Sensibility Theory and Conservative Complancency

In Ruling Passions , Simon Blackburn contends that we should reject sensibility theory because it serves to support a conservative complacency. Blackburn’s strategy is attractive in that it seeks to win this metaethical dispute – which ultimately stems from a deep disagreement over antireductionism – on the basis of an uncontroversial normative consideration. Therefore, Blackburn seems to offer an easy solution to an apparently intractable debate. We will show, however, that Blackburn’s argument against sensibility theory does not succeed; it is no more supportive of conservative complacency than Blackburn’s noncognitivism. A victory for noncognitivism cannot be so easily won. Simon Blackburn draws the traditional noncognitivist distinction between fact and value in terms of what he calls an input-output function. 1 According to the noncognitivist distinction moral judgments don’t represent moral facts, but rather express moral proor con-attitudes. 2 On Blackburn’s input-output rendering of the distinction, inputs are representational mental states with descriptive direction of fit: they represent ways one takes the world to be. Thus inputs are states such as beliefs and perceptions. Outputs are also representational mental states, but by contrast have directive direction of fit: they represent ways one would like the world to be, and so express (positive and negative) goals. Outputs, then, include desire-like states. 3 Accordingly, Blackburn’s noncognitivism holds that moral judgments are classified as outputs; they are a kind of directive state which he circumscribes as stable concerns. 4 Blackburn has long defended this position against John McDowell’s sensibility theory, a form of moral realism which claims that moral judgments are representational mental states with both descriptive and directive SENSIBILITY THEORY AND CONSERVATIVE COMPLANCENCY 545 © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. aspects. 5 Identifying moral judgments as hybrid states – called besires in J. E. J. Altham’s useful terminology – allows McDowell to show how a theory which holds that moral judgments represent moral facts can also make sense of the intuition that moral judgments are inherently motivational. For while states that are merely descriptive are motivationally inert, besires, in virtue of their directive aspect, are motivationally active. Furthermore, McDowell argues that the possibility of such hybrid states has been overlooked due to adherence to a crude sharp division of mental states into beliefs and desires, or, more generally, descriptive and directive. This division, claims McDowell, is founded on the wrongheaded Humean distinction between fact and value, and should be superseded. 6 McDowell’s reasons for thinking that the fact-value distinction is wrongheaded come, in turn, from an antireductionist worry that there is no absolute conception of reality, that is, a characterization of reality such as the characterization science claims to provide, which avoids values and norms altogether. 7 Blackburn’s reductionism, by contrast, holds that values and norms can be characterized in terms of directive states which can be understood in nonnormative terms, namely, in causal terms. 8 Ruling Passions offers Blackburn’s most sustained attempt to defend a sharp distinction between descriptive and directive states. Blackburn argues there that old fashioned Humeanism prevails over newfangled sensibility theory on the basis of an uncontroversial normative consideration. He contends that McDowell’s proposal that moral judgments are fused descriptive-and-directive besires serves to support a “. . . conservative and ultimately self-serving complacency”, 9 where conservative complacency involves a dogmatic resistance to moral criticism. However, Blackburn maintains, noncognitivism’s sharp division between descriptive and directive states provides the resources to counter conservative complacency. Thus noncognitivism prevails on uncontroversial normative grounds. Blackburn’s strategy is attractive in that it seeks to win this metaethical dispute – which ultimately stems from a deep disagreement over antireductionism – on the basis of a normative consideration that anyone could accept. Therefore, Blackburn seems to offer an easy solution to a debate which appears to be intractable from the standpoint of metaphysics. However, while we are in general agreement with Blackburn’s noncognitivism, we think that the suggestion of an easy solution should be met with suspicion. Arguments that purport to bring resolution to deep metaphysical disagreements are typically deflected with the charge of begging the question, and the deep disagreement continues unresolved. 10 We’ll show that Blackburn’s argument against besires from considerations about conservative complacency provides a case in point. 11 We’ll first lay out the argument. We’ll then demonstrate that McDowell’s sensibility theory has the same general resources for countering conservative complacency as noncognitivism. Finally, we’ll diagnose Blackburn’s underestimation of 546 PACIFIC PHILOSOPHICAL QUARTERLY © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. McDowell’s resources for countering complacency as the result of ignoring a crucial aspect of McDowell’s antireductionist view. A victory for noncognitivism cannot be so easily won. Blackburn’s argument against besires Any account of moral judgment provides roles for states that are descriptive and states that are directive. The issue dividing Blackburn and McDowell is whether descriptive and directive states can be sharply separated, as Blackburn’s proposal of an input-output function contends, or whether there are some states with both descriptive and directive aspects, as McDowell’s proposal of besires claims. Since the idea of hybrid descriptive-and-directive states is unfamiliar, an initial concern is whether we can make sense of characterizing moral judgments as such hybrid states. Descriptive states such as beliefs represent ways one takes the world to be. Accordingly, such states tend to go out of existence in the face of evidence that their contents are false. By contrast, directive states such as desires represent ways one would like the world to be, and don’t tend to go out of existence in the face of evidence that their contents are false. In view of this characterization, how can a single state be both directive and descriptive? It would seem that a state which is both would both tend to and not tend to go out of existence in the face of falsifying evidence. 