textual details in ways that limit the potential freedom of the recon­struction (p. 4). She is much concerned with the picture of Kant as "the harsh 'duty philosopher,' unsympathetic to human emotions" (p. 1), and as too "action<entered," and inattentive to "affect" (p. 4). But it is often hard to tell when she is claiming that Kant himself does not fit this description, or that a Kantian ,;ew need not do so, or simply that any ethical ,;ew that does so is therein deficient, though it seems clear that she advocates all three claims (see, e.g., pp. 20, 146). Copyright © 2000. All rights reserved. 190 Ethics October 2000 One difficulty in understanding Sherman's line of thought arises from her not having explained what she takes "emotions" to be. She makes clear thatemo­tions are not simply "feelings" (p. 57), but elsewhere she seems to use the two words ~ithout distinction (pp. 259,96-97, 168), and she says that emotion is "a way of being affected" (p. 77). Emotions are also "sensitivities" (p. 77), and they also give "reports" (p. 250). "Emotion," she says, "is not a univocal thing" (p. 51), and that seems fair enough, but it would often be good to know what thing is in play in a given context. One troubling feature of Sherman's treatment is that she almost invariably speaks of emotions as an undifferentiated class, \\ithout noting that certain points may hold for some emotions but not for others. The occasions on which such differentiation might have been helpful seem to me almost innumerable. When she says, for example, that cultivation of emotions that are "receptivities" might have "moral worth" (p. 125), one needs to know whether this really holds for all emotions or only for some. For even if it were granted that all emotions are "re­cepti\ ities" to morally salient features-which it might well not be-it would not immediately follow that they would all have value, let alone moral worth. More­over the "moral worth" of emotions that are not such recepthities could still be called into question. It hardly seems right in these contexts to speak of emotions en bloc. I shall recur to this point later. Moreover, Sherman draws no explicit distinction between an emotion as a standing state and an emotion as a particular episode, for example, a particular reaction to a particular situation. She also says nothing specific about the differ­ence between treating either kind of state as a response (to speak loosely) either to a particular event or state as particular or to a general characteristic of states of affairs. She sometimes appears to allow us to think of emotions as responses to "kinds of actions and circumstances" (p. 269, "ith my emphasis). This failure to differentiate casts some doubt on her attempt to use cases in which (she main­tains) Aristotle invokes emotions to support her ascription to him of "particular­ism" (p. 251 and in general chap. 6) and her' claim that "rules" play only "a limited role" (p. 284) in his \iews on deliberation. It is especially unhelpful that here, as elsewhere, Sherman does not link her discussion in any detailed way to long-standing scholarly controversies about how to interpret Aristotle on this point. Another obscure matter, philosophically and also for the interpretation of Aristotle's view, is the connection between emotions and tendencies to feel plea­sure and pain. Sherman often indiscriminately refers to "emotion," "pleasure and pain," and "moral sentiments (pity, fear, hope, anger, and, in general, plea­sures and pains in that [?] sense)" (p. 205). As for Kant, it is confusing that Sherman seems, "ithout any discussion, to treat the Kantian notion of "respect" as a "feeling," which is "responsive to per­sons simply as persons," and as "the moral emotion" (p. 181, her emphasis). It does not seem at all clear that Kant thinks of "respect," either for persons as ends or for the moral law itself, as an emotion or for that matter a feeling at all, though whether he would have been justified in not thinking of it thus is of course prob­lematic. It does seem clear, however, that this issue deserves more discussion in the context in which Sherman takes it up. This is not, however, the main point on which Sherman tries to show that Kant pays attention to the role of emotions in ethics. Rather, she focuses mainly Copyright © 2000. All rights reserved. Book Reviews 191 on the idea that, according to Kant, certain emotions "support" a tendency to d:> one's moral duty (p. 126) and we accordingl), have a dut), to cultivate those ver), emotions (pp. 134-35). Here, however, another question arises. 