Code: DP-7

Emotion, Judgment and Autonomy: Buchlerian Resources for Addressing Feminist Concerns

I. Agency precludes emotion

In"The Conscience of Huckleberry Finn," Bennett draws an intriguing comparison between Huckleberry Finn, Heinrich Himmler, and Jonathan Edwards.(1) None is a woman and there are no feminist issues explicitly raised by Bennett. However, some issues that have been central feminist concerns -- such as the role of emotion in moral judgment and the conceptualization and valorization of autonomous agency -- are so vividly drawn in this article, at least as I understand it, that I have decided to use the article as a touchstone for the following discussion. I am going to focus on the contrast drawn between Huck Finn and Heinrich Himmler. Bennett focuses on the incident from the Twain story when Huck is on a raft with his friend and fugitive slave, Jim. Huck Finn is torn between his belief that Jim is someone's property and therefore, ought to be returned to his owner and his feelings of sympathy and friendship for Jim, who communicates to Huck his joyful anticipation of freedom and his gratitude to Huck for helping him as a friend. When Huck has to respond to the white slave hunters who call to him from another boat -- whether his (Huck's) companion his white or black -- Huck responds "white," while at the same time expressing the thought that he has consigned his soul to hell and has abandoned morality. Huck feels and acts on sympathy for his friend, the slave Jim, while believing he is wrong and "weak" for doing so. Bennett juxtaposes the case of Heinrich Himmler, the Nazi SS commander, to that of Huck. According to Bennett, Himmler (allegedly) feels sympathy for his victims and asserts that one ought to feel sympathy in order to remain "human" in doing what is otherwise one's duty. Duty is defined by adherence to principles that one self-consciously endorses. Himmler does his "duty" and exhorts others, too, as well, in spite of whatever sympathy and compassion he and they might feel for their victims.(2)

Bennett argues that while Huck, in fact, may have dones the right thing, he has failed in moral agency. Even if his moral principles (conscience) were "bad" or ill-[in]formed, in abandoning them for sympathy, he failed in some important way, in moral agency itself. In contrast, Bennett argues, Himmler's moral principles, like Huck's, are "bad" but in acting on them, in feeling himself committed to them as guidelines for action, Himmler does not fail in being a moral agent. With Himmler "morality," albeit "bad morality," wins out over sympathy and Himmler retains moral agency and Huck does not. Bennett goes on to suggest that the moral function of sympathy is to prompt reevaluation of principles. Thus, Huck and Himmler are similar in that each fails morally (neither uses sympathy to revise principles), and each has the potential to improve his morality (because each has sympathy). But on Bennett's analysis only Himmler judges as a moral agent, albeit a bad one, because Himmler judges and acts on principles rather than feeling.

What has always bothered me about Bennett's analysis is that Himmler gets characterized as making a mistake, but he is still a moral agent. Huck, on the other hand, gets characterized as failing to be a moral agent or judger altogether, because he acts on feeling and against what he professes to believe as a moral principle. This same kind of criticism is levelled against women, or anyone who acts from feeling, e.g., sympathy, benevolence. The criticism is that the person is not even making a moral judgment -- because feeling is not a legitimate moral interest; the moral failure therefore is on a different order from that of Himmler. Himmler just needs to re-think, is educable, about moral principles, whereas Huck needs a major overhaul as a person, in order to be a moral agent at all. On this view, Himmler is just lacking knowledge of the right principles and once he had that, he would be a (good or correct) moral agent. But Huck is presumably lacking moral agency altogether, in so far as he lacks the ability to identify and act on a principle. Himmler fits in to classical Socratic view, that virtue is knowledge and vice is ignorance.

But, why isn't Himmler's failure to act on feeling a failure of moral agency and judgment? Why isn't Himmler's apparent inability to be guided by a sensitivity to the experiences of others a failure of moral agency? The reason is that Bennett's categorizations rely on Kantian and well-entrenched assumptions about moral agency, agency that is defined in terms of rational, self-conscious articulation and endorsement of moral principles. Now, I am not going to argue that that shouldn't count as moral agency. Rather, my claim is that as the only game in town, it constitutes a narrow view of moral agency, one that fails to recognize the scope of moral judgment and that has an artifically restricted vision of what agency and autonomy consist in.

The Kantian claim that moral judgment and agency rest on reason derives its force from a metaphysics which sharply distinguishes between the interests of pure reason and those of experience, feeling, life in general. Only the former can form the basis of genuine autonomy and morality; the latter are always by definition, heteronomous, and are at best prudential rather than moral. An appeal of this approach, is that it provides a clear way of differentiating between "merely" determined, caused behavior -- emotion falls into this category -- and choice as free because "purely" rationally determined. In so doing, it provides grounds for holding persons responsible for choices and characterizing them as autonomous, self-determining beings. However, the Kantian metaphysical framework is not subscribed to, not even by most contemporary Kantian moral theorists. And assuming there are good reasons for abandoning that metaphysical framework, then, having done so, it seems to me that the conception of moral judgment, agency and autonomy as necessarily determined by reason and reason alone has to give way to a reconceptualization of reason or, more broadly to a more generous conceptualization of judgment, agency and autonomy. While the alternatives are not mutually exclusive, to start out accepting the distinction between reason and emotion, and then reconceptualizing them, is starting out with loaded terms. Rather, what is needed is an account of judgment that will allow emotion itself to count as judicative and therefore, as central to agency and autonomy, no less so than reason has classically been thought to be.