12 This concern is not damaging, however, because besires can be characterized in terms of more than one content; moreover, they are descriptive relative to one content and directive relative to another. For example, taking any moral judgment of the form φ -ing is right, this state describes a situation in which φ -ing happens as right, and so is descriptive. As well, the state holds φ -ing as a goal, and so is directive. 13 Allowing that we can make sense of such hybrid states, Blackburn contends that the besire account can be normatively faulted. Blackburn argues that according to McDowell’s theory of the nature of moral judgment, moral judgment has a feature that allows one to dogmatically resist moral criticism; while noncognitivism, rejecting this feature, has the resources to counter such resistance. 14 The feature in question is the putative descriptive aspect of moral judgment. According to sensibility theory, moral judgment involves an inextricable meld of a quasi-perceptual detection of a moral property and a moral goal, thus fusing description and moral attitude. Blackburn contends, however, that accepting this meld serves to support complacency. At bottom, Blackburn’s argument seeks to show that any form of moral realism that appeals to a quasi-perceptual capacity is subject to an objection targeted at Moore’s intuitionism. The realist characterization of SENSIBILITY THEORY AND CONSERVATIVE COMPLANCENCY 547 © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. moral judgment as descriptive indicates that we “just see” genuine moral features in the world. But in this case, the realist characterization has the consequence that we can’t angle ourselves into a standpoint from which we’re able to consider whether a situation nonmorally described (input) deserves a certain moral response (output). Rather, if we “just see” genuine moral features, we seem to have a grip on reality that allows us to be resistant to moral criticism: we can rebuff moral criticism as blind to certain genuine features, and thus ignore it. 15 Consequently, by providing this “perfect defense” against criticism, 16 sensibility theory provides us with a reason to be close-minded. As Blackburn makes clear, the point isn’t that sensibility theory indicates that changing a moral transgressor’s mind is impossible: “Perhaps McDowell can admit that, as a bare fact, when we declaim against the [moral transgressor] as suggested, this may influence him for the better.” 17 Presumably the moral transgressor’s mind could be changed through means apart from rational argument – through threats or brainwashing, for example. Rather, the point is that sensibility theory provides us with a reason to be resistant to criticism. Thus attempts at changing the moral transgressor’s mind run up against an obstacle: a reason – the critic’s blindness – for the transgressor to stand pat. It is in this particular way that sensibility theory’s proposal of besires serves to foster a conservative complacency, a fault which apparently is inherent in quasi-perceptual accounts of moral judgment. Furthermore, McDowell motivates the besire account of moral judgment on the basis of an analogy between color and moral properties, according to which moral properties, like colors, are dispositions of objects to produce a certain psychological response. 18 However, the analogy with color can be taken to open the besire account to the charge of conservative complacency. We are complacent in our color attributions, and for good reason. We rarely face criticism of our color attributions. And in the rare case where we do – in some borderline case of yellowish green versus a green with no yellow at all – we assume that we cannot angle ourselves into a standpoint from which we can consider whether a surface physically characterized is really yellowish green versus a green with no yellow. In fact there is no known basis in the physical properties of surfaces which settles the issue. 19 Thus, it seems, with respect to some of our color attributions we have a reason to be resistant to criticism; we just see a surface as yellowish green. Thus we can reject criticism as blind to this genuine feature, and ignore it. Certainly a metaethical theory which serves to foster complacent resistance to moral criticism can be faulted on uncontroversial normative grounds. And Blackburn seems to demonstrate that a perceptual model of moral judgment is vulnerable to this decisive objection. Thus, Blackburn seems to show that the metaethical theory McDowell establishes on the basis of the analogy is not, in the end, sufficiently different from Moorean intuitionism to be a tenable option. 