1 do not think anyone at alJ would den)' that according to Kant, emotions can and indeed need to support moralit), in some sense. This is at the ver), least because some emotions can interfere \,;th our doing our duty, since they involve "inclinations" that can work against our doing it or even being aware of it. In that case, however, other emotions, which work against those duty-inhibiting emotions, seem to need cultivation, as indeed Kant indicates. But-I have already aIJuded to this point-it is not clear that this makes alJ emotions worth cultivating, rather than only some. Thus it is not clear that Kantian thinking should here broadly endorse "the cultivation of our pat;­sional selves," as Sherman seems to contend (p. 125). In this connection, it would have been instructive to read a discussion of whether and how, according to Kant, one emotion or inclination can be, so to speak, enlisted by reason against another duty-inhibiting emotion or inclination. And of course a discussion of the relation between emotion (in whatever sense) and inclination, in Kant's sense or one like it, would have been helpful, though Sherman gives no attention to the idea of inclination itself. Related to this issue is the fact that the kind of value thus far observed to attach to emotions, whether some or alJ of them, is only instrumental value. As Sherman notes (p. 158), this is quite different from the claim that emotions are ever "intrinsicaIJy valued." I cannot see, however, that Sherman ever makes a comincing case for this latter claim, though she says that at some points that "Kant's language seems to suggest" it (p. 159). Here she appears to me to be OIl weak ground, since, as she acknowledges, his "self-conscious" ,iew is that the value of emotions "is always conditional, supporting rather than grounding mo­rality" (p. 123). Elsewhere she argues for her interpretation on the basis of what seems to be a false contrast, maintaining that because Kant does not "merely tolerate" emotions, he makes them "a part of human perfection" (p. 182). StiIJ, it seems fair enough for her to have stressed the instrumental value of emotion:1 in Kant, even though that idea is not unfamiliar. It would also have been interesting in this connection to have a far mort: explicit and extensive treatment of benevolence. Perhaps this word does not des­ignate an emotion, but it often seems to designate a feeling, and, as noted, Sher· man sometimes treats these as the same. Moreover, benevolence seems to have at least as much right to be caIJed an emotion as respect does. It is also associated with 'affect'. 'Benevolence~ does not appear in the index (though 'Clinton, W: does), and \\;th casual effort I noticed the word onl)' once (p. 359). Now, in the Grundlegung, Kant, famously or notoriousl)" contends that benevolence is not (to put it in a compressed way) a moral motivation, even though he notes that it might encourage moraIJ)' right actions. Surely this matter should have been treated. It raises prima facie difficulty, at the least, for the contention that Kant counts as "a part of human perfection" at least this emotion (if indeed benevo· lence is an emotion), and perhaps even any emotion at all. Here it is also relevant to recalJ the distinction that I drew earlier, between expounding Kant's own ideas, expounding the \;ews of the traditions that have developed his ideas, and shm,;ng how one might develop them using one's own philosophical resources. What Sherman mainly wants to show, I suspect, is thaI 192 Ethics October 2000 emotions would figure in certain ways in a philosophically defensible develop­ment of Kantian thinking. If so, then she may well have spent too much energy trying to find that development in Kant himself. Sherman's neglect of issues about the notion ofrespect is connected with a neglect of other issues having to do with the notion of moral obligation. Sherman tends to focus on whether Kant's treatment of "morality" is accurate and suffi­ciently broad. She maintains that it places too much emphasis on "rules" and­though she maintains that this charge is frequently exaggerated-too little em­phasis on emotions. Now these criticisms should be assessed in the light of assumptions about what Kant's theoretical aims were or what a Kantian's theoretical aims should be. One might concede that if Kant's aim is to describe morality as a whole, or even more broadly ethics as a whole, then his approach is in some ways limited and distorted in both respects: too much about rules and perhaps not enough about emotions as well as other things. On the other hand, if his aim is narrower, namely, to describe and explain moral obligation (such a thing being presumed to exist), then one might saYlthat Kantians are right to emphasize rules and to deemphasize emotion. A disagreement between anti- and pro-Kantians would then be taken to concern, not whether morality is rightly depicted, but rather whether moral obligation,-agreed to be itself rightly depicted, is as important within morality or ethics, as t whole, as Kant or certain Kantians present it as being. Thus perhaps ethics as a wh.ole is less austerely or severely obligation­oriented than, but nevertheless contains as a component, moral obligations which are every bit as severe and auste;e, in their place, as the most astringent reading of Kant pictures them as being. The book raises interesting issues and contains some suggestive discussions, especially (I think) some interesting comparisons between Aristotle and Kant in chapter 7. I hope that Sherman goes on to sharpen the formulations of the issues and the articulations of positions concerning them, and that she pursues some relevant exegetical questions further. NICHOLAS WHITE University of Utah