I have argued elsewhere and in greater detail that the Buchlerian theory of judgment provides the kind of account that would begin to do justice to the range of behavior and choice that ought to count as moral judgment.(3) I will briefly summarize the main features of the account of judgment that would accommodate emotion as judicative. I will then turn to making some suggestions for how agency and autonomy might be rethought following the lead of this account.

II. Emotion as Judicative

Taking off from the pragmatic expansion of belief as including active and emotional values, Buchler takes judgment beyond its association with belief (NJ 32). (4) Judgment is an "actualization of a relation between the individual and some natural complex, a relation consummated by [the individual]" (NJ 10). For Buchler, judgment is "done" not by the mind or reason or any other particular faculty but by the individual, the whole self. Second, judgment is constitutive of a self; it "actualizes a relation" of the self. The basic question then is, what kind of relation or actualization is judgment?

To judge is to take a position, to produce by "manifesting the natural commitments of a self" (NJ 11), to "appraise" and "pronounce" on some phase of the individual's world (NJ 12). To judge is to define something by bringing it into relation to us (NJ 13). Every judgment is "perspectival."

When we judge we are partially determining the properties of a sphere of existence within a given perspective--from a given "position." We combine and select. (NJ 15)

Judgments are both manifestations of the self and discriminative of a subject matter. Judgments are not necessarily intellectual or even self-conscious, although they can be.

The man who takes a short cut on his walk home is ipso facto making a judgment...with respect to means that fulfill his habits, desires, or needs. He is devising or applying a technique that arises out of what he is and what he has been. To say that in taking the short cut he is making a judgment does not mean that he is asserting to himself what goals the action will accomplish. It is his action that is the judgment. He may, in addition to his action, represent the action verbally. But whether he does so or not, the action subserves the same function. It is as much an expression by him as the assertion....It is of course customary, both popularly and philosophically, to apply the term "judgment" to the assertion. "He judged the shortest way home" is supposed to mean "He said to himself, this is the shortest way home." But to limit the usage is to limit the analysis. If the man walking home were completely preoccupied with other matters and took the shortest path automatically, this habitual action would still, fully as much as the verbal representation, embody a policy relating him to his environment and to his own past history, and characterize the existences among which he is located. (NJ 11-12, italics added) (5)

The theory aims to avoid three misconceptions about judgment: (1) that it is antecedent to action or making; (2) that it is a specific operation, typically mental, of a special faculty (e.g., "reason," a "moral sense"); (3) that it takes the form of thinking, venturing truth claims, and is essentially propositional in form (ML 93-95).

Judgment occurs in three basic modes: assertive, active, and exhibitive. We judge assertively when we "predicate, state, or affirm, by the use of words or by any other means; when the underlying direction is to achieve or support belief; when it is relevant to cite evidence in behalf of our product"; actively when in doing or acting "the underlying direction is toward effecting a result; when `bringing about' is the central trait attributable to our product"; and exhibitively "when we contrive or make, in so far as the contrivance rather than its role in the action is what dominates and is of underlying concern; when the process of shaping and the product as shaped is central." Finally, "[o]n the methodic level...assertive judgment is exemplified by science, or more generally, inquiry (including any discipline that makes a truth claim); active judgment, by deliberate conduct morally assessable; exhibitive judgment, by art" (ML 97-98).

Anything--a sentence, a gesture, an action--may function as a judgment in one or more mode (ML 98-99). (6) For example, an action may function as an assertive judgment, a linguistic utterance as an active judgment. (7) When Billy Budd, whose stammer renders him temporarily unable to speak, strikes the master-at-arms (Claggart), who falsely accuses Budd of mutiny, Budd's action is an assertive judgment; it is a denial of Claggart's accusation. (8) Claggart's knowingly false accusation, on the other hand, is an active judgment, part of an explicit attempt to destroy Budd. Captain Vere's subsequent silence during the drumhead court that condemns Billy is also an active judgment aimed to effect a result (suppression of a perceived threat of mutiny).

If silence, a nod, a purchase or any other piece of behavior can function as a judgment, then so too can feeling and emotion. "A feeling that functions appraisively is a judgment" (NJ 149-50).