548 PACIFIC PHILOSOPHICAL QUARTERLY © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. Why Blackburn’s argument fails In considering Blackburn’s argument, however, it’s important to make clear that while McDowell motivates his proposal about moral properties and moral judgment on the basis of an analogy with color and perceptual states of color, he limits the analogy to some specific points about how a property can be both mind dependent (in one sense) and mind independent (in another sense). In particular, holding that a property is a disposition of objects to produce a certain psychological response fills out how this can be so. Such a property is both mind dependent (being in part constituted by a psychological response) and mind independent (being a dispositional property, and thus not requiring activation for its existence). 20 Yet, the analogy between color and moral properties, being merely an analogy, allows for important differences between these sorts of properties. Consequently, use of this analogy doesn’t force McDowell to accept that our complacency with respect to color attributions is paralleled by a similar complacency with respect to moral property attributions. While this general point about the analogy goes some distance in responding to Blackburn’s argument, it doesn’t go far enough. The question remains as to whether, on closer scrutiny, McDowell’s quasi-perceptual account of moral judgment, through its claim that moral judgment incorporates a descriptive aspect, does serve to support dogmatic resistance to moral criticism. Let’s look at an example, namely Blackburn’s example of a shy newcomer to a boarding school who is teased by a resident bully. 21 Given the circumstances we criticize the bully with the moral judgment that teasing shy people is wrong (or cruel). 22 According to McDowell, our moral judgment is a besire formed by a moral sensibility. The besire in question is an inextricable meld of a quasi-perceptual detection of a moral property (the wrongness of teasing shy people) and a moral attitude which provides moral motivation (in which the ending of teasing shy people is taken as a goal). Thus, in virtue of our besire we “just see” that teasing Jimmy is morally wrong. However, the bully has a different besire, involving a quasi-perceptual detection of a different moral property (the moral appropriateness of teasing shy people). The bully “just sees” that teasing Jimmy is morally appropriate. This “seeing”, Blackburn contends, involves “ lock[ing the bully] into some barely interpretable amalgam of besire.” 23 Thus, we and the bully are at odds with regard to how the situation is even described, and so it seems that the bully has a “perfect defense” against our criticism. If we can’t “just see” genuine moral features that he can, so much the worse for us. It seems that the bully has a reason to be resistant to criticism; he has carved out a space for conservative complacency. To determine whether sensibility theory is at a relative disadvantage on the score of conservative complacency, however, we need to consider a SENSIBILITY THEORY AND CONSERVATIVE COMPLANCENCY 549 © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. sensibility theory’s resources for countering it. With regard to the bully, chances are the bully shares our moral sensibility. 24 In this case, the bully in fact has the general besire that teasing shy people is wrong, but this besire, when he is faced with newcomer Jimmy, is overridden by other conflicting attitudes. 25 The bully just needs a good talking to for this sort of overriding to cease. But, sadly, we can imagine a circumstance where the bully simply doesn’t have the appropriate general besire, unusual as this circumstance is. Presumably, this is the difficult case that Blackburn has in mind, where sensibility theory seems to give the bully a reason to resist criticism. The sensibility theorist’s initial response is to point out that while besires are products of sensibilities, a moral sensibility is, in turn, a product of culture (and to some extent physiology), and is culturally malleable. Thus, we can try to get the bully to share our sensibility through acculturation (which, on McDowell’s view involves training which develops conceptual resources). If we’re successful, we can get the bully to form the besire that teasing shy people is wrong; we can get him to “see” that his behavior is wrong. Moreover, the sensibility theorist can contend that the noncognitivist’s resources for countering complacency are not so dissimilar as to convey an advantage. In characterizing an input-output function, Blackburn focuses our attention on the beginning and end points of this function – on a nonmoral description of a situation and our response to it in the form of moral judgment. Of course, this emphasis serves his purpose of rendering the traditional noncognitivist fact-value distinction. But this emphasis also obscures a point which Blackburn states but doesn’t stress enough: the transition between input and output is not a “. . . . simple chute or conveyor belt whereby we mechanically or automatically find some things generating some responses”, 26 but rather is mediated by a complex set of background states. So, for example, between the input of representing little Jimmy as easily frightened and hesitant to respond to aggressive treatment and the output of the specific moral judgment that teasing Jimmy is wrong, there are more general moral goals, such as that the teasing of shy people ends, and the suffering of all people ends. These goals are more general in the sense that they apply to many specific situations, and so combine to form a transition between input and output. When we consider the noncognitivist’s resources for countering complacency we must look to the transition. With regard to the bully, chances are that he shares our more general moral goals. But again, we can imagine a circumstance where the bully simply doesn’t have the relevant more general moral goals. Nevertheless, according to Blackburn such goals are products of culture (and to some extent physiology), and are culturally malleable. Thus, where the sensibility theorist talks about modifying a moral sensibility, the noncognitivist talks about modifying such general goals. 