12

VIRTUE WITHOUT AGENCY

Sentiment, Behavior, and Habituation in A. Smith

Throughout the Theory of Moral Sentiments, there is a marked reversal of emphasis, away from the drama of volatile and non-cognitive passions and toward reaffirming the continuity of a different type of affect. The course correction here takes the form of retranslating the passions—not back into a metaphysics of the will, to be sure—but into a firmly empirical, at times seemingly actuarial understanding of reason as it manifests itself in established customs, prevailing manners, average forms of behavior, and a mimetic conception of virtue. Viewing his arguments as post-metaphysical, yet also wishing to move beyond the rationalist, emotivist, and skeptical critiques of metaphysics that had dominated since the Restoration, Adam Smith seeks to overcome the antagonism of will and intellect—a dilemma that, unbeknownst to him, modernity had not so much discovered as created. To David Marshall, Smith “seems less concerned about the constitution of the self” and indeed “presupposes a certain instability of the self; it depends upon an eclipsing of identity, a transfer of persons.”1 Marshall’s compact formula risks obscuring, however, that such a transferential model of sociality achieved by continued imaginative substitution constitutes something of a logical paradox. For “how can one become another person without suffering the dramatic change that is self-liquidation?” Furthermore, “if my identity is caught up with yours, and yours with another’s, and so in a perpetually spawning web of affiliations, how can I ever know that your approving glance is your glance, rather than the effect of an unreadable palimpsest of selves?” After all, any such knowledge hinges on “entering into another experience while retaining enough rational capacity of one’s own to assess what one finds there. The cognitive distance which such judgements require cuts against the grain of an imaginary ethics.”2

Arguably, none of these logical paradoxes can be resolved in the terms in which they are here being stated—that is, in a vocabulary still committed to knowledge as propositional and tethered to a solitary and self-aware epistemological agent. Though far from meaning to present an apologia for Smith’s transferential conception of moral agency, the reading here undertaken suggests that it is precisely this mentalist idea of knowledge—viz., as a type of intentionality issuing in a distinct representation—which Smith means to leave behind. In fact, his solution to Hume’s epistemological dilemma rests on distilling how the inherently non-cognitive conduct of individuals will yield rational, systemic effects that could never be secured if social meaning were to depend on subjective intention. To this end, Smith comprehensively re-describes passion as “sentiment” and, in turn, sentiment as a social transaction or “behavior.” The result of Smith’s sweeping account of sociality as the circulation and mimetic appraisal of sentiments is a moral theory that bears more than a passing affinity to modern behaviorism.

As is evident from Hobbes’s and Locke’s strictly epistemological approach to the self, the modern conception of truth as an objective to be realized by specific epistemological method is the most significant legacy of late Scholastic nominalism. In repudiating the idea of knowledge as a result of active contemplation (theoria) for which the cosmos had once furnished the ontological source and ethical telos, modern inquiry after Bacon and Descartes instead proceeds by isolating singular entities as the only viable locus of meaning. The resulting paradigm of knowledge as “information” thus gives rise to a fact/value divide that Hume eventually sets forth as an axiom of modern rational inquiry, the result being that knowledge as an intellectual commodity has become terminally estranged from the broader ideal of wisdom and human flourishing. After 1750, epistemology and ethics are fundamentally conceived as distinct and, increasingly, as unrelated pursuits, and it is in Adam Smith that this bifurcation is completed as the project of moral philosophy migrates from a theological to a sociological, and from a normative to a descriptive endeavor. Such a shift completes a tendency that had first taken shape in the methodological treatises of Bacon and Descartes, and that was subsequently radicalized by Newton and Locke: viz., the wholesale redefinition of knowledge as a quest for “certainty,” which is to say, as a possession bound up with its inter-subjective communicability and accredited with reality only insofar as the self can exercise dominion over it in propositional, syllogistic form. Having already traced some of the effects of this anti-humanist strain in modern thought, we might also allude to later, yet more radical versions of epistemological skepticism, such as Frege’s and Wittgenstein’s systematic de-psychologizing of cognition, of which behaviorist and neuro-scientific models are the two most prominent reductionist offshoots. Notwithstanding their substantially different objectives, all these successors of modern nominalism share at least one aim: viz., to expunge the idea of human interiority and introspection altogether by driving a wedge between the phenomenological event of a thought (deemed inaccessible and thus irrelevant to empirical inquiry) and its content—judged real and pertinent only insofar as it can be objectively captured either as an ordinary-language proposition, a statistically observable behavioral pattern, or a measurable neural event.

The following reading of Smith’s Theory of Moral Sentiments seeks to draw attention to the prehistory of these more recent projects by exploring how some key terms (e.g., passion, sentiment, sympathy) traditionally associated with introspective accounts of moral judgment and human flourishing are for the first time being systematically reinterpreted as social and objectively classifiable phenomena. To be sure, Smith’s own project rests on a few distinguished precursors, including the Stoics, Shaftesbury, Hutcheson (Smith’s teacher), and Hume. While Shaftesbury’s role in the Theory of Moral Sentiments is oblique and has gone mostly unrecognized, his principal thesis that moral and spiritual self-cultivation is circumscribed by an intricate social cum aesthetic grammar crucially shapes Smith’s argument.3 One need only recall a passage like the following to pick up on the deep connection between Shaftesbury’s and Smith’s model of benevolence and sympathy; in it Shaftesbury ponders

the narrowest of all Conversations, that of SOLILOQUY or Self-discourse. But this Correspondence . . . is wholly impracticable without a previous Commerce with the World: and the larger this Commerce is, the more practicable and improving the other, he thinks, is likely to prove. The Sources of this improving Art of Self-correspondence he derives from the highest Politeness and Elegance of antient Dialogue, and Debate . . . And nothing, according to our Author, can so well revive this self-corresponding Practice, as the same Search and Study of the highest Politeness in modern Conversation. (SC, 3:96)

Likewise preferring a meliorist, conversational tone to Hume’s sharp-edged analytical idiom, Smith in 1759 locates the “selfish passions” as occupying “a sort of middle place” between “the social and the unsocial” (TMS, 40). Long before Kant’s late utopian musings advanced “the most extreme counterposition to the [Hobbesian] principle auctoritas non veritas facit legem,” Scottish political economists had already begun to shift from a voluntarist command ethic toward a narrative model that conceives the will or, at least, its empirical heirs—the passions—as susceptible of “improvement.”4 The trajectory in question typically proceeds from the mindless force of mute desire, advancing to an expressive but inadequately socialized “passion,” and culminating in the eventual lucidity of a stable set of “interests” or, in Hutcheson’s phrase, “secondary desires” (EPA, 19) responsive to the systemic cues of modern commerce. Even for the skeptical Hume, the project’s logic is quite irresistible. Acknowledging that it simply is not feasible to “infuse into each breast . . . a passion for the public good,” Hume instead suggests that it is “requisite to govern men by other passions, and animate them with a spirit of avarice and industry.”5

Albert O. Hirschman has called this the “marvelous metamorphosis of destructive ‘passions into virtues’” through the new paradigm of “interest.”6 Anticipating Hirschman’s thesis, John Pocock had previously qualified this view in one important respect; for even as credit was being “translated into virtue,” this “restoration of virtue was subject to a single sharp limitation.” Thus, insofar as “virtue was now the cognition of social, moral and commercial reality, . . . imagination . . . is replaced in the Whig literature of 1710–1711 by nothing more than opinion” and “rationality is only that of opinion and experience.”7 More than any of his contemporaries, Smith bears out this insight by systematically disabling the Augustinian conception of virtue as predicated on sustained introspection and standing counterfactually vis-à-vis a fallen world. In its place, Smith ventures a mimetic conception of the good bound up with the vicarious transfer of prevailing, socially sanctioned sentiments, an approach that “made it possible to dismiss the egoistic motivational theory underpinning Mandeville’s paradoxes.”8 While Smith’s focus on the passions is a familiar feature of eighteenth-century moral philosophy, his objectives differ markedly from those of Hutcheson and Hume. For underlying his concern with the passions, and their potential commutation into durable sentiments, is no longer Locke’s or Hume’s epistemological quest for a viable and verifiable successor to the metaphysical notion of the will. Rather, Smith seeks to disengage moral reflection from the stranglehold of Hume’s radical skepticism, which had pushed Lockean nominalism to such extremes as to render basic humanistic key concepts (e.g., will, action, consciousness, self, introspection, judgment) all but meaningless. Retreating from this philosophical dead-end, Smith finds Stoicism to be of particular value to his objectives, in part because the Stoics, even as they had acknowledged the intrinsic deficiency of human intelligence (moral and otherwise), had also sought to remedy that predicament dialectically by arguing for the complementarity of finite and imperfect human perspectives. From Smith’s neo-Stoic viewpoint, this fortuitous alignment of blind volition and systemic rationality reflects a divinely ordained and providential arrangement: “Nature, accordingly, has endowed [man], not only with a desire of being approved of, but with a desire of being what ought to be approved of; or of being what he himself approves of in other men.” Moral and economic self-improvement thus follow the same behavioral template, viz., a quest for “self-approbation” whose pursuit constitutes “the principal object, about which [a wise man] can or ought to be anxious. The love of it, is the love of virtue.”9

At first glance, this new credit-based and self-interested, entrepreneurial self promises to offer “a countervailing strategy” to the Hobbesian, all-consuming will “on a continuing day-to-day basis” rather than demanding for its containment the Leviathan’s ad hoc projection of overwhelming force.10 In short, Hobbesian voluntarism with its strictly negative conception of power periodically reviving the unruly subject’s “Feare of Death” is transformed into a systemic principle. For Smith, it is above all sympathy that counterbalances the specter of a radically particularist and pluralist nation where “every individual is . . . attached to his own particular order, . . . his own interest, [and] his own vanity.” In the innumerable, serendipitous encounters of which quotidian sociality is composed, the practice of sympathy gradually effects an enduring, if imaginary model of community. Smith calls it “that more gentle public spirit” (TMS, 230–232). Yet to postulate that “every man is . . . by nature, first and principally recommended to his own care” (TMS, 82) makes clear that voluntarism remains very much the default model of agency. Indeed, far from remedying Hobbes’s and Hume’s anti-rationalism, the Whig conception of commercial society proved in its own ways just as “fantastic and nonrational” and threatened “to submerge the world in a flood of fantasy.”11 Parallel to this transformation of the will into potentially rational, predictable, and manageable “interests” run various cultural narratives about the rise of modern “refinement,” “manners,” and “taste,” as well as a host of new institutions dedicated to their advancement.12 To the Scottish political economists of the mid-century it was apparent that governmental authority and individual will were increasingly mediated by the impersonal rationality of complex networks and “expert languages” (Anthony Giddens), rather than by a sovereign and centralized power. As Adam Smith puts it, “in the great chess-board of human society, every single piece has a principle of motion of its own, altogether different from that which the legislature might chuse to impress upon it” (TMS, 234).

Though differing widely in their critical stance vis-à-vis the story they wish to tell, Norbert Elias, John Brewer, J. G. A. Pocock, Linda Colley, and Michel Foucault (to name but a few), all converge in interpreting this “rise of disciplinary society” (as Charles Taylor has most recently labeled it) as a process that increasingly disperses and so obscures the force that had loomed so conspicuous and ominous in Hobbes’s Leviathan and Mandeville’s Fable.13 Hutcheson’s Essay (1728) had pointed the way here, arguing that it was “foolish” to infer “from the universal Prevalence of these Desires [of Wealth and Power] that human Nature is wholly selfish, or that each one is only studious of his own Advantages; since Wealth or Power are as naturally fit to gratify our Publick Desires, or to serve virtuous Purposes, as the selfish ones” (EPA, 19). With the opportunities for economic and social mobility largely cordoned off by the mercantilist system of his native France, Montesquieu’s theory concerning the separation of powers pursues a cognate objective within the field of constitutional philosophy. Thus he also posits a systemic balancing of interests as the most auspicious strategy for containing a will that, like Locke and Mandeville, he also regards as inherently volatile, blind, and irremediably selfish. This shift from an auratic to a technocratic and from a personal to a systemic understanding of power would, of course, continue and in time spawn often brilliant analyses of modern constitutional, political, and economic thought (from Montesquieu to the Federalist Papers to Hegel’s Philosophy of Right to Marx’s critique of political economy and even Max Weber’s sociological account of institutionalized politics, science, and culture).