All three of the modes of judgment may generate feeling, either in their producers or in others....All three may entail feeling on other levels. A feeling may function as an event, a concomitant of judgment among other concomitants....It may function as an aspect of judgment...or as the guiding tone whose relative intensity and fluctuations help to determine force of judgment and change of judgment. (ML 99-100)

What is distinctive about judgment is how it reveals and commits the self, not that it is distinctively assertive or linguistic. All three modes are judicative and equally fundamental--hence the insistence "on the relevance of all three functions in any adequate account of the human process and its products" (ML 100, italics added). Buchler's tripartite theory of judgment provides conceptual language for articulating how other modes of judgment besides the assertive function, discriminate, and actualize the self.

Thinking back to our starting point, we saw that in a classical, e.g., Kantian analysis, the case was prejudiced against Huck as moral agent and judger. We can now say, using Buchlerian categories, that this is because "rationality" was identified with linguistic or formally symbolic assertive judgment (a narrow view of assertive judgment and of rationality) and because judgment was equated with assertive judgment (a narrow view of judgment). (9) But if there are three equally fundamental modes of judgment (10) and if the range of behavior that is judicative (in one or more mode) is not confined to linguistic assertion assessable as true or false, then we are in a much improved position to interpret as judgments conduct that in traditional theories is assimilated to an inappropriate model, remains uninterpretable, or is classified as humanly inferior.

Let us now examine how Huck's judgment might be reinterpreted using the Buchlerian theory of judgment. Huck is characterized by Bennett as giving up morality for sympathy and hence as not acting as a moral agent. But from the Buchlerian framework Huck's action and experience can look quite different: His sympathy discriminates features of the situation (the consequences to Jim of being caught, the nature of Huck's relationship to Jim) and weights them in relation to choice. Huck makes an active judgment: his action itself is a judgment and not only his thoughts about his action. His sympathy is directed toward effecting a result and it, as well as Huck's action, may be a component in Huck's broader project in the novel of reevaluating morality. His sympathy may also function as a sign that exhibits, even if only to or for himself, the conflict between his feelings for Jim and the conventions and rules that Huck had taken to be principles. Thus sympathy may also be an (exhibitive) phase in the larger project of reevaluating principles and the very meaning of morality.

In Huck's case it is actually ambiguous what his genuine motives or interests are. Huck is feeling, undergoing, an inconsistency; he cannot resolve his sympathy with his "principles." In abandoning them at the moment of choice, he feels he has abandoned morality (and Bennett agrees). But on the Buchlerian theory the situation is more complicated. Huck's self-reflective judgment (probably an assertive judgment) is not the only judgment he is making. If his action itself is a judgment, it is not necessarily the case that he has abandoned morality, his own self-conscious reflection notwithstanding (see also ML 98). He acts to effect a felt good for Jim, and to actualize (recognize in action a feature of) his relationship (of friendship) with Jim. He characterizes himself as "akratic," as weak-willed, but it seems to me that he also exemplifies considerable courage in accepting what he believes to be the consequences of his action (damnation of his soul) in opting to support the felt good for another and for a relationship rather than the principles with which they are in conflict. It seems to me we have missed something important in morally assessing deliberate choice and conduct if our notion of moral judgment doesn't see this.

In Himmler's case, he certainly is to be faulted for failing to know better, for using his rational powers in evil and perverse ways. These are Kantian terms for condemning Himmler and the Buchlerian theory of judgment would allow for such moral condemnation of Himmler. However, the Buchlerian theory also allows room for condemning Himmler for failing to follow his (alleged) feelings of sympathy and to act against his principles, even if, like Huck, he didn't have a clear articulated version of what a better principle would be. Himmler is condemnable not only because he deliberately acts on evil principles and intends to do harm to people; he is equally condemnable for deliberately resisting an alternative basis for conduct, namely, sympathy for the feelings and experiences of others. Bennett's analysis leaves us with the conclusion that Himmler failed because he had "bad principles" (so did Huck in believing in slavery) and because he did not use his (alleged) sympathy to reevaluate his principles (neither did Huck, at least not in the immediate situation), but that Himmler would have been worse as a moral agent had he, like Huck, abandoned his principles without reevaluating them and simply acted on sympathy. Again, it seems to me that we have missed something important in morally assessing deliberate choice and conduct if our notion of moral judgment leads us to this conclusion.

On the Buchlerian theory of judgment that I have introduced here neither "reason" nor feeling is privileged, and Huck is no less engaged in judgment and hence, is no less a moral agent than Himmler. The Buchlerian tripartite theory of judgment frees judgment from its association with a narrow conception of reason and allows for parity of feeling and emotion in moral judgment. And it does so without automatically prejudicing the case against those features of moral judgment that have often been associated more with women than with men (or with a certain class of men). If there is parity between Huck Finn and Himmler as moral judgers, then there is parity between them as moral agents, and I want to say, as autonomous moral agents. This seems to contradict the standard conception of autonomy and I now want to turn to that issue.