550 PACIFIC PHILOSOPHICAL QUARTERLY © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. On the face of it, the sensibility theorist’s initial response is successful. The two theories indeed have the same general resources for countering complacency. According to both sensibility theory and noncognitivism, complacency is countered by conveying certain moral goals. Furthermore, for both theories these goals are understood as products of culture. And these points seem to defuse Blackburn’s worry that sensibility theory prevents us from angling ourselves into a standpoint from which we can consider whether a situation nonmorally described deserves a certain moral response. For the prevention of such angling only matters with respect to complacency insofar as it hinders us from conveying certain moral goals. But since both McDowell and Blackburn hold that moral goals are products of culture, their views have the same general resources for conveying moral goals. However, this initial response still doesn’t go far enough in addressing Blackburn’s argument. Blackburn’s charge of conservative complacency stems from sensibility theory’s incorporation of a descriptive aspect of moral judgment by which we can “just see” genuine moral features. And this putative descriptive aspect by which we represent genuine features sets up a particular obstruction to conveying moral goals. Thus, even if sensibility theory and noncognitivism have the same general resources for conveying moral goals, these views are not on a par because besires’ supposed grip on reality provides us with a reason to be close-minded. And the idea that we “just see” features that are “genuine” does seem to give us the sort of grip on reality that would justify close-mindedness – at least this is a natural understanding for the reductionist. Thus the charge of conservative complacency may be seen to have a point from the perspective of Blackburn’s reductionism. But this argument fails. For, in line with his antireductionism, McDowell has a radically different conception of genuine features. 27 According to this conception, features of objects or situations are genuine so long as they are mind independent in the sense that they are dispositional properties, and thus don’t require activation of a psychological response for their existence – the sense in which McDowell holds that both colors and moral properties are mind independent. On this conception, the idea that we “just see” features that are “genuine” isn’t a matter of gaining access to features that are mind independent in the strong sense Blackburn suggests in his comment that the bully is “locked into” a besire, and so locked into a characterization of reality. To the contrary, since McDowell’s antireductionism rejects an absolute characterization of reality, the idea of locking into a characterization of reality has no place in McDowell’s view. 28 Rather, characterizations of reality are as malleable as the sensibilities that tune into reality. Thus “just seeing” genuine moral features doesn’t give us the kind of grip on reality that would justify close-mindedness. Indeed, built into McDowell’s antireductionist conception of genuine SENSIBILITY THEORY AND CONSERVATIVE COMPLANCENCY 551 © 2005 The Authors Journal compilation © 2005 University of Southern California and Blackwell Publishing Ltd. features is a malleability of “seeing” formed by training which develops conceptual resources that open us to new features in the world – as for example the training which develops the conceptual resources that opens us to “hearing” jazz as more than mere noise. 29 Furthermore, the debate between Blackburn and McDowell is of interest largely because it stems from a deep disagreement over antireductionism. McDowell’s antireductionist conception of genuine features is not an idiosyncrasy of McDowell’s version of sensibility theory. A sensibility theory isn’t necessarily antireductionist. However, its basic claim that moral properties are dispositions of objects to produce certain psychological responses accommodates antireductionism, for such psychological responses may be characterized in terms of an irreducible subjectivity. And antireductionism is part of what makes sensibility theory attractive to some ( just as antireductionism is part of what makes the standard version of color dispositionalism – which characterizes color experience in terms of irreducible subjectivity – attractive to some). Consequently, McDowell’s conception of genuine facts can be used to defend sensibility theories apart from McDowell’s version. Sensibility theory is no more vulnerable to the charge of conservative complacency either from the perspective of general critical resources, or from the standpoint of particular obstructing reasons. Blackburn’s charge ignores McDowell’s alternative conception of genuine features. We diagnose Blackburn’s underestimation of McDowell’s resources for countering complacency as the result of ignoring this crucial aspect of McDowell’s antireductionist view. Blackburn’s argument fails because it tacitly appeals to a question-begging conception of “genuine features” which incorporates the idea of locking into a characterization of reality. The debate then cycles back to an apparently intractable issue in metaphysics. Blackburn’s argument against besires notwithstanding, there’s no easy victory for noncognitivism. 30 Department of Philosophy California State Polytechnic University, Pomona