Their highly diverse ideological commitments notwithstanding, however, the mid-eighteenth-century works of political, economic, and moral theory converge in this one point: they all regard the will as essentially opaque, unfree, and irrational. Such is the case wherever the individual will is subject to the impersonal and remedial discipline of the law and to institutions of education and cultural literacy broadly speaking—all of which are gradually being aligned with the instrumental rationality said to govern economic behavior. Smith’s account of the “impartial spectator” perfectly embodies this meliorist logic by positing moral development as a complex process of transference: just as the other’s “sympathy makes them look at [suffering], in some measure, with his eyes, so his sympathy makes him look at it, in some measure with theirs, especially when in their presence and under their observation.” A new, notably transactional conception of sociality thus takes shape, one that pivots on reconciled customs and manners of mid-eighteenth-century polite and commercial culture, fueled by a rhetoric of “improvement” that oscillates between the obliquely moral and the emphatically sentimental. Under this new dispensation, the containment of the (Hobbesian) will and the restoration of a certain kind of rationality is conceived as a procedural question, a matter of the right technique being brought to bear on the passions. In classical Stoic fashion, Smith thus views his own philosophical enterprise as therapeutic in nature and, in particular, as a quest for incentives such as will effectively and efficiently regulate behavior.14 No longer considered are Aristotelian and Thomistic conceptions of the will as susceptible of internal clarification and as an integral part of moral cognition. Hobbes’s view of the will as a strictly appetitive, volatile, and opaque pathos devoid of intellectual potential and impervious to introspective remediation thus is not so much opposed by his Enlightenment heirs as it is turned into a bleak premise for their meliorist view of politics, culture, and economics as the most auspicious venues for recovering a socially responsible model of human agency.

In fundamental if not entirely obvious ways, Smith accepts Locke’s and Hume’s account of a self driven by hedonistic pleasures and all but bereft of personal identity due to the discontinuity and incommensurability of all sensation. The self as an inner agent, while perhaps real, is ultimately deemed inaccessible to philosophical speculation; it is constitutively opaque and, to judge by such evidence of the inner life as still reaches us, it lacks any coherent and sustained sense of its reality as a person. Hence, Smith’s strategic investment in the notion of sympathy arises from his deep-seated conviction “that others’ states of mind are naturally inaccessible to us, concealed as they are by the fleshly encasements of their bodies.” Moreover, Smith “was not convinced that sympathy could, on its own, maintain social order” and, where the limits of sympathetic sociality arose, was quite willing to maintain such order “through the fear of death operating in conjunction with a sovereign power.”15 It is not merely a breakdown of inter-subjective discourse, then, but the underlying, radically nominalist view of a self dissolved into heterogeneous impressions, sensations, states, desires, and affections that prompts Smith to concede from the outset that “we have no immediate experience of what other men feel” (TMS, 9). The apparent tension between Locke’s epistemology of the self and his political theory of rational agents electively and deliberatively entering into social relations by means of propositional, quasi-contractual arrangements had been thrown into disarray by Hume’s dispersal of personal identity, his atomistic view of the passions, and his disjunction of fact from value. Smith’s Theory of Moral Sentiments thus attempts to rethink the very project of moral philosophy following the massive onslaught of post-Hobbesian skepticism on the self’s epistemological coherence and moral integrity. Crucially, passion in Smith’s Theory is no longer viewed as a purely impulsive, hedonistic, and self-consuming mental event. Rather its seemingly irrational thrust is being attenuated, even reversed, in that Smith regards emotion as an inter-subjective phenomenon, something to be reconstituted in the simulacrum of “sentiment”—less an expressive act than a behavioral norm designed and displayed so as to maximize prospects of “approval” by others.

While the Stoic project of overcoming the emotions (as de facto misjudgments) remains a central feature of Smith’s argument, it is being deployed here for very different strategic purposes. For the objective is no longer, as it had been for the Stoics, a quest for inner balance and wisdom (apatheia) but, rather, the smooth operation of social life as an end in its own right. In anticipating and accommodating itself to projected conditions of reception, Smith’s individual will “conceiv[e] some degree of that coolness about his own fortune, with which he is sensible that they will view it” (TMS, 22). In sharp contrast to classical Stoicism, then, self-possession and self-mastery (autarkeia) are at most incidental to the decidedly un-Stoic project of an affect-based and strictly “imagined” community. Indeed, the meaning, value, and significance of the inner life will only disclose themselves insofar as passion has been successfully converted into the social currency of behavior, a term that will occupy us shortly. If a primal passion still furnishes the raw material for this transformation, it signifies and is credited with reality only to the extent that it has been successfully transposed into “sentiment”—that is, a type of behavior sanctioned by the social grammar that undergirds Smith’s sympathetic community. The imagined sphere of discursive sociality at once recovers passion from the netherworld of mute and unfocused animal desire and in so doing redeems it for purposive human life: “Society and conversation, therefore, are the most powerful remedies for restoring the mind to its tranquillity” (TMS, 23). Conversely, an inner life that refuses to accommodate the sociality of sentiments is at best value-neutral, though more likely of a pathological or criminal nature—a vagrant, opaque, and intractable symptom on the order of Nietzsche’s “extra-moral” (aussermoralisch) sense.

It is just this conversion that shows Smith’s use of Stoic philosophy to be rather peculiar and selective, and which also reveals the acute modernity of his Theory of Moral Sentiments. For the Stoics, who follow Aristotle rather than Plato, lack of self-command or irresoluteness of will (ἀκρασία) is not a result of hedonistic and compulsive appetition but of a flawed judgment, which is to say, of precipitous or ill-conceived “assent” to some desire or other. Inasmuch as Stoicism views the mind as substantially one and not to be partitioned in the spirit of modern faculty psychology, the emotions cannot be understood as physiologically conditioned surges intruding on the otherwise distinct and separate superagency of reason.16 Seneca remarks that “if any one supposes that pallor, falling tears, prurient itching or deep-drawn sigh, a sudden brightening of the eyes, and the like, are an evidence of passion and a manifestation of the mind, he is mistaken and fails to understand that these are disturbances of the body.” It is not that the mind is victimized by quasi-physical passions (pathē) but, rather, that it “suffers” them to acquire excessive or distorted epistemic and moral force. An emotion thus “does not consist in being moved by the impressions that are presented to the mind, but in the surrendering to these.” It is an act of precipitous and misguided “assent,” which is to say, an intellectual operation that has eluded the kind of scrupulous supervision that the Stoics mean to instill in their disciples. Speaking of the paradigmatic emotion of anger, Seneca thus notes that it “must not only be aroused, but it must rush forth, for it is an active impulse; but an active impulse never comes without the consent of the will [numquam autem impetus sine adsensu mentis est].”17 While a fuller exploration of this complex and still contested issue of the Stoics’ concept of emotion is not feasible here, a number of points can be extracted that will reveal the very different thrust of Smith’s seemingly neo-Stoic argument.

First, the Stoics understand all acts of mind—including the passions—as propositional in nature. Second, to the extent that a passion has any hold on the mind it does so solely in consequence of the intellect’s assent to the (oblique) proposition with which it is presented. Passion, then, is not a distinct antagonist of reason but evidence of the latter’s as yet incomplete cultivation. To the extent that passion holds sway over someone’s mind, it points to one’s failure to apprehend and appraise the propositional nature of “impressions” with the requisite care. Third, it is just this type of failure that gives rise to a disorder of judgment and, ultimately, to ill-conceived action, a syndrome for which the Stoics adopt the Aristotelian concept of akrasia. Here again, it is important not to misidentify akrasia as some isolated failure to act on a good clearly perceived; neither is it some deficient act of mind or “opinion” randomly insinuating itself into the otherwise rational and responsible narrative of a life. Far from some contingent psychological mishap obtruded from without, akratic “weakness of will” is integral to human psychology. For Aristotle, akrasia thus involves a specific kind of desire (orexis), one that is not determinative like the craving for food or sex but, rather, something unpredictable and self-fuelling. Unsurprisingly, it is “anger” (ira) that the Stoics, and Seneca in particular, identify as the very embodiment of akrasia—viz., a pointedly a-social passion that can be either prospective (e.g., revenge) or retrospective (e.g., resentment). In the first case, there is no determinate causal link between the akratic disposition and a specific course of action, for as we well realize, Hamlet may or may not act to avenge his father’s murder. Conversely, in the case of looking back in anger it is even more obvious that akrasia does not compel a specific action since the past cannot be changed.18 Crucially, the Stoic objective of heading off misjudgment—that is, the precipitous assent to an emotionally colored and distorted perception—was to be realized by methodical and sustained introspection.

Not so in the Theory of Moral Sentiments; for while Smith appears to echo the Stoic quest for isolating and disabling the sources of a disordered inner life, his solution is precisely not one of introspection but, rather, to declare all passion akratic until and unless it has been socialized as a benevolent sentiment. The result is a marked shift in emphasis, which in turn yields a distinctively modern conception of moral agency. To begin with, it is apparent that Smith no longer views the passions as propositional in nature. Both in its primitive state and once converted into a socially recognized “sentiment,” emotion in Smith’s Theory proves to be altogether extra-rational and non-propositional. The telos of Smith’s account is thus neither knowledge nor the self’s ability to form logical and critical judgments on the quasi-propositional character of the emotions. Rather, Smith seeks to effect a mimetic alignment of the self’s emotively charged impressions with what are hypostatized to be the cognate affective experiences of other individuals. Simply put, the modern objective is emulation, not cognition. From this a second point follows; for contrary to views that the Enlightenment came to hold, classical Stoicism’s methodical quest for inward composure and rational self-governance had always served the final objective of building up the self’s capacity for active and selfless citizenship. Its structured regimen of self-mastery was aimed at furnishing certain and unbiased representations (not sentiments) such as would prove conducive to political “action.”19 Indeed, it is just this framework of a vita contemplativa integrated with, rather than opposed to, the vita activa that was also shared by Augustine and Aquinas (the latter taking his cue directly from Aristotle rather than the Stoics).20 By contrast, Smith’s critique of Stoicism as expiring in the languor of “sublime contemplation” and after-the-fact “consolation” is not only misleading but also obscures his own theory’s far more equivocal outlook on action. In Smith’s rather one-sided portrayal, Stoicism limits the self to a cultivation of “apathy” and to a concerted attempt at “eradicat[ing] all our private, partial, and selfish affections, by suffering us to feel . . . not even the sympathetic and reduced passions of the impartial spectator.” The result, we are told, is to have removed the self from “every thing which Nature has prescribed to us as the proper business and occupation of our lives” (TMS, 292–293).