III. Autonomy: the standard account

I have been bothered by Bennett's treatment of Huck Finn and Heinrich Himmler not only because it misrepresents the complexity and sources of morality and immorality in their respective conduct. It also reflects widely held, Kantian assumptions about autonomy and agency. It treats Huck as someone who, in abandoning principles and acting on the impulse of sympathy, has abandoned the possibility of exercising self-direction or self-governance.(11) Himmler, if not autonomous in the Kantian sense, is at least closer to being autonomous because he employs rational, self-reflective capacities in guiding his own behavior. I am not so much interested in rescuing Huck Finn as autonomous, but in exploring a more generous way of thinking about autonomy such that emotion based judgments are not automatically ruled out or relegated to second place as far as autonomy is concerned.

There are a variety of approaches in the contemporary literature to autonomy. Some characterize it negatively, that is, as a right to noninterference in defining and leading one's own life (12); others posit that the possession of certain capabilities (13) or the aiming for or realization of certain kinds of goals(14) constitute the substance of autonomy.

It seems to me that views which characterize autonomy in terms of a capability or power of some kind are on the right track. In some of these accounts, it seems that autonomy is thought of as aiming for or enacting a kind of [capacity for] self-expression; in others, as aiming for or enacting a kind of [capacity for] self-transcendence.(15) The former would seem to capture the idea that the self has some power to be self-determining whatever the sources of its desires and so on happen to be; the latter to capture the idea that autonomy also seems to suggest the idea that the self can be self-transformative, can direct itself to [at least partly self-articulated] ideals. I'm inclined to think that these two dimensions need not be mutually exclusive, but for the purposes of this paper I am not going to pursue the differences between them. I am going to suppose, though, that the suggestions I do make in what follows, will allow for both as dimensions of autonomy.

A "capability" approach to autonomy, even when not self-consciously invoking Kant, reflects a concern similar to one that Kant had, namely, how to account for the fact that human beings [seem to] have the power to be self-directive or self-governed. For Kant, such power depended on rationality and on an explicit metaphysical framework. Some contemporary "capability" views of autonomy reject the Kantian emphasis on rationality and broaden the conception of what the powers of self-direction involve.(16) No contemporary views, at least among the ones with which I am familiar, accept the Kantian metaphysical framework and its sharp demarcation of pure practical reason (as at least a hypothesis) as the world of autonomy from the causally determined natural world of desire, interest, inclination and so on as the world of heteronomy. If however, one were to accept the second piece of the Kantian dichotomy as an account of the natural world, then autonomy analyzed in terms of capacities of self-reflection would reduce to a largely psychological matter. Without a metaphysical framework ensuring the independence of reason (at least when operating "purely") or whatever the specified autonomy capacities, there is nothing that ensures that "self-direction" is not "really" heteronomous (e.g., simply a "rationalization" of desires, inclinations and so on).(17)

One could provide a meaningful account of human processes, psychologcal and other capacities, that constitute autonomy without addressing the broader metaphysical issues. However, some of those broader metaphysical issues may influence the account of autonomy. For example, an approach that argues that certain types of reflection count as "autonomous", that certain capacities for critical self-reflection and self-evaluation count as "autonomous" presumes that most of the time human beings are not autonomous. (18) Human activity is largely reactive caused by desires, or forces "outside" oneself. One responds to those felt desires, needs, impulses or forces and in doing so employs "norms" that are themselves other conditions (of culture, society, religion, etc.). In critical self-reflection on those norms, or in having the capacity for such reflection, lies the possibility of autonomy. Autonomy is something that is not only hard to achieve, but dubiously and rarely achieved.

Carrying the deterministic half of the metaphysical framework to its logical conclusion, some contemporary schools of thought view human beings as biologically, historically, socially and culturally constituted -- in post-modern terms, as locations or foci of power relations -- suggesting that autonomy is not only never achieved; it is in principle impossible. (19) If this were right about autonomy, then rational activity is no more likely to lead to autonomy than emotional or desiring activity is. Each is motivated (determined) by a set of beliefs, desires, inclinations and so on. Himmler's "principles" might be more thoroughly rationalized and thus more firmly embedded in his character as bases for action than Huck's beliefs about the rightness, say, of slavery, but Himmler would be no more autonomous than Huck. For he, like Huck, would just be expressing, albeit in more rationalized terms, his personal and cultural beliefs, desires, preferences and so on; even his alleged "self" identification with or affirmaton of the "second order" desire produced by rational self-reflection would reduce to the same deterministic conclusion upon critical analysis. Well, this would be one way to get the parity between Himmler and Huck for which I earlier tried to argue! However, the cost seems rather high if autonomy and agency have been eradicated altogether.(20)