Notably, Smith’s critique of Stoicism in his survey of moral philosophy (added as Part VII to the 1790 edition of his Theory of Moral Sentiments) also departs from the neo-Stoic thrust of his book’s earlier sections, written some three decades earlier. There Smith had firmly aligned himself with the Stoic ideal of affective self-governance: “to restrain our selfish, and to indulge our benevolent affections, constitutes the perfection of human nature; and can alone produce among mankind that harmony of sentiments and passions in which consists their whole grace and propriety” (TMS, 25). Echoing Joseph Addison’s Cato and anticipating similar claims in Gotthold Ephraim Lessing’s Laokoon, Smith expresses admiration for “the man who has lost his leg by a cannon shot, and who, the moment after, speaks and acts with his usual coolness and tranquillity.” What is put on theatrical display here is the ostensibly anti-theatrical Stoic virtue of “self-command.” Yet there is something decidedly equivocal about Smith’s claim that “in proportion to the degree of self-command which is necessary in order to conquer our natural sensibility, the pleasure and pride of the conquest are so much the greater” (TMS, 147). For it is just this ostensibly Stoic conquest of the passions that, in Smith’s account, has to be put on display inasmuch as its success remains in epistemological limbo until and unless it has been confirmed by outside spectators.21 Elsewhere, Smith thus extols

that reserved, that silent and majestic sorrow, which discovers itself only in the swelling of the eyes, in the quivering of the lips and cheeks, and in the distant, but affecting coldness of the whole behaviour. It imposes the like silence upon us. We regard it with respectful attention, and watch with anxious concern over our whole behaviour, lest by any impropriety we should disturb that concerted tranquillity, which it requires so great an effort to support.22

This is a fine instance of what we might call faux apathy, a modern, theatrical simulation of the Stoic ideal whose original purposes, however, substantially elude Smith. What stands out is not its professed extirpation of passion but, rather, the manner in which its seeming conquest is put on display. The conspicuous theatricality with which the raw and unfiltered psychic data of passion are shown to have been commuted into the artifact of a “concerted tranquillity” is lauded as a worthy achievement precisely because it enables the spectator to admire the “effort” expended in that very transformation. What accounts for the pivotal role of both behavior and sympathy here is that, being socially intimated with such “propriety,” they call for no action whatsoever on the part of the spectator. Following Rousseau, Smith consistently holds “that sympathy be something that one feels rather than something that one does,” and as Nancy Armstrong has argued it is this very premise “that our most compelling feelings might have a source external to ourselves [that] becomes especially apparent in writing associated with ‘sensibility.’”23

As in Smith’s later economic arguments in the Wealth of Nations—a book now understood to have grown out of his moral philosophy—sympathy gives rise to a distinctly modern conception of sociality as an imagined, indeed simulated lateral bond between anonymous, hermetic, and substantially unrelated individuals.24 The compact that defines the modern economic and social order is strictly virtual; sympathy here functions as the moral equivalent of the speculative commodity of modern stock. Its value pivots on the shrewd management of how it is socially perceived, that is, on the accommodating and confident “behaviour” (an important word used twice in the above passage) with which it is introduced into social space. Notably, the following rhetorical question (whose closing exclamation mark suggests that it only admits of an affirmative reply) shows Smith entirely blinded to the possibility of deception—collectively felt and reinforced—about the import and value of moral sentiments: “How amiable does he appear to be, whose sympathetic heart seems to re-echo all the sentiments of those with whom he converses, who grieves for their calamities, who resents their injuries, and who rejoices at their good fortune!” (TMS, 24). To embrace this stance is to delegate one’s moral judgment and orientation to others, to make it dependent on behavioral cues furnished by others, and to collapse the good and the true into the contingent (and possibly opportunistic) affirmation that some particular view is perceived to have been accorded by others. Few models could be farther removed from the Stoic ideal of autarkeia than this transferential and manifestly heteronomous cultivation of moral agency.

For a canonical counterexample, one might look at Augustine’s suspicious probing of theatrically induced “sympathy” (misericordia) in Book 3 of the Confessions. Sheer spectatorship, in Augustine’s view, does not constitute but merely simulates moral meanings: “What compassion is to be shown at those feigned and scenical passions? For the auditors here are not provoked to help the sufferer [sed qualis tandem misericordia in rebus fictis et scenicis? Non enim ad subveniendum provocatur auditor].” Spectatorship by itself is corrosive of action and agency in that it shifts the focus from the practical realization of the good to a narcissistic delight in the very failure of it.25 If staged suffering elicits sympathy, an increase of it will augment the sympathetic emotion and the spectators’ gratification in its experience: “they so much the more love the author of these fictions, by how much the more he can move passion in them [et auctori earum imaginum amplius favet, cum amplius dolet].” Indeed, not only do theatrical simulations of suffering beget in the spectator an analogous, virtual attachment to “sympathy” (“Are tears therefore loved, and passions? [lacrimae ergo amantur et dolores]”), but they thereby also create an incentive to desist from practical action. The attenuation of moral agency, Augustine contends, stems from the spectator’s attachment to the gratifying emotion of sympathy, over and against envisioning himself or herself as capable of achieving the good. To do the latter, in fact, is to conjure the logically absurd scenario of wishing for both the flourishing and the suffering of others at the same time: “For if there be a good will that is ill-willed (which can never be), then only may he, who is truly and sincerely compassionate, wish there might still be some men miserable, that he might still be compassionate [si enim est malevola benevolentia, quod fieri non potest, potest et ille, qui veraciter sinceriterque miseretur, cupere esse miseros, ut misereatur].” As embodied by the stage and the hedonistic model of spectatorship to which it gives rise, Augustine’s view of sympathy (misericordia) is that of a false good, a “joy that enchains” (vinculum fruendi), as he puts it with deliberate emphasis on the sexual connotations of vinculum (fetters, bondage). For Augustine, emotion is legitimate only inasmuch as it is a source of action, not a self-consuming, narcissistic experience; it enjoins the self “to relieve” (ad subveniendum) genuine “suffering” (miseria). The anti-theatrical passage that opens Book 3 thus becomes a template for similar scenes in subsequent books, such as Augustine’s deeply suspicious hermeneutic of his own grief at the death of a young friend in the book following (Confessions, 4.4–7); and it only finds its completion when, mourning the death of his mother in Book 9, Augustine appears at last to have achieved the proper ratio of grief, sympathy, and purposive action.

Fifteen hundred years later, the Theory of Moral Sentiments offers a starkly different picture by construing sympathy as a type of virtual action, rather than the inner condition to be complemented by an active quest for providing material relief to those suffering. For Smith, both the spectator and “the sufferer long less for relief from pain than for the relief that is afforded only by sympathy,” and the latter necessarily requires some material suffering (or theatrical simulation of it) as its precondition.26 Discussing sympathy under the heading of the “social passions” (TMS, 38–40), Smith is at pains to conceive moral sentiments as the implicit approval of a prevailing social consensus. Having all but lost the hortatory and potentially transformative function as “source” that it held in Augustine, sympathy instead names (and affirms as valid) an existing and manifestly self-certifying social consensus: “Generosity, humanity, kindness, compassion, mutual friendship and esteem, all the social and benevolent affections, when expressed in the countenance or behaviour, even towards those who are not peculiarly connected with ourselves, please the indifferent spectator . . . We have always, therefore, the strongest disposition to sympathize with the benevolent affections” (TMS, 38–39). What facilitates this lateral, albeit inarticulate comradeship where “these affects, that harmony, this commerce, are felt (TMS, 39) is the virtual agency of the impartial spectator, a heuristic fiction that effectively collapses the distinction between self and other, thereby suspending the individual’s self-awareness as a responsible being.

A coded answer to Hume’s dystopia of a world where empirical fact and normative meaning have become terminally estranged, Smith’s moral sentiment goes to the other extreme. By its very nature, he repeatedly argues, sympathy is a psychological fact that implies its own value—something suggested by the equivalent position of “countenance and behaviour” or “sentiments and behaviour” (TMS, 162).27 The moral significance of emotion is thus confined to its inter-subjective realization qua sentiment, a “fellow-feeling” whose meaning and value expire in the narcissistic dramaturgy of sentiments displayed and approved, respectively. David Marshall’s well-known account of Smith’s theatrical aesthetic seems rather oblivious to the question of just what kind of theater it is that the Theory of Moral Sentiments stages. On his account, Smith’s “self is theatricalized [sic] in its relation to others and in its self-conscious relation to itself,” even as it “must be an actor who can dramatize or represent to himself the spectacle of self-division in which the self personates two different persons who try to play each other’s part, change positions, and identify with each other.” Quoting Hume on how “the minds of men are mirrors to one another” (HT, 236), Terry Eagleton likewise remarks on the narcissistic logic underpinning both Hume’s and Smith’s social theories: here “the [Lacanian] imaginary . . . is a sort of mutual admiration society, in which in a kind of mise-en-abyme each act of reflection gives birth to another,” thus revealing to us “the cyclical time of the imaginary rather than the linear evolution of the symbolic.”28 In so orchestrating a kind of égoïsme à deux, sympathy no longer furnishes the occasion for focused deliberation and judgment as to what action a given situation calls for. Rather, Smith’s sociality is theatrical to its very core; here all the world is indeed a stage, albeit with this peculiar qualification that everyone is always on stage, but takes himself or herself to be a mere spectator. At the same time, action and plot have been all but supplanted by an invariant display of epistemological narcissism inasmuch as every individual’s performance is mimetically enslaved to the same role and script. We are much closer to the realm of Pirandello and Beckett than to that of Joseph Addison or even Shakespeare. The result is a system based on “substitution as a foundational principle [eine Art fundamentaler Stellvertretung]” where the “spectator is above all spectator of himself [dieser Zuschauer ist vor allem auch Zuschauer seiner selbst]”; subjective experience and meaning thus unfold within a hypothetical, indeed, virtual matrix of “as if” relations that render “truth values inseparable from role-playing, illusion, semblance, and stage dynamics.”29

It is this structural ambiguity of Smith’s “moral sentiment” that prompts Vivasvan Soni to speak of a persistent “double meaning of . . . sympathy as affective identification and sympathy as pity” and, hence, to read Smith’s Theory as inaugurating “an epochal shift . . . from an ethical to an epistemological and identificatory structure of reading.” According to the latter model, for the first time fully realized in Smith’s account, “the purpose of a narrative is to allow one to reproduce imaginatively the world and the experiences of the protagonist, instead of producing an ethical relation to the narrative situation of the other.” In Smith’s post-teleological world, action has been supplanted by the merely transactional, and rational deliberation by a cascade of minute transferences.30 No longer is human flourishing conceived as a dialectical (and potentially tragic) narrative composed of introspection, (mis-)judgment, and purposive action, however imperfect. Instead, by shifting focus to the socialization of potentially isomorphous selves—types rather than persons—Smith’s project pivots on the mimetic acquisition and circulation of virtual sentiments in the guise of approved forms of “behaviour.” Action, understood as the teleological fulfillment of rational personhood in Aristotelian and Thomistic thought, has been all but absorbed into this “sentiment of approbation” wherein, as Smith tells us, “there are two things to be taken notice of; first, the sympathetic passion of the spectator; and, secondly, the emotion which arises from his observing the perfect coincidence between this sympathetic passion in himself and the original passion in the person principally concerned” (TMS, 46; italics mine). More pointedly yet, a later section probes the “difference between the approbation of propriety and that of merit or beneficence,” with Smith concluding that “’till I perceive the harmony between his emotions and mine, I cannot be said to approve the sentiments which influence his behaviour” (TMS, 78). In both cases, Smith is responding to Hume’s astute query whether it is reasonable to suppose that “all kinds of Sympathy are necessarily Agreeable,” considering that “the Sympathetic Passion is a reflex image of the principal, [and hence] it must partake of its Qualities, and be painful where it is so.”31 In the event, Hume’s question reflects something of a misunderstanding of Smith’s argument, which is no longer premised on some garden-variety intersubjective relation between otherwise autonomous agents but, on the contrary, understands moral agency as something transferentially constituted.

In this regard, the word “sentiment” in Smith’s eponymous moral theory is rather misleading in that it seemingly posits an interior and authentic emotional certitude as the epistemological point of departure for its argument. Yet there are multiple indications that this not the case. For one thing, the concept of the will has now definitively vanished from the vocabulary of moral theory, a development only intelligible when seen in the broader context of Smith’s overall retreat from an interiorist, rescogitans model of subjectivity. To understand the nature and significance of that shift, it helps to interrogate the work’s central concept: what does Smith mean by “sentiment,” and what can that word signify in a philosophical context that has rendered the notion of a unified and autonomous self substantially inoperative? In Part VII of the Theory of Moral Sentiments, added to the sixth edition (1790), a short but incisive critique of Hutcheson’s “moral sense” signals what Smith had taken to be his overall retreat from an internalist account of human agency. As Smith points out, Hutcheson’s hypothesis of a moral sense operating wholly unconditioned by external contingency or interest actually breaks down when certain qualities are introduced that, “belong[ing] to the objects of any sense, cannot, without the greatest absurdity, be ascribed to the sense itself” (TMS, 323). What Smith objects to is the very supposition that vice and virtue could ever be established by an inner sense operating independent of any contextual awareness or feedback; and his fictitious scenario of a “man shouting with admiration and applause at a barbarous and unmerited execution, which some insolent tyrant had ordered,” is meant to illustrate the way that moral judgment is fully enmeshed with some ambient, social dynamic. Characteristically, Smith thus flags the susceptibility of Hutcheson’s “moral sense” to possible misjudgments by depicting how the surrounding spectators should “feel nothing but horror and detestation” at such an inappropriate response, and how they are likely to “abominate him even more than the tyrant” who had ordered the execution. As he sees it, absent a reciprocally constituting social awareness, Hutcheson’s moral sense lacks all criteria. For Smith, then, there can be no such thing as synderesis, for to hypothesize a purely inward source of moral orientation is to hazard a “perversion of sentiment” that may rise to “the very last and most dreadful stage of moral depravity” (TMS, 323).