IV. Autonomy Reconsidered

I now want to pursue some thoughts about a"capability approach" to autonomy from the point view of a Buchlerian framework. Taking autonomy to mean the capacity for and the exercise of self-direction or "self-governance," then autonomy involves the generation of norms. By "norm," I do not necessarily mean something social, moral, rational or fixed. All I mean to suggest is that autonomy means that the self has the capacity to generate guidelines ("norms") by which it directs or governs itself.(21) If autonomy as self-direction is to mean something more than a rationalization of desires, needs, and goals, then it needs a broad account of human judgment, of human powers, on the one hand, and, on the other, a background metaphysical framework or ontology in which possibility and indeterminacy are genuine features. Thus, autonomy requires a two-pronged account: one of human powers, the other, of the background ontology. Since some resources for addressing the first have already been sketched in an earlier part of this paper, I will focus on developing that in the remainder of this paper. I would like, though, to indicate what features of the Buchlerian theory of judgment itself point to the second, broader background ontology.

The theory of judgment defines judgment as the "actualization of a relation between the individual and some natural complex, a relation consummated by [the individual]" (NJ 10). If judgment is an actualization, that means that it presupposes possibility; for, every actualization is the actualization of a possibility. If the possibility is not a necessity (that is, the sole possibility), then judgment presupposes the possibility of a gap. A judgment, then, closes an opening of indeterminacy in the world and in so doing, creates, at the same time, a new gap or opening. Now, it need not be the case that every judgment does so; some judgments may in fact be actualizations of necessity. But, if possibility does not reduce to necessity (22), then the "natural world" does not preclude, by definition, autonomous judgment and action. The Buchlerian general ontology does explicitly affirm the reality of possibility in general and not only in relation to human judgment. (23) While not pursuing here the ontological grounds for this affirmation, assuming it stands up, the reality of possibility would provide a background metaphysical framework for autonomy.

I now turn to the issue of human powers. I confess to being in uncharted waters when it comes to autonomy in the Buchlerian framework. The notion of autonomy was not explicitly addressed by Buchler. Moreover, recent challenges to the notion of autonomy -- for example, from feminist perspectives -- frame a context for discussion of the issues that is very different from the historical, social and philosophical context in which Buchler thought and wrote. (24) Nonetheless, I believe there are valuable resources in Buchler's work for meeting some of those challenges and for articulating a robust notion of autonomy. In what remains, I would like to initiate an exploration, however tentative, of what some of those resources might be.

If autonomy meaning self-direction or self-governance is a human possibility, then

1) Human judgment must be a power; human processes must have the power to make a difference in shaping the contour and direction of the self. Therefore, human judgment must be more than the mere expression of other forces (biological, historical, cultural, political and so on) even if it is located in and thus influenced by those forces;

2) Human judgment must generate, be the source of, norms. This in two senses:

a) Human judgment creates (25) norms as procedures in the service of goals or results beyond judgment itself (e.g., human judgment invents methods and procedures for the production of food, for the sake of biological sustenance of the species). (26)

b) Human judgment creates norms for the sake of judgment itself, for its own unfolding; human judgment invents norms for the critique and extension of judgment itself. (27)

The two senses of norm generation are not necessarily mutually exclusive (as the distinction between hypothetical and categorical would be in a Kantian framework), for results or products of judgment may be not only results extended beyond judgment itself, but conditions of judicative norm generation. For example, in one sense, I produce this paper as a product in response to a call for papers for a philosophical society's annual meeting. In another sense, this product, both in my producing of it as well as (hopefully!) in its assimilative and critical reception by other judgers, contributes to (functions as a condition of) judicative norm generation -- specifically, to the creation and implementation of norms for philosophical judgment which guide my own philosophical judgments. (28)

In saying that autonomy involves norm generation, I am not suggesting that it reduces to the Kantian notion of moral legislation. Rather, I am proposing to relocate an analysis of what that involves to a framework which is more expansive as an account of human functioning and hence, as an account of what would count as autonomous (norm-generating and hence, self-directive/self-governing). We have seen that the Buchlerian theory of judgment is an expansive theory of what counts as judgment; it provides resources for analyzing an emotion-based choice as a genuine judgment. Perhaps the same implications can be drawn for autonomy, namely, that the theory of judgment provides resources for analyzing more expansively how judgment can be a norm-generating activity that is self-directive/governing.

On the Buchlerian account, as we saw earlier, there are three modes of judgment, assertive, active and exhibitive. Judgment is not the activity or expression of a special faculty, such as reason (versus, say, emotion or the passions). Rather, judgment is a human functioning that in any one of its modes may involve a diverse and wide-range of specific human capacities. If that is right, then there are two possible approaches one could take to autonomy: (1) that autonomy is restricted to a specific kind of judgment or to the use of specific (e.g., rational) capacities in judgment; or (2) that any mode of judgment, any specific kind of judgment, could be autonomous, that there is no a priori limitation on what kind of judgment could qualify as autonomous. The latter is the approach I favor.

Now, in what sense could an emotion or sympathy-based judgment be regarded as norm-generating and self-directing? Going back to Huck Finn, one's first thought might be that Huck has given up norms ("principles") altogether and, in turning himself over to whatever impulse is the strongest at the moment, has abandoned the possibility of self-governance. Maybe that's right about Huck.(29) But, there is another way of looking at, if not Huck, at least the kind of judgment being made. One possibility is that the "norm" being generated goes something like this: "when 'principles' conflict irreconciliably with sympathy, then abandon principle and act on sympathy." (30) I am not suggesting that this is stated as a principle, but that the judgment (the emotion based action) itself shapes a normative policy by which the judger guides him or her self. Such a "norm" is generated by the self as a[n inventive] judger which the self in action takes as its own guide. Such a norm need not be self-consciously justified in order to function as a norm; someone may not be able to state what the norm is or why a judgment is autonomous (a form of self-governance), but that wouldn't entail that it isn't.

Other kinds of human judgment illustrate the point. Consider an athlete: in physical action she assesses the distance between herself and the next hurdle and adjusts her pace so as to clear the hurdle. Suppose this is something that she is able to do repeatedly. It seems to me that her physical movement capabilities constitute autonomy (self-direction/governance) in the realm of [a particular kind of] physical activity. Similarly, a carpenter may articulate methods in the process and activity of crafting woodwork which serve as norms for her self-direction qua carpenter. Each, the athlete and the carpenter, is judging and is generating norms which are [self-]directing of each one's judicative process.

Consider yet another kind of human judgment (that typically is not thought of as involving skill let alone as having anything to do with autonomy!), a parent interacting with her infant. She experiments with different ways of holding the infant and in the process of this [physical, emotional, etc.] activity develops norms by which she recurrently directs herself in the ongoing care of the infant. She is self-directive (has invented norms through her own judicative process) in her capacity to comfort, communicate with, and succor the infant. Presumably such norms evolve over time as the infant grows and the parent's relationship to her child develops, but the point is, that this too is autonomous and no less so for being parental and social than more typical, paradigm examples of autonomous action, such as a person making an "autonomous" career choice.

These examples might not seem to get to the heart of autonomy, since they are "episodic" and not necessarily characteristics of the self as a whole. There are a couple of responses to this sort of objection:

(1) Autonomy need not be of the whole self, but the "subject" of autonomy could be the self in a particular respect. Presumably, it could also be the self as a whole (although more discussion of what "the whole self" means is needed before this claim could be secured); the point is only that it need not be.(31)

(2) That a self is autonomous in some respects, does not entail that it is in others and conversely, that a self is not autonomous in some respect does not entail that it is not in others. To say that a quadriplegic is not autonomous with respect to [most] physical movement or activity would not entail that she is not autonomous with respect to verbal expression.

(3) Autonomy of the whole self, whatever exactly that turns out to mean, would seem to depend on or require that the self have at least some specific autonomy capabilities. This is not a high hurdle to meet on this theory, since a human self is by definition one who judges and at least some judgments will generate norms that are self-directive (however unsophisticated they might be).

I said at the beginning of this section that I thought Buchler's theory offered resources for a robust theory of autonomy and for addressing some of the challenges to autonomy raised by contemporary feminists. A robust theory of autonomy would be one which articulates autonomy in terms of the [norm] inventive powers of the self. The Buchlerian theory of judgment does just that. Moreover, the catholicity of the Buchlerian theory of judgment includes as judgment activities, processes and capabilities that are often not even considered as judicative, let alone as contributing to autonomy (e.g., emotion, physical activity, action in general). Thus, the Buchlerian framework allows for a broadening of what autonomy is and for recognizing as autonomous behavior that has often been dismissed as "merely emotional, intuitive, natural, manual"... and so on. This latter point is particularly a propos contemporary feminists' challenges to autonomy. If "emotion" is no less a contributor to autonomy than "reason" and if social and communicative processes can count as autonomous, then autonomy would not have to be defined as either an exclusively "male" (rational, individualistic) province or value and autonomy as a value would not necessarily be inconsistent with many feminist values (e.g., the value of relationships). These comments are only suggestive. But, if the prospects for the Buchlerian theory of judgment play out as I think they will, then the theory offers a much improved basis on which to understand and evaluate autonomy.

Notes

1 Bennett, Jonathan. 1974. The conscience of Huckleberry Finn. Philosophy 49 (188): 123-34.

2 Bennett takes Himmler's words in the speeches from which he quotes at face value, but one could question the sincerity and meaning of what Himmler was saying. However, if the point is to contrast two different ways of acting in response to sympathy, which is Bennett's point, then we can leave that more troublesome issue about Himmler aside for the purposes of this paper.

3 author, "title" in journal vol., no., (date), pp. xx-xx.

4 Justus Buchler Nature and Judgment (NJ), University Press of America, 1985. I will also cite passages from Buchler's Main of Light (ML), Oxford University Press, 1974. "Natural complex" is Buchler's term for "being."

5 Marjorie Miller pointed out to me that in Buchler's work, except where the reference is to a specific male individual, the pronoun "she" can be substituted for "he" without any change in meaning.

6 See also NJ 21-22: "Assertions usually occur in some form of symbolism which has syntactical structure....But an assertion may be made without using words at all, for instance, by the act of nodding in answer to a question. Nodding as such could be considered an active judgment which, in this context, also functions assertively....In another context, nodding may be an act of lamentation, or a distracted utterance of grief; and in such cases the assertive function may be negligible or absent."

7 Austin's theory of "performative utterances" attempts to broaden the conceptualization of language (Austin 1975) but not of judgment. In Buchlerian terms, a performative utterance is an active judgment in language. "Performative utterance" is the species, "judgment" the more general category. See Singer (1983, 69-71) on Austin and Buchler.

8 Billy Budd by Herman Melville.

9 Both of these are limitations with Habermas' s view of communication (Habermas 1983; Young 1987).

10 The parity refers to their equal relevance as categories of human nature, not to their equal importance in every given instance of judgment.

11A more post-modern term for "self-direction" or "self-governance" would be "self-invention." I have no objection to the term, although I would if it was intended to suggest that invention is either total and absolute or not possible at all. Autonomy does not it seem to me require total reinvention or independence from all "external" influence. If it did, it's hard to see what resources the self would have with which to invent. While I do think autonomy involves invention (through judgment), autonomy or "self-invention" cannot preclude historicity or the influence of social relations. (Foucault himself addresses this point. See, for example, Michel Foucault, "What is Enlightenment?" in The Foucault Reader, ed. Paul Rabinow, Pantheon Books (1984), pp. 32-50.)

12See for example, Isaiah Berlin, "Two Concepts of Liberty," in Four Essays on Liberty (1969) Oxford University Press and John Stuart Mill, On Liberty. Ronald Dworkin's conception of autonomy as "moral independence," that is, being entitled to respect for one's chosen way of life seems closer to this "negative" way of defining autonomy than it does to the second way of defining autonomy which I go on to distinguish. Brison's distinguishing of six different philosophical accounts of autonomy in the free speech literature has been helpful to me in sorting out various approaches to autonomy (Susan J. Brison, "The Autonomy Defense of Free Speech," Ethics 108 (January 1998]: 312-339.)

13Such as (1) critical reflection on desires, capacity for reflection and identification with "first order" desires, and/or a capacity for legislation or (2) the development of a repertory of "autonomy skills" or equal capabilities among persons. For (1) see, e.g., H. Frankfurt, "Freedom of the Will and the Concept of a Person," in The Importance of What We Care About, Cambridge University Press, 1988; John Rawls, A Theory of Justice (1971); Thomas Scanlon, "A Theory of Freedom of Expression," Philosophy and Public Affairs 1 (1972): 204-226, and Scanlon, "Freedom of Expression and Categories of Expression," University of Pittsburgh Law Review 40 (1979): 519-550; G. Dworkin, The Theory and Practice of Autonomy, Cambridge University Press, 1988. For (2) see, e.g., Diana Meyers, Self, Society and Personal Choice, Columbia University Press, 1989; John Christman, "Autonomy and Personal History," Canadian Journal of Philosophy 21 (1991): 1-24; Amartya Sen, "Gender Inequality and Theories of Justice," in Women, Culture, and Development, ed. Martha Nussbaum and Jonathan Glover, Oxford University Press, 1995, Sen, Inequality Reexamined, Harvard University Press, 1992, Sen, "Well-being, Agency, and Freedom: The Dewey Lectures 1984," The Journal of Phlosophy 82 (April 1985): 169-220, and Sen, "Rights and Agency," Philosophy and Public Affairs 11 (Winter 1982): 187-223.

14Such as self-fulfillment, self-realization, personal integration. In Meyers' (1989) account of autonomy or competency skills the aim is personal integration and presumably some degree of such integration would have to be achieved if the competency skills are the skills they purport to be. According to Brison (1988) Redish and Baker posit a self-realization condition for autonomy (Martin H. Redish, Freedom of Expression: A Critical Analysis, Charlottesville, VA: Michi Company, 1984; C. Edwin Baker, "Scope of the First Amendment Freedom of Speech," UCLA Law Review 25 (1978): 964-990.

15The distinction comes from Yaffe who distinguishes "self-expression" theories of freedom of will as theories which equate freedom of will with expression of certain crucial psychological states and "self-transcendence" theories as those which equate freedom of will with responsiveness to evaluative facts. (Gideon Yaffe, "Free Will and Agency at Its Best" in "Philosophical Perspectives, 14, Action and Freedom, 2000 (A Supplement to Nous)", Tomberlin, James E (ed), Blackwell : Cambridge, 2000). I would want to modify the notion of "self-transcendence" to mean not just responsiveness, but inventiveness. In my inventory, then, "personal self-realization" theories could be instances of either the first or the second and "self-invention" theories instances of the second.

16For example, Meyers (1989), Sen (1982, 1985, 1992, 1995).

17Kant, of course, recognized that reason could be subverted to heteronomous purposes, but maintained that reason's possible independence from heteronomous concerns in its pure practical exercise is what guaranteed autonomy. Hence, for Kant, contemporary views which conceputalize autonomy in terms of rational, critical self-reflection on desires would by definition be heteronomous and not autonomous at all. I am not suggesting that they have to be by definition heteronomous, but that by itself rational critical self-reflection is barely distinguishable from rationalization and hence, a weak conception of autonomy.

18This weakness resides in the Kantian analysis as well, since the effort required to reason in a purely rational fashion about practical, i.e., moral, matters, was not an effort that one makes routinely or even with any assurance. As Kant himself admits, one can never be sure that there are not hidden motives undermining the "purity" of the rational operations.

19Or, on one approach, taken by Rorty, autonomy is only "privately" achievable, a sort of personal self-experimentation that is alleged to take place "in private," not in the public sphere, where laws, norms, standards and so on are "public," that is, equally applicable, and therefore determinative of, guiding for, all. (Richard Rorty, "Moral Identity and Private Autonomy: the case of Foucault" in Essays on Heidegger and Others (1991), Cambridge University Press, 193-8.

20This is the complaint of many feminists against the post-modern conclusion that the notion of an autonomous self is a myth. For just as women begin to attain recognition as autonomous selves, autonomous selfhood is declared null and void and the reality of power relations is the only game in town. The goal then, if one lacks power, is to gain it, a reversion to a kind of Hobbesian universe. There may be something to understanding feminist politics in these terms. In that case, the declaration that autonomous self-hood is an illusion might better be understood as a rhetorical power move, rather than a theoretically sound conclusion.

21I am torn between the ideas of self-direction and self-governance, because the former seems to lack reference to norm invention and the latter seems to have too many Kantian associations. For the present, then I am using them both or alternating between them as seems appropriate to the context.

22In other words, even if some possibilities are necessities, not every possibility is.

23Justus Buchler, Metaphysics of Natural Complexes, second expanded edition, SUNY Press, 1989, Chapter IV. Buchler develops the Jamesian (and Bergsonian) insight in terms of his own ontological categories, the analysis of which is at this point beyond the scope of this paper.

24These would include issues about whether autonomy is compatible with the value of relationships, whether it reflects a "male" individualistic perspective, and with respect to morally legislative views of autonomy whether they undermine unique individuality. Also raised are issues about the very constitution of the self as "relational" which I have not addressed in this paper at all. A recent representative resource for some of those discussions is Relational Autonomy: Feminist Perspectives on Autonomy, Agency and the Social Self, ed. Catriona Mackenzie & Natalie Stoljar (2000) Oxford University Press.

25"Creation" could mean modification, alteration, extension, refinement, rejection of existing norms, as well as generation of different norms altogether.

26In a Kantian framework, this would be "hypothetical." For Kant, the operative power would be reason and reason in the service of other, non-rational goals is a conditional or instrumental exercise of reason. McGandy, in his discussion of Buchler's notion of query, proposes a four-fold structure to the normative character of query, employing the Kantian terms "hypothetical" and "categorical" to mark the difference between two distinct dimensions, of what he calls "finite pragmatics" and "absolute pragmatism." (Michael J. McGandy, "Buchler's Notion of Query," Journal of Speculative Philosophy XI, 2 (1997), 203-224.) I draw inspiration from the distinctions McGandy makes and employ some of them in this discussion.

27Employing the hypothetical/categorical framework mentioned in the previous footnote, this would be "categorical." Norms in this sense would be "unconditional" in the sense that they are not for the sake of results per se, but for the furtherance of judgment itself.

28Such norms may at some point be superceded by other norms generated by my own philosophical activity and judgments. They might also contribute to the autonomy of the community of philosophers, as well. That is, philosophers invent; they generate the norms by which they produce and assess their own products. I am not going to pursue here the idea that communities might be meaningfully be said to be "autonomous."

29Huck's case is complicated. In so far as he seems to explicitly repudiate the idea that his behavior is right, he may not be autonomous. But in so far as he seems to think that he ought to guide himself by his feelings because morality doesn't seem to satisfactorily address the situation perhaps he is autonomous.

30The Himmler norm would be ""when 'principles' conflict irreconciliably with sympathy, act on principle and keep sympathy in the background, or suppress it."

31Views of autonomy which attribute it only to the moral self hold that the subject of the autonomy is only some part or dimension of the self and not the whole self. I am not taking a position in this paper on the issue of the autonomy of the whole self, just pointing out what would be the implications of the approach I am taking.