Yet Smith’s alternative of moral sentiments bound up with a spectatorial infrastructure that supplants judgment with “wonder and applause” is not without problems of its own. Above all, it is not easy to conceive of “sentiment,” as Smith repeatedly insists we must, as socially conditioned and only arising from “mutual regard” (TMS, 39). Would not such sentiment be but a more genteel term for a prolonged bout of Freudian “transference” (Übertragung) or “projection”? As Vivasvan Soni notes, “affective communion does not occur simply on the evidence of the other’s affect; a narrative is required to engender analogous affects in us.” Yet if “a narrative understanding of sentiments is already built into [Smith’s] theory,” the self-certifying nature of affect allows us to take that narrative for granted and allows us to “focus our attention on the emotional state of the other without regard for narrative.”32 To this one might add that a “narrative” that tacitly regulates the specific dynamics of affect, both as it is expressed and received, should perhaps be thought of as an underlying “grammar” rather than narrative. In what is to follow, we shall think of it under the heading of “behavior.”

Moreover, how is a spectator to “admire the delicate precision of [someone else’s] moral sentiments” if not by autonomously interpreting the symbolic and gestural language ostensibly “expressing” such sentiments? If sentiment is inter-subjectively constituted, would this not preclude any critical or counterintuitive perspective on it? Does the Theory of Moral Sentiments recognize the need for a genuine hermeneutic of sentiment? To judge by contemporary responses to Smith’s, the answer is “no.” Characteristic of early reactions to the Theory is a peculiar blend of the enthusiastic and the unreflexive; reporting to the author that his Theory of Moral Sentiments “is in the hands of all persons of the best fashion,” William Robertson notes “that it meets with great approbation both on account of the matter and stile” and that it is “impossible for any book on so serious a subject to be received in a more gracious manner.”33 A notice published in the Monthly Review echoes this appraisal and singles out the author’s “agreeable manner of illustrating his argument, by the frequent appeals he makes to fact and experience.” While declining to endorse Smith’s account of sympathy outright, the reviewer depicts the book’s argument as “extremely ingenious and plausible.”34 Finally making good on a promise (to Hume) that he would write to the author of the Theory of Moral Sentiments, Edmund Burke is quick to zero in on the favorable ratio between intellectual gain (large) and intellectual effort (small) involved in the perusal of Smith’s work: “I do not know that it ever cost me less trouble to admit to so many things to which I had been a stranger before.” Like so many other readers, Burke also dwells on “those easy and happy illustrations from common Life and manners in which your work abounds more than any other that I know by far,” and he echoes similar praise by referring to Smith’s “lively and elegant” style.

Writing for his Annual Register, Burke extends the latter comment in a particularly revealing manner by remarking how Smith’s “language is easy and spirited . . . it is rather painting than writing.”35 Smith’s stylistic accomplishments are well received on the Continent, too, with an early French review approving “par la beauté et la noblesse des sentiments.” More surprising might be the same review’s affirmation that “religion is respected throughout the work [que la Religion y est par-tout respectée],” considering that the same had been affirmed in the Monthly Review—viz., that “the strictest regard [is] preserved, throughout, to the principles of religion.”36 That both Anglo-Protestant and Catholic readers should reach such uniform (if notably vague) conclusions on the lingering question of Smith’s attitude to religion brings us to the one point observable throughout virtually all of the responses to Smith’s book: none of them actually identify, let alone engage, its thesis. Bearing a striking resemblance to Thomas Gray’s contemporaneous “Elegy Written in an Country Churchyard,” Smith’s book draws responses that prove both consistently sympathetic and positively non-cognitive. Working by accretion, the Theory sets forth “a system of moral philosophy able to propagate itself by accommodating ever more examples of its central claims. Where Hume had sought, through his Treatise, to regulate social systems by means of systematic reflection on sympathy, Smith opted for a form of regulation that denied the appearance of systematicity.”37 The peculiar appeal of the Theory of Moral Sentiments stems from its having so successfully harnessed the power of the moral commonplace as both medium and message. Conspicuous for the low interpretive demands it imposes on the reader, Smith’s Theory truly embodies its implicit conception of virtue, morality, and the good as so many behaviorally instantiated sentiments.

Yet another set of questions to press concerns the traditions of inquiry, ancient and modern, that are being linked by Smith’s unprecedented emphasis on the sociality of the emotions. The two conceptual traditions that intersect in his work are classical (and early modern) Stoicism and, from what we may call the “future-past,” the twentieth-century project of “behaviorism,” whose conceptual roots have often been traced back to Locke’s and Hume’s epistemology, yet whose conceptual intent is uniquely anticipated in the Theory of Moral Sentiments. Though ostensibly consumed with psychological discriminations of all sorts, Smith’s argument undertakes a comprehensive de-psychologizing of human agency. Hints to this effect come early, and they typically take the form of shifting focus from the (seemingly inscrutable) inner phenomenology of passion toward an analysis of its social, mediated character: “Sympathy . . . does not arise so much from the view of the passion, as from that of the situation which excites it.” By choosing emotions such as only ever occur in social settings (flustering, awkwardness, embarrassment, wit, etc.) rather than those liable to crystallize by way of sustained introspection (doubt, despair, hope), Smith draws attention to sociality as positively constitutive of the meaning of “passion”—a term that throughout the Theory of Moral Sentiments is all but synonymous with “emotion” yet categorically distinct from “sentiment.” Smith’s passing observation that “we blush for the impudence and rudeness of another” (TMS, 12) emphasizes how emotion involves an element of imagination, projection, and transference. Its semantics are conditioned not by an inner certitude but, rather, by an agent’s appraisal of a social situation as it is (probabilistically) taken to be perceived by multiple individuals. Moreover, Smith contends that this sublation of individual emotions into socially constituted sentiments is not some adventitious occurrence but is teleologically inscribed into the passions themselves. For intrinsic to every emotion there is a distinctive “motion,” a gravitational force or tropism oriented toward the other person, much like Jacques Lacan’s account of the “symptom” as something both real and intelligible only insofar as it is directed at an addressee.

Yet for Smith, inasmuch as that other is conjectured to be experiencing the same inner state, she or he is not simply the recipient of some seemingly interior, private feeling. Rather, she mimetically confers reality on that emotion as a socially viable sentiment: “Whatever may be the cause of sympathy, or however it may be excited, nothing pleases us more than to observe in other men a fellow-feeling with all the emotions in our own breast” (TMS, 13). If the first part of this sentence discounts the Platonic and Christian conception of emotion as an inner “source,” it is not so much that Smith means to reject that view outright but that it no longer holds a central role within his overall philosophical project. Instead, an emotion in his Theory acquires distinctness, significance, and prima facie reality only if and inasmuch as it is reflected back by another self. What Smith calls “fellow-feeling” furnishes, if not the material ground (ratio essendi), then certainly the ground whereby an emotion becomes positively intelligible and significant (ratio cognoscendi). To the extent that “our approbation is ultimately founded upon a sympathy or correspondence [of sentiments]” (TMS, 17), an emotion can be credited with epistemological standing and moral significance only insofar as it reflects and reaffirms an underlying social consensus.

Central to the Theory of Moral Sentiments, and widely recognized as the book’s main contribution to modern thought, is Smith’s spectatorial model of moral cognition. Arguing that every sentiment encrypts an instance of approbation or disapprobation, Smith suggests that any judgment ventured about the merits or demerits of one’s own (proposed) action is achieved by imagining whether the motive underlying it would meet with someone else’s approval. Merely to put the question in that way is to have already migrated from an introspective to a transferential account of moral value. Sentiments in Smith do not identify but constitute their objects; social “facts” do not precede, let alone exist independent of, but are only realized by the interpretive and evaluative process of social exchange. Having famously enjoined his readers that “we must become the impartial spectators of our own character and conduct,” Smith notes how any instance of subjective “approbation necessarily confirms our own self-approbation.” The praise of others “necessarily strengthens our own sense of our praiseworthiness,” an interesting update on the problem of Aristotelian megalopsychia, which had vexed virtue ethics for more than two millennia. For Smith, praiseworthiness is not inwardly felt, let alone unilaterally asserted by a moral agent about herself or himself. Rather, it is an inference compelled by how others have assessed one’s own conduct: “So far is the love of praise-worthiness being derived altogether from that of praise; that the love of praise seems, at least in a great measure, to be derived from that of praise-worthiness” (TMS, 114).

Smith’s claim that a spectatorial type of moral judgment performatively creates the very values to which individuals take themselves to be responding tends to come in two preferred metaphors, the theatrical and the optical. In one of his programmatic accounts of the impartial spectator (109–113), Smith thus insists that this virtual agent “is the only looking-glass by which we can, in some measure, with the eyes of other people, scrutinize the propriety of our own conduct” (TMS, 112). That all moral judgment thus rests on, and is conditioned by, the a priori sociality of human beings becomes, according to Smith, evident if one entertains the counterfactual scenario. For a completely isolated self such as the noble savage of Rousseau’s second Discourse “is provided with no mirror which can present . . . to his view” the “beauty or deformity of his mind” any more than that of his body. Yet “bring him into society, and he is immediately provided with the mirror which he wanted before. It is placed in the countenance and behaviour of those he lives with, which always mark when they enter into, and when they disapprove of his sentiments; and it is here that he first views the propriety and impropriety of his own passions” (TMS, 112). For strategic reasons, we note the peculiar conjunction of a regime of vision (“countenances” looking out, and being looked into, like looking glasses) with a term suddenly rising to prominence in post-metaphysical social theory: “behavior.” Striking about both is the absence of any deliberative, discursively reasoning component. Indeed, behavior and vision alike are valued precisely on account of their supposedly seamless, transparent, and effortless operation. At least for Smith—herein markedly differing from the forensic and evolving concept of vision that we find in Wordsworth, Goethe, Darwin, Ruskin, and Gerard Manley Hopkins—neither sight nor behaviors involve any sustained cognitive effort.38 Furthermore, both the concept of vision and of behavior can only signify on the basis of an already established (and likely complex) matrix of socio-cultural values. To derive moral approval or disapproval from the gaze of the other, like the kind of implicit social orientation denoted by the concept of “behavior,” requires that a grammar of social propriety and order has already been internalized and thus has enabled the self to look for specific meanings rather than gazing outward without either purpose or comprehension. Yet to premise moral orientation on a specular and performative model of sociality comes at a price. At the very least, it means that moral cognition and judgment have been demoted from inner reflection to a mere reflex gesture, and that the counterfactual or creative potential of thought has yielded to the mimetic affirmation of some already established, albeit inarticulate notion of social propriety.39

To develop a clearer sense of Smith’s overarching objectives in the Theory of Moral Sentiments, and its overall implications for subsequent accounts of practical reason and ethics, one must scrutinize precisely this transposition of passion into sentiment. What prompts Smith’s sweeping re-description of the inner life is a desire to restore to the self the epistemological coherence and moral authority that it had lost in Locke’s Essay, Mandeville’s Fable, and, especially, in Hume’s Treatise and second Enquiry. As a result, stress is no longer placed on the ideational content of emotion, a point on which Smith appears cannily agnostic. Rather, scrutiny is brought to bear on the dramaturgy of emotion as it is “actualized” (in the Hegelian sense of Verwirklichung) in social exchange. The reality of emotion, Smith suggests, proves altogether inseparable from its social phenomenology, that is, from its mode of appearance as “behavior.” For passion to be commensurate with (rather than disruptive of) social relations, it must take the form of a recognized symbolic practice (as opposed to subjective “expression”). Such practice, Smith shows, is likely to take the form of subtly evaluative discriminations whose referent is no longer some separate object “out there” but, rather, a set of circumstances socially shared and acknowledged as pertinent to the present situation. Insofar as it has been successfully transmuted into outward symbolic practice—viz., has been enacted as a specific “sentiment” for others—and only then, does passion acquire meaning and social value. Its true significance resides not in the moral value or appraisal that it ostensibly “expresses” about some particular issue or fact. Rather, sentiment by its very nature accredits and reinforces a shared understanding of social relations as they are presently taken to be constituted.

The language of Smith’s sentiment ultimately functions as a meta-language in Roman Jakobson’s sense of appraising a prevailing “code” rather than denoting a distinct “referent.”40 As Smith puts it: “to approve of the passions of another . . . as suitable to their objects, is the same thing as to observe that we entirely sympathize with them” (TMS, 16); and again: “we approve of another man’s judgment, not as something useful, but as right, as accurate, as agreeable to truth and reality: and it is evident we attribute those qualities to it for no other reason but because we find that it agrees with our own” (TMS, 20). Moral sentiments, then, are not stand-alone units of subjective experience. On the contrary, they constitute prima facie acts of “assent” or value judgments about a social situation, and as such they do not signify an inner (mental) action but a social transaction. By its very nature, such a type of “agreement may produce peace but it cannot produce truth.”41 Anticipating Kant’s judgment of taste, which “imputes” (ansinnen) universal agreement to those for whom it is voiced, Smith here is able to recover from the extreme nominalism of Hobbes, Locke, and Hume whose strictly epistemological focus had led them to characterize passion as a hermetic, ineffable, and altogether transient phenomenon.

While there is an obvious and significant Stoic dimension to Smith’s view of moral sentiments as quasi-judgments, his Theory no longer furnishes any frame of reference for such evaluation independent of (or anterior to) the social interaction whose success or failure such judgments ratify. In this regard, Smith’s appeal to “the great machine of the universe” and the “secret wheels and springs which produce” its countless appearances (TMS, 19), or his later affirmation that “in every part of the universe we observe means adjusted with the nicest artifice to the ends which they are intended to produce” (TMS, 87), is less a nod to Stoic physics than an echo of the pervasive deism and the commonplace, “just-so” story endlessly recycled by eighteenth-century natural theology, a subject on which Smith repeatedly lectured between 1752 and 1764 as professor of moral philosophy at the University of Glasgow. Particularly revealing in this regard are the uninspired, token references that Smith inserted into the sixth edition (1790) of the Theory of Moral Sentiments, such as his tribute to “the great Director of the universe” and to “the idea of that divine Being, whose benevolence and wisdom have, from all eternity, contrived and conducted the immense machine of the universe, so as at all times to produce the greatest possible quantity of happiness.” There is a strong deist tendency in this notion of a God who makes sure the cosmic train runs on time but who has delegated day-to-day operations of the material world to “his viceregent upon earth,” including the task of “superintend[ing] the behaviour of his brethren” (TMS, 130). While affirming that “the administration of the great system of the universe . . . is the business of God and not of man,” Smith’s quick shift of focus strongly suggests that his God remains de facto invisible and, lacking credible revelation, has become all but irrelevant to moral thought: “To man is allotted a much humbler department, but one more suitable to the weakness of his power and to the narrowness of his comprehension; the care of his own happiness, and that of his family, his friends, and his country” (TMS, 236–237).

The focus on ethics, which for the ancient Stoics was altogether inseparable from their physics, has here been compartmentalized as a separate province governed by finite, fallible, and self-determining human agents. There is no vertical connection between the judgments and appraisals ventured by the latter and the divine logos as disclosed in a cosmological system.42 Indeed, precisely because judgment appears terminally estranged from any transcendent, metaphysical source, its present-day incarnation as “moral sentiment” can only be focused on, and is exclusively licensed by, endlessly shifting social circumstances. It is just this definitive separation of judgment from ontology that attests to the modernity of Smith’s entire project. Additionally, it bears recalling that whatever “social reality” happens to be contingently at issue in a case of moral sentiment does not pre-exist the specific judgment in question—say as a Platonic idea or Aristotelian/Thomist “substantial form.” On the contrary, such reality is only instantiated by a vast number of affective micro-judgments and acts of transference.

In this regard, Smith’s choice of laughter as an example is both revealing and shrewd inasmuch as it vividly displays the non-propositional and quintessentially social nature of moral cognition:

He who laughs at the same joke, and laughs along with me, cannot well deny the propriety of my laughter . . . If I laugh loud and heartily when he only smiles, or, on the contrary, only smile when he laughs loud and heartily; in all these cases, as soon as he comes from considering the object, to observe how I am affected by it, according as there is more or less disproportion between his sentiments and mine, I must incur a greater or less degree of his disapprobation: and upon all occasions his own sentiments are the standards and measures by which he judges of mine. (TMS, 16–17)

In ways that Keats would later explore to rich aesthetic effect, Smith identifies the misalignment of emotion as prima facie evidence for the way that all moral values are social in essence. That is, embarrassment of the kind here described shows Smith to understand moral interaction to be no longer about inner certitudes or metaphysical norms and aspirations. Rather, the objective at hand is the success (or failure) of social interaction itself.43 Irving Goffman thus notes that contrary to a “breach” of some moral norm, which is likely to “give rise to resolute moral indignation,” embarrassment amounts to a constitutively social emotion. For even as “the expectations relevant to embarrassment are moral . . . we should look to those moral obligations which surround the individual in only one of his capacities, that of someone who carries on social encounters. The individual, of course, is obliged to remain composed, but this tells us that things are going well, not why. And things go well or badly because of what is perceived about the social identities of those present.”44

It is no accident that Smith and Goffman should both have happened upon the seeming benefits of embarrassment. Long before there was to be a discipline called sociology and a subsidiary specialization known as behaviorism, Smith offers a sophisticated (if deeply problematic) “account of the formation of groups”; and it is this account’s structural reliance on the specter of embarrassment that prompts us to wonder “what sort of unity” sympathy could possibly generate.45 As it turns out, the psychological constellation known as embarrassment operates at all times, and not merely when it is manifestly unfolding—along with all the outward symptoms of blushing, flustering, sweaty palms, and stammering speech so prominent in John Keats’s canny fusion of depth-psychology and class-consciousness. In a remarkable passage, Smith thus contends that both the outward conduct and inner constitution of the self are at all times circumscribed by the specter of embarrassment. Even at its most introspective, the individual is intelligible to itself only as a socially constituted reality. However free from embarrassment at moments of introspection, it is precisely the constant possibility of some performative misadventure in public that furnishes the motivational prompt for how to cultivate one’s persona:

The man of real constancy and firmness, the wise and just man who has been thoroughly bred in the great school of self-command . . . has never dared to forget for one moment the judgment which the impartial spectator would pass upon his sentiments and conduct. He has never dared to suffer the man within the breast to be absent one moment from his attention. With the eyes of this great inmate he has always been accustomed to regard whatever relates to himself. This habit has become perfectly familiar to him. He has been in the constant practice, and, indeed, under the constant necessity, of modeling, or of endeavouring to model, not only his outward conduct and behaviour, but, as much as he can, even his inward sentiments and feelings, according to those of this awful and respectable judge. He does not merely affect the sentiments of the impartial spectator. He really adopts them. He almost identifies himself with, he almost becomes himself that impartial spectator, and scarce feels but as that great arbiter of his conduct directs him to feel. (TMS, 146–147)

This is about as vivid and palpable an account of the concept of “introjection” (a term first coined by Sándor Ferenczi in 1909) as one could ask for. Unlike Ferenczi’s and Freud’s subject, however, Smith’s impartial spectator is stripped of all magical or fantasy-like elements and, thus, appears much closer to Lacan’s model according to which “introjection is always the introjection of the speech of the other.”46 Introjection thus attests to the incomplete and seemingly illegitimate nature of the individual understood in strictly inward, mentalist terms. Smith thus insists how the virtual superego of the impartial spectator is not merely to be accommodated in “outward conduct and behaviour,” but that it is to be internalized as the real substance of moral agency. The “great inmate” or “man within the breast” is not merely some occasional complement of the self, nor is he to be construed as some didactic allegory of righteousness or virtue.

At first glance, the above passage appears to present us with a genuinely (neo-) Stoic model of self-cultivation. Still, the emphasis on “self-satisfaction” stands out as discordant, if for no other reason than that it revives the dilemma that Aristotelian megalopsychia had bequeathed Christian virtue ethics, which obviously looks upon pride with acute misgiving. Moreover, Smith here tells essentially a story of emancipation from a debilitating inner life. Stoic habituation—whose telos was wisdom—yields to modern, neo-Stoic self-discipline, whose telos is approval; and it is only for the sake of such approval (and the successful socialization it betokens) that the self embraces the surrogate or virtual interiority of the “man within the breast.” The person qua unique, spiritual self is expunged, much in the way that seventeenth-century Calvinist and Jesuit thought had appropriated the neo-Stoic language of self-discipline. With the virtual agency of the impartial spectator having supplanted the metaphysical idea of the person, Smith’s Theory of Moral Sentiments no longer grants epistemological legitimacy to the idea of the person—understood as a unique, unclassifiable, and “incommunicable” being (as Boethius had defined it). Concurrently, the emotions enjoy epistemic standing only insofar as they can be converted into socially sanctioned sentiments. At the same time, the inner life has reality and legitimacy solely as a socially constituted and accredited performance, a process for which Smith does not (indeed, cannot) identify any telos or terminus ad quem. Echoing Smith’s position, Goffman’s closing discussion of the “social function of embarrassment” thus stresses the extent to which this particular affect is properly generative of the self, rather than being the inward property of an already constituted individual: “By showing embarrassment when he can be neither of two people, the individual leaves open the possibility that in the future he may effectively be either. His role in the current interaction may be sacrificed, and even the encounter itself, but he demonstrates that while he cannot present a sustainable and coherent self on this occasion, he is at least disturbed by the fact and may prove worthy at another time.”47 Both Smith’s impartial spectator as a behavioral template to be introjected, and Goffman’s embarrassment as the exemplary (because intrinsically “social”) emotion betrays an eagerness to abandon mentalistic and psychologizing conceptions of moral meaning and agency. Crucial to understanding Smith’s objectives is that “sentiments” are not, properly speaking, inner states but arise from our unwitting negotiation of a complex grammar of social values and behavioral patterns or averages.

One of the abiding consequences of this de-psychologizing, quasi-behaviorist reorientation of moral theory is the contraction of any temporal perspective of agency to the currently prevailing social dynamics. Inasmuch as the teleological nature of “action” has been supplanted by concern with the success of present “interaction,” a sentiment-based “appraisal of social facts no longer distinguishes between motive, implementation, and consequences of discrete actions.” Instead, the social field appears “determined by complex interdependencies wherein events and affective states reciprocally induce and generate one another.”48 The atrophying of volition, intentionality, and action is palpable when, in his discussion of the impartial spectator, Smith emphasizes the non-participatory nature of this model: “we must imagine ourselves not the actors, but the spectators of our own character and conduct” (TMS, 111). To be sure, just a few pages earlier Smith had stressed that sentiment without action is but “indolent benevolence,” and that “the man who has performed no single action of importance, but whose whole conversation and deportment express the justest, the noblest, and most generous sentiments, can be entitled to demand no very high reward” (TMS, 106). Yet inasmuch as Smith regards the value of action to depend on the motives with which it is being undertaken, it is not the action but the social currency of the motive qua sentiment that furnishes the basis for the agent’s moral appraisal by others. Indeed, Smith’s concerted attempt to move beyond internalist accounts of moral meaning at times causes him to speak rather diffidently about the intrinsic value of ideas, however large or small. In a revealing aside, Smith thus notes how difference of opinion is easily tolerated where no present business is pending: “I can much more easily overlook the want of this correspondence of sentiments with regard to such indifferent objects as concern neither me nor my companion.” Yet Smith here is not thinking of weather patterns in eastern Mongolia or the imbalance of trade between ancient Athens and Sicily but, in fact, of just about any topic whatsoever, including “that picture, or that poem, or even that system of philosophy, which I admire. They ought all of them to be matters of great indifference to us both.” Indeed, it appears that the intrinsic value and pragmatic relevance of moral sentiments requires the de facto expurgation of all ideational content from them. Moral sentiments merely simulate propositional utterances, while actually furnishing a merely formal assurance of sympathy or “fellow-feeling for the misfortunes I have met with” (TMS, 21).

Early on in Smith’s Theory, this virtual logic of “sentiment” is stated quite explicitly. To be sure, everyone “passionately desires a more complete sympathy” than what can reasonably be expected of others. Yet precisely because this narcissistic desire (“to see the emotions of their hearts, in every respect, beat time to his own”) cannot be satisfied in unadulterated form, every individual feels induced to lower his passion to that pitch, in which the spectators are capable of going along with him: “He must flatten, if I may be allowed to say so, the sharpness of his moral tone, in order to reduce it to harmony and concord with the emotions of those who are about him.” As Smith well realizes, this transposition does not simply involve a reduction in the force of sentiments but amounts to a qualitative transformation. Consequently, the socially successful and, in that rather unique sense, “moral” agent thus is imbued with “the secret consciousness that the change of situations, from which the sympathetic sentiment arises, is but imaginary” and thus “not only lowers it in degree, but, in some measure, varies it in kind, and gives it a quite different modification.”49 The qualitative shift here is from emotion as a unique, inner, and unverifiable occurrence of questionable rational standing to sentiment as a form of “behavior.” By definition, “behavior” is an abstract concept. It signifies a “class” or “type” of outwardly manifest, observable, and recurrent practices and, by implication, it appraises human action only within a matrix of average practices or what, in identifying his overarching concern with the “propriety of every passion,” Smith calls “emulation . . . of the excellence of others” (TMS, 114) and “a certain mediocrity” (TMS, 27; italics mine). Similar formulations recur throughout the Theory of Moral Sentiments, such as in Smith’s acknowledgment that, happily, “the reflected passion . . . is much weaker than the original one” (TMS, 22). Moral agency thus is said to pivot on a kind of actuarial balancing of emotional states, an aspiration still couched in neo-Stoic language as the virtue of “temperance” (TMS, 28) or, in statistical terms, as aiming at “the most ordinary degree of kindness or beneficence” (TMS, 80; italics mine). Just as Locke, Mandeville, and Hume had dissolved the metaphysical concept of the will into a radically nominalist model of ephemeral and heterogeneous passions, so the concept of passion in turn is now systematically being converted from a phenomenon open to introspective appraisal into a reflex-like, behavioral pattern. To sharpen the point, a brief and necessarily selective review of behaviorism’s central claims and objectives is needed.

Modern behaviorism’s principal objective is to render mental phenomena objectively verifiable and actuarially measurable, a project that, in the view of behaviorism’s founder, John B. Watson, is likely to succeed only once the whole mentalist vocabulary of consciousness, introspection, emotion, and so forth has been definitively abandoned. It is this proposed conversion of enigmatic qualia into quantifiable, social facts that explains the sudden appeal of “behavior” as a new paradigm for the manifestly ailing discipline of experimental psychology just before World War I. As Watson puts it in his 1913 manifesto (“Psychology as the Behaviorist Views It”), “the end is the production of mental states that may be ‘inspected’ and ‘observed.’” In asserting “the independent value of behavior material” and foreswearing traditional introspective concepts (“we will no longer have to work under false pretenses”), Watson intends nothing less than to abandon all mentalist and introspective terms, “to discard all reference to consciousness” and, in quasi-revolutionary fashion, “to throw off the yoke of consciousness.”50 The aspiration here is not merely to extricate social analysis from a half century of inconclusive experimental psychology, which according to Watson “has failed signally” (163) but to expunge a 2,500-year-old tradition (going back to Plato and Augustine) that had traced all moral and social meaning back to a variety of “inner” processes.51 Vaguely following in the footsteps of Frege, modern behaviorism contests the reality or, at least, the relevancy of mental phenomena and introspective methods. Watson, whose manifesto will have to suffice as synecdoche for the larger movement, thus envisions a psychology that will “never use the terms consciousness, mental states, mind, content, introspectively verifiable imagery, and the like” (166). The basic objection to the entire conceptual inventory associated with inner experience is that the states to which these concepts refer are knowable only insofar as the observer is able to establish a strong correlation between behavioral patterns. At most, then, mind constitutes an “inference” that may be drawn on the strength of such patterns, though for Watson, and even more so for his successors, it becomes increasingly unclear why anyone might wish to draw such an inference at all.

In time, a more strident variant known as “analytic” behaviorism abandons any residual mentalist vocabulary by insisting that there is properly no need to speculate on inner states or dispositions. On this view, behavior is no longer evidence for a mental process relative to which it might be said to function epiphenomenally (i.e., behavior as the appearance of inner states). Instead, analytic behaviorism conceives behavior as something that “can be described and explained without making ultimate reference to mental events or to internal psychological processes.”52 Here, then, the stabilization of (formerly) psychological, mental states in the form of social conditioning and manifestly associative patterns of social cognition is said to render human behavior intelligible. Yet behaviorism’s wholesale abandonment of psychological and spiritual concepts in favor of externally observable and measurable criteria of human behavior is pressed further. True to its origins in a “logical positivism” that had restricted the meaning of scientific statements to experimental conditions such as could be inter-subjectively observed (a.k.a. “verificationism”), modern behaviorism draws on the same spectatorial logic first articulated in Smith’s Theory of Moral Sentiments. It cannot surprise, then, that behaviorism’s disciplinary criterion of strict verifiability reflects an underlying investment in the predictability of its objects of inquiry. Human behavior, if it is to be known at all, must be re-described in such ways as will render its subjects predictably responsive to external mechanisms of control. Watson thus opens his 1913 manifesto by classifying this new field as “a purely objective experimental branch of natural science” whose “theoretical goal is the prediction and control of behavior.” What drives behaviorism’s proposed dissolution of the enigmatic qualia of human consciousness into causally determinate and quantifiable behavioral patterns is a quest for applicable, instrumental knowledge. The investigation into predictable and verifiable human responses to stimuli is undertaken solely for the sake of “conditioning.” Watson, it seems, might have approvingly chuckled at the old joke, which starts with the question of why of late lawyers have replaced rats as the behaviorist’s preferred test subject (Answer: “1. There are more of them; 2. Experimenters run no risk of forming attachments to their subjects; and 3. There are certain things that even rats won’t do”). Then again, the joke’s underlying premise might have eluded his naturalist viewpoint altogether. For the success of behaviorism as a discipline clearly hinges on pushing the re-description of formerly mental phenomena to the point where the line between animals conditioned by stimuli and humans actuated by motives has been permanently erased. Watson explicitly commits himself to supplanting the hermeneutic concept of “experience” with a strictly material and physiological account of the way that stimuli can be shown to “cause” particular types of behavior.

It is precisely their apparent “mindlessness” that recommends stimuli as the centerpiece of behaviorism’s (in origin Hobbesian) attempt to posit a “unitary scheme of animal response, [which] recognizes no dividing line between man and brute.”53 Still, the final challenge that behaviorism has yet to meet involves accounting for “more complex forms of behavior, such as imagination, judgment, reasoning, and conception.” It is only in a long and truly startling footnote late in the essay that the other shoe drops, with Watson now venturing as “tenable” the

hypothesis that all of the so-called “higher thought” processes go on in terms of faint reinstatements of the original muscular act (including speech here) and that these are integrated into systems which respond in serial order (associative mechanisms) . . . Paucity of “imagery” would be the rule. In other words, wherever there are thought processes there are faint contractions of the systems of musculature involved in the overt exercise of the customary act, and especially in the still finer systems of musculature involved in speech. If this is true, and I do not see how it can be gainsaid, imagery becomes a mental luxury (even if it really exists) without any functional significance whatsoever . . . I should throw out imagery altogether and attempt to show that practically all natural thought goes on in terms of sensori-motor processes in the larynx (but not in terms of “imageless thought”).54

To be sure, as with most other manifestos allowances have to be made for the exuberant, at times utopian fervor with which Watson sketches his arguments here. Moreover, his analeptic conception according to which thought-like effects are produced by “motor habits in the larynx” would appear to receive ample corroboration by what transpires on most Sunday morning political talk shows and most elsewhere in the feeding frenzy of hourly news-cycles today. Nonetheless, Watson’s suggestion that “faint contractions” in the musculature of the larynx have, for the past 2,500 years, been misidentified as phenomena of mind, consciousness, imagery, and so forth is likely to disconcert even the most hardened of social constructivists. While there is good reason to view Smith as a precursor of modern behaviorism, it would be gratuitous, not to mention anachronistic, to hold his Theory of Moral Sentiments accountable for the conceptual problems eventually experienced by modern behaviorism, a field whose intellectual currency, in any event, has lost nearly all its value during the past few decades.55

The present objective, however, is not another, by now redundant critique of behaviorism. Rather, it is to show that the critique of an interiorist or introspective (Christian-Platonist) concept of agency in the eighteenth century did not always assume the overtly anti-humanist character that Hobbes, Locke, Mandeville, and Hume had given to it. For rather than extending the epistemological critique of that model—which, in the wake of Hume’s Treatise, seemed hardly possible—the Theory of Moral Sentiments opts to reconstitute moral meaning as positively arising from patterns of social transaction such as can be ascertained, quantified, and classified by means of sustained empirical observation. Mind in Smith’s account is above all a habit, a virtual relay station for meanings said to originate in social and cultural exchange (speech, manners, gestures, and a vast inventory of codified expressions). Hence, in pressing the question of how Watson could possibly assert that “practically all natural thought goes on in terms of sensori-motor processes in the larynx,” one is ultimately confronted with a distinctly modern (associationist) revival of the Aristotelian and Scholastic concept of habit. To connect habit and habituation with the behaviorist vocabulary of “stimulus-response” and “conditioning”—while not intended to rehabilitate Watson’s extreme theory—is to recognize a strong link between modern behaviorism and the concept of socially induced sentiment that rises to sudden prominence in Smith’s Theory. According to Watson, habit comprises the myriad associative motor processes that justify behaviorism’s basic hypotheses and research program: for “improvement in habit comes unconsciously. The first we know of it is when it is achieved—when it becomes an object.” Without any hesitation, Watson extends this (substantially flawed) observation into an argument to the effect that “the so-called ‘higher thought’ processes” are no different: “I believe that ‘consciousness’ has just as little to do with improvement in thought processes . . . [I]mprovements, short cuts, changes, etc., in these habits are brought about in the same way that such changes are produced in other motor habits. This view carries with it the implication that there are no reflective processes (centrally initiated processes).”56

For present purposes the question becomes how “habit” and “behavior” are linked in modern social theory and how that configuration might differ from a premodern account of habit. A first step is to attend to the specific word(s) associated with the two concepts of habit and behavior, viz., the Latin habitus and the early modern English hauyoure, respectively. In his detailed account of habitus, Aquinas opens by observing that “this word habitus is derived from habere [to have],” and true to form he launches into a complex distinction of meanings, most of which need not concern us.57 What does matter, however, is his decision to follow Aristotle’s cue and define habit as a so-called post-predicament, that is, as a concept made possible by some of the ten categories (praedicamenta), such as quantity, quality, relation, priority, posteriority, etc. Which is to say (still nothing original here) that habit is not a substantial form, is not nature or essence, but something “had” only insofar as it has been made. Yet implicitly, this otherwise simple claim also tells us that, being something “constructed,” habit derives its intrinsic rationality from the ultimate end whose attainment by the human subject it is meant to facilitate. Having established that habit is something constructed over time rather than ontologically given (in which case nothing could be done about good or bad habits), it follows for Aquinas that the concept can only be understood if integrated into a narrative of human flourishing. To be sure, there is nothing wrong with calling habit second nature provided we understand this to mean that it is, precisely, not nature. For present purposes, the most relevant meaning of habit is that of a relation between the “haver and that which is had,” which is to say “a quality.” Having quoted from Simplicius’s commentary on Aristotle’s Metaphysics (ST, Ia IIae Q 49 A 2), Aquinas emphasizes that habits, inasmuch as they bear on theology, arise from strictly “adventitious” dispositions. Like any other quality that has to be cultivated by methodical and sustained practice, we only “have” these qualities in the sense of a possessive relation, which is to say that “they can be lost.”58

Now, in Aristotelian and Thomistic thought anything “made” or created is rational and intelligible only insofar as it can be shown to have been fashioned in accordance with an overarching teleological framework. Minimally, this means that habit must not be confused with nature, instinct, or anything on the order of a cause that is at once materially heterogeneous and efficient or determinative. Recalling Augustine (no. 36 from the Book of 83 Questions), Aquinas thus subtly demurs when it comes to the bishop of Hippo’s claim that an animal tamed by means of rewards or incentives has effectively “commuted its nature into something habitual [quod in earum consuetudinem verterit].” But the habit is incomplete, as to the use of the will, “for they have not that power of using or of refraining, which seems to belong to the notion of habit: and therefore, properly speaking, there can be no habits in them” [quia non habent dominium utendi vel non utendi, quod videtur ad rationem habitus pertinere].” Habit thus is a form of mental causation and in a very specific and limited sense can even be viewed as a precursor of the modern idea of autonomy. Put differently, habit cannot be explained as a result of outward conditioning; it is not, in Aquinas’s phrase, “a disposition of the object to the power” but, rather, an act of self-fashioning or “disposition of the power to the object” (ST, Ia IIae Q 50 A 4). For Aquinas, certainly, habit and the extended process of habituation that gives rise to it have for their logical premise the essential indeterminacy of the human person—what Nietzsche would much later call that “still undetermined animal” (das noch nicht festgestellte Tier).59 Yet for Aquinas, such indeterminacy constitutes no defect but, on the contrary, is the condition for the human person as a rational, free, and self-creating being. As such, it pertains to intellect and will alike, for both are powers capable of varied development, including the possibility of not properly developing at all or even regressing: “Every power which may be variously directed to act, needs a habit whereby it is well disposed to its act.” Crucially, that is, the relation of powers to habits is a result of the fact that powers are not simply “faculties” in the modern, Kantian sense but potentialities in whose nature it is to develop, both in the sense of an ongoing intrinsic cultivation and perfection and with a view to an ultimate end: “Since it is necessary, for the end of human life, that the appetitive power be inclined to something fixed, to which it is not inclined by the nature of the power, which has a relation to many and various things, therefore it is necessary that, in the will and in the other appetitive powers, there be certain qualities to incline them, and these are called habits” (ST, Ia II ae Q 50 A 5).

Aquinas now points out that habit, simply because it is formed and sustained by repeated acts, may be misconstrued as something passively and extraneously received: “if the acts be multiplied a certain quality is formed in the power which is passive and moved, which quality is called a habit” (ST, Ia IIae Q 51 A 2). Yet this is a misleading view to take, for all repetition implies and produces difference, rather than amounting to an invariant and mindless mechanism corrosive of rational agency (as most Enlightenment thinkers tend to suppose). For an example, one may think of a violinist practicing over and over her scales or some intractable passage in a score. She will initially do so by monitoring all her movements with great alacrity, deactivating those uncalled for (redundant movements of the bowing arm, for example) and fine-tuning those that are crucial for the end (of playing the piece well). Cumulative practice will gradually diminish the need for such intentional awareness of movement as each instance of repetition imperceptibly stabilizes and perfects interval spacing, position shifts, vibrato control, bow markings, and countless other details susceptible of being assimilated to “muscle memory.” The result is a freeing up of mental awareness for higher-level interpretive considerations (dynamics, phrasing, ensemble work, etc.). Habit becomes a substitute for an instinct of which human beings in particular seem to be largely deprived and, as such, rebuilds a strong, seemingly spontaneous connection between will and nature.

Consistent with such a scenario, Aquinas thus remarks that, unlike the stable properties of material substances, habituated actions and passions are susceptible of increase. Put differently, the law of habit is nothing like the law of inertia, for “other qualities which are further removed from substance, and are connected with passions and actions, are susceptible of more or less, in respect of their participation by the subject.” Against the second objection (referencing Aristotle’s Physics, 246^a13) that habit is a perfection term and, thus, does not admit of any “more or less,” Aquinas notes that in the sphere of human action perfection is not constrained by an already existing, determinative nature but, on the contrary, continually helps to define that nature more fully: “habit is indeed a perfection, but not a perfection which is terminal in respect to its subject; for instance a term giving the subject its specific being.”60 The following article (Q 52 A 2) reinforces the basic conception of habit and perfection as jointly delineating the teleological character of human development. Within the context of human action and habituation, repetition and perfection cannot be understood as accumulative but, rather, as transformative: “such increase of habits and other forms is not caused by an addition of form to form; but by the subject participating more or less perfectly, in one and the same form [sed fit per hoc quod subiectum magis vel minus perfect participat unam et eandem formam].” What stands out about the classical conception of habit presented here, and bearing close affinity to Aristotle’s account in the Nicomachean Ethics, is the active, self-determining, and teleological character of habit. In its very essence, then, habit is transformational and dynamic, for its very cultivation presupposes an overarching end and, in turn, shapes and develops a conscious intelligence dedicated to its attainment.

If modernity has not looked kindly on habit, this is the case not because of its avowed overcoming of the alleged intellectual shortcomings of habit and habituation but, on the contrary, because if has proven itself unable to grasp their active, evolving, and purposive structure. Descartes and Kant in particular variously identify habit with inauthenticity, mindless routine, a numbing of sense perception, and, especially, purely mechanical and empty repetition.61 Though just as critical of habit’s alleged epistemological limitations, Hume only differs from Descartes and Kant in that he sees habit, or “custom,” as providing the self with a kind of shelter from the epistemological paralysis where his own skepticism had left it. To understand Smith’s overall affirmative (if deeply problematic) view of habit as “behavior,” it helps to complement Aquinas’s account of habit with Félix Ravaisson’s early phenomenological discussion of habit presented in his remarkable 1838 thesis De l’habitude (Of Habit), a book that was to exercise considerable influence on Bergson, Heidegger, Sartre, Emmanuel Levinas, and Paul Ricoeur. Like Aquinas, Ravaisson sharply distinguishes between the law of inertia and self-sameness that defines physical substance and the law of habit that cannot be applied to the “empire of immediacy and homogeneity that is the Inorganic realm.” Inasmuch as habit is to be understood as the internalization of change as (second) nature, “nothing . . . is capable of habit that is not capable of change.” Habit is a process, a law of growth and development, and hence a manifestation of reason. Opposing the modern critique of habit as mindless, mechanistic, and invariantly repetitive, Ravaisson emphasizes that habituation is not even conceivable without “a centre at which reactions arrive and from which actions depart,” which is to say, consciousness: “in consciousness . . . the same being at once acts and sees the act; or, better, the act and the apprehension of the act are fused together.”62

For Ravaisson, then, the phenomenon of habit is only intelligible within an overarching “economic” model of consciousness that measures how and when different quantities of energy are being expended. Presaging similar arguments in Freud’s meta-psychological writings of 1911–1915, Ravaisson thus sees consciousness not only as intentionality but also as a concomitant awareness of the energy required to work through the object or phenomenon with which it happens to be engaged: “its force is its own measure, [and] it is also measured in that it measures itself out sparingly, making its present energy proportional to the resistance it has to overcome.” Crucial for understanding habit is to notice how consciousness at all times monitors its own expenditure of energy and, like any economic system, seeks to minimize that expenditure in proportion to the task at hand: “Effort is . . . not only the primary condition, but also the archetype and essence, of consciousness.” Now, if habituation helps to draw down that energy expenditure, it does so without compromising the individual’s ability to fulfill a given task; indeed, it facilitates execution of that task while also freeing up consciousness for other purposes. Repetition diminishes both the effort of consciousness and, thus, the consciousness of effort: “As effort fades away in movement and as action becomes freer and swifter, the action itself becomes more of a tendency, an inclination that no longer awaits the commandments of the will but rather anticipates them, and which escapes entirely and irremediably both will and consciousness . . . In this way, then, continuity or repetition brings about a sort of obscure activity that increasingly anticipates both the impression of external objects in sensibility, and the will in activity.” As volition yields to inclination, we can see how fully psychological and physiological processes are enmeshed. No other phenomenon reveals action more clearly to be a fusion of intentionality and execution than habit. Here execution is not some “after-thought” or seemingly mindless implementation of an antecedent conscious process. Rather, habit reveals how motor processes are themselves thoroughly imbued with intelligence. As Ravaisson puts it, “it is not in the will but in the passive element of the movement itself that a secret activity gradually develops. To be precise, it is not action that gives birth to or strengthens the continuity or repetition of locomotion; it is a more obscure and unreflective tendency, which goes further down into the organism.”63

Precisely this last insight, however, appears to have been lost in the modern era, certainly by the time Descartes disparages habit as the very antithesis of philosophical method. That is, even as habit evacuates the principle of action from conscious volition, it does not constitute a case of mindless and aimless repetition. Conversely, and no less important, Ravaisson (as Aquinas long before him) rejects the modern premise of the body as mere unintelligent, inert “stuff” against whose imposition (or imposture) the Cartesian cogito means to assert and defend its autonomy. The modern conception of mind/body is obviously divisive, whereas the stance adopted by the Aristotelian-Thomist tradition and by exponents of modern vitalism and phenomenology (Goethe, Coleridge, John Henry Newman, Félix Ravaisson, Franz Brentano, Maurice Merleau-Ponty, Jean-Luc Marion) is fundamentally integrative. For Ravaisson, to understand habit as “inclination” is to acknowledge both its intelligential nature and its teleological orientation. Hence the body’s “sensibility increasingly demands this sensation that the will abandons.” Even as the habituated movement “leaves the sphere of will and reflection, it does not leave that of intelligence. It does not become the mechanical effect of an external impulse, but rather the effect of an inclination that follows from the will.” In fact, habituation closes the gap between will and telos in what Ravaisson shows to be a metonymic, narrative progression. It intensifies the degree to which consciousness participates in its object or, rather, in the very idea of which the object is the phenomenon: “In the progress of habit, inclination, as it takes over from the will, comes closer and closer to the actuality that it aims to realize; it increasingly adopts its form. The duration of movement gradually transforms the potentiality, the virtuality, into a tendency, and gradually the tendency is transformed into action. The interval that the understanding represents between the movement and its goal gradually diminishes; the distinction is effaced; the end whose idea gave rise to the inclination comes closer to it, touches it, becomes fused with it.” Ravaisson’s remarkable forensic account of habit culminates in the following extended passage, which is worth quoting in full:

In reflection and will, the end of movement is an idea, an ideal to be accomplished: something that should be, that can be and which is not yet. It is a possibility to be realized. But as the end becomes fused with the movement, and the movement with the tendency, possibility, the ideal, is realized in it. The idea becomes being, the very being of the movement and of the tendency that it determines. Habit becomes more and more a substantial idea. The obscure intelligence that through habit comes to replace reflection, this immediate intelligence where subject and object are confounded, is a real intuition, in which the real and the ideal, being and thought are fused together.64

Ravaisson’s main point here is to showcase that habit is a particular manifestation of intelligence, and that it consummates the teleological nature of human thought. For the classical tradition—which here would comprise both Platonic and Aristotelian, as well as Augustinian and Thomist, models—it is the essence of all thought to aspire to the most complete participation in an idea and not, as the Enlightenment typically insists, to an open-ended, critical, and prevaricating distance vis-à-vis the object of thought.

It is just this axiomatic disparagement of anything on the order of immediacy, intuition, and unreserved commitment to an idea that prevents modernity from grasping any stance (such as habit) that aims at a “fusion” of being with idea. Habit works by minimizing distance and suspending detachment, concepts that define the intellectual stance of modernity. Another way of drawing the picture might be to understand habit as an inconspicuous, almost quotidian form of mystical experience. A writer who feels words, phrases, sentences effortlessly arising and capturing the object of his attention; musicians executing a scrupulously rehearsed piece of chamber music with seeming ease; a practiced long-distance runner conscious of her near-effortless control of pace, posture, breathing, and so forth: they all experience the benefits of habituation and the way that volition and execution have merged in a single, intelligent, and purposive psycho-physiological continuum. Operating outside the modern subject’s antagonistic and distrustful outlook on embodied phenomena, habit points a way beyond the subject-object divide. Yet it does so not by investing either the self or some object “out there” with improbable or fantastic powers; rather, it effects a unio mystica by moving beyond the gratuitous and constricting premise of the self as a being legitimated solely by its principled distrust of inner and outer phenomena alike. Indeed, there is no other road toward that mystical union than the long, slow, and admittedly uncertain one through habituation. What renders habit a significant philosophical concept is that it opens up a domain of meaning that cannot be described in terms of autonomy or heteronomy, of the merely “subjective” or “objective” at all.

Notably, it is just this intelligent, teleological, and metaphysical dimension of habit that vanishes in modern social theory, a shift that brings us to the second term already prominent in Smith’s Theory and, obviously, in Watson’s eponymous disciplinary innovation: “behavior.” As it turns out, that concept shares the same etymological root with “habit,” being obliquely associated with habere and, more obviously, with its Old French derivative, aveir, avoir.65 Appearing a total of eighty-three times in the sixth edition of Smith’s Theory of Moral Sentiments, “behaviour” already surfaces in the very first chapter (“the furious behaviour of an angry man”); and throughout the book the term denotes not an inner disposition but an outwardly manifested pattern of conduct sufficiently stable to be assigned a proper name (e.g., “we call his behaviour mean-spiritedness” [TMS, 35]). Though regularly used as a synonym for “habit,” “behavior” in Smith’s Theory does not denote a dynamic, teleological progression within the self but, instead, refers to the seemingly ubiquitous phenomenon of social adaptation. As such, the term supports Smith’s overarching objective of re-describing and thus capturing the individual as an aggregate of statistical probabilities. The logic of “behavior” in Smith’s account is plainly inductive in that it serves to distill concepts from myriad cases whose likeness and meaning, however, are simply assumed rather than interpretively demonstrated. As a result, one might say that for each instance of “behavior” isolated by the colloquial “we” of Smith’s meandering account there has to exist within the observer a perceptual habitus no less rigid and reflex-like than how the book portrays the cultural conditioning of its empirical subjects. Such is the methodological price exacted by a theory that has reduced moral cognition to mimetic practice, rational action to compliant behavior. A strange hybrid of budding sociologist and virtual superego, Smith’s authorial voice (ventriloquized as the “impartial spectator”) seeks to extract faintly normative meanings from an abundance of supposedly equivalent and self-interpreting social phenomena.

A few cases will help illustrate the prominent role assigned to moral commonplaces throughout Smith’s argument, as well as that peculiarly modern, actuarial deformation of habit as impersonal and invariantly repetitive “behavior.” Of the social upstart—no doubt a frequent character in the dynamic entrepreneurial world of mid-Georgian England—Smith notes that, whatever his merits, he is “generally disagreeable.” Hence, “if he has any judgment, . . . he redoubles his attention to his old friends and endeavours more than ever to be humble, assiduous, and complaisant. And this is the behaviour which in his situation we most approve of; because we expect, it seems, that he should have more sympathy with our envy and aversion to his happiness, than we have with his happiness” (TMS, 41). By contrast, the coxcomb pretending to nobility and “affect[ing] to be eminent by the superior propriety of his ordinary behaviour” (TMS, 54–55) meets with contempt. His failure is to emulate a social and symbolic grammar outside his native domain and, in so doing, to render the theatrical project of social emulation excessively apparent. What Lord Byron would later so cannily identify as Keats’s “shabby genteel” also vexes Smith, even as the latter is far more sympathetic than Byron to middle-class “virtues” such as industry, frugality, and mainstream piety. For Smith, there still is such a thing as genuine virtue, though he insists that its authenticity is not achieved by the categorical disavowal of theatrical display but, rather, by its measured and class-specific cultivation. What Smith understands by virtue involves the subtle emulation of aesthetic, rhetorical, and social practices that cumulatively define a class-specific behavioral template. In the Theory of Moral Sentiments, the transposition of habit into behavior thus unfolds in strictly mechanical ways. Insofar as the theater of public behavior seeks to merge self and other in the virtual superego of the impartial spectator, habituation involves repetition without difference. Its sole objective is the individual’s compliance with or internalization of a process of social conditioning that no longer has any intelligible telos outside its own operation. Supplanting a premodern, teleological progression that had fused will and object in habituated action, “behavior” in Smith’s Theory aims at an isomorphism of self and other that strips either individual of the ability to claim credit and take responsibility for the meanings thus generated. Smith’s “behaviour” discovers nothing, leads nowhere, but merely reaffirms what is hypostatized or projected as a meaning supposed to be already held by the other: “The man who is conscious to himself that he has exactly observed those measures of conduct which experience informs him are generally agreeable, reflects with satisfaction on the propriety of his own behaviour” (TMS, 116).

At times, Smith comes close to perceiving some of his project’s more troubling implications, in particular, the threatening disappearance of any line demarcating moral qualia from their strictly opportunistic imitation. At what point does the emulation of behavior or “conduct” turn into sheer simulation? And how would a strictly mimetic moral agent fare in a thoroughly corrupt or totalitarian community? It is in a chapter on “the corruption of our moral sentiments, which is occasioned by this disposition to admire the rich and the great, and to despise or neglect persons of poor and mean condition” (TMS, 61–66) that we find Smith struggling with the unintended consequences of his spectatorial model. For the first time, the almost complete absence of a narrative, teleological, and counterfactual dimension from Smith’s sentimental account presents itself as a serious liability. Once again drawing on theatrical or painterly tropes, Smith offers “two different models, two different pictures . . . the one more gaudy and glittering in its colouring; the other more correct and more exquisitely beautiful in its outline.” Yet as it turns out, only “the most studious and careful observer, . . . a select, though, I am afraid, but a small party” will appreciate the inconspicuous virtues of “temperance and propriety” (TMS, 62). The problem, in other words, is that a strictly mimetic conception of moral agency is prone to be drawn to the most glaring and conspicuous patterns of conduct or (what may amount to the same thing) to a type of behavior associated with the most palpable material rewards. The specular mechanism of the “man within the breast” (TMS, 130) is not some culturally or morally neutral view from nowhere but has already inscribed within it those moral judgments and valuations to which it is said to give rise. For Smith, the only road out of this dilemma is the one that he was to chart with requisite detail in his Wealth of Nations nearly two decades later. Yet the outlines of it are already hinted at in the Theory of Moral Sentiments, specifically in Smith’s hopeful conjecture that “in the middling and inferior stations of life, the road to virtue and that to fortune, to such fortune, at least, as men in such stations can reasonably expect to acquire, are, happily in most cases, very nearly the same. In all the middling and inferior professions, real and solid professional abilities, joined to prudent, just, firm, and temperate conduct, can very seldom fail of success” (TMS, 63).

It is no accident that the word “profession” should show up twice within the same sentence, for its self-regulating view of propriety serves the same purpose in the realm of economics that “behaviour” does in the domain of morals. Clearly, Smith’s behaviorism is meant to describe and further consolidate the fluctuating and emergent demographic phenomenon of the middling classes, which lack a viable strategy for self-description and self-legitimation of the kind enjoyed by the “man of rank and distinction.” Central to his entire moral argument is a conception of society as a harmonized aggregate of essentially atomistic and selfish individuals: “Every man, therefore, is much more deeply interested in whatever immediately concerns himself, than in what concerns any other man” (TMS, 83). With revealing terminological imprecision, Smith thus shifts back and forth between the “impartial” and the “indifferent spectator” (TMS, 83, 85). Composed of so many indifferent or “punctual” selves (as Charles Taylor has more recently dubbed them), social space itself has become an extrapolated, virtual domain, rather than a historically grown and socially elaborated, empirical framework.66 Its evanescent, abstract qualities also explain why the quintessentially modern category of the “social”—so markedly different from the sharp division between the “private” (oikos) and the “public” (politeia) in Aristotle—can no longer compel individuals to support it as a positive goal but merely insists on the minimal condition of their compliance with the law. Unaware of how a radical, atomistic notion of the individual “vitiates arguments for connections,” Smith maintains that, even though “society may subsist among different men, . . . no man in it should owe it any obligation.”67

By positing individual liberty as ontologically prior and substantially unrelated to any shared norms, values, or virtues, Smith conceives of society in strictly contractual, institutional, and pragmatic terms. Its coherence rests not on a normative (Aristotelian) notion of justice and the good but is sustained only by the punitive post facto intervention of the law. As Smith concedes, without the specter of legal retribution, modern society cannot exist at all: “If it is removed, the great, the immense fabric of human society . . . must in a moment crumble into atoms” (TMS, 86). While uneasily retaining ancient Stoic prescriptions as a vague defense against the troubling prospect of a wholly disembedded subjectivity, Smith’s classical liberalism ultimately can only conceive the self in actuarial and virtual terms. Anticipating similar concerns in Goethe’s Wilhelm Meister’s Apprenticeship (1796), Smith is intent on bridging the divide that separates the economically self-created “burgher” (Bürger) from the public persona (öffentliche Person) of the gentry and aristocracy.68 For the latter is constitutively identified with the public sphere and its appraisal of conduct as a grammar of meticulously rehearsed and internalized gestures and rhetorical modes. Always on stage, as it were, the public person natively inhabits and projects social authority (“if his behaviour is not altogether absurd” [TMS, 51–54]). By contrast, members of “the middling and inferior professions” lack as yet a conceptual matrix and practical regimen by means of which they may understand, shape, and legitimate their persona in “approved” and behaviorally objective ways.

A development already palpable in post-Hobbesian thought thus culminates in the Theory of Moral Sentiments. Viz., the displacement of a metaphysical concept of the will by the empiricist (nominalist) notion of contingent and discontinuous desires entails the loss of several other concepts that had previously been just as central to the project of humanistic inquiry. The Aristotelian/Thomist concept of “action” (praxis, operatio) loses its rational dimension and, as a result, is absorbed into a Lockean calculus of countless disaggregated “reactions” or, in the case of Smith, into a mimetic logic of “transactions.” Concurrently, the concept of deliberation (phronēsis, prudentia) and judgment (prohairesis, electio) gives way to a proto-behaviorist model wherein the self constitutes itself transferentially—viz., by emulation rather than reflection. By the time that we get to Smith’s Theory of Moral Sentiments, the (neo-) Platonic idea of mind as truly active and shaping its world has become almost entirely foreign. Perceiving action, judgment, and willing as casualties of a prolonged campaign of epistemological skepticism that begins with Hobbes, mutates in Locke’s and Mandeville’s hedonistic theory, and reaches its apex in Hume’s Treatise, Smith acquiesces to work with what is left: a dramatically weakened conception of the human being, substantially devoid of internal development or progression except insofar as it merges with an actuarial and mimetic construct of successful socialization or “behaviour.” The striking loss of the founding category of “action” is rendered symptomatic in Smith’s Theory by his apparent (and decidedly un-Stoic) failure to grasp emotion as an integral part of a unified intellect, and in turn to understand the human intellect itself as teleologically ordered toward active participation in the ratio of the world. Further compounding this erosion of a differentiated conceptual apparatus for humanistic inquiry, Smith’s Theory premises its central category of “behavior” on what turns out to be a misapprehension of the idea of “habit”—viz., as the strictly passive, mechanistic, and non-rational response to external stimuli or some introjected equivalent. Such a view rests on a logical fallacy (endemic to early Enlightenment thought) whereby it is supposed that mind is not only premised on the body as its “material” condition of possibility, but that it is therefore also exclusively and exhaustively determined by the body.

Though not nearly as dogmatic as Locke, Mandeville, and Hume about reducing all final causation to material and, ultimately, to efficient causation, Smith nonetheless lacks any phenomenological curiosity (as do Watson and his behaviorist heirs) about the operation of mind. Instead, his Theory of Moral Sentiments signals moral philosophy’s gradual retreat from internalist models of human agency. It also affirms the disappearance of a fully articulated idea of human flourishing by confining human awareness to a continuous present filled with an endless stream of incidental and unreflected desires. Supplanting normative frameworks with contingent interestedness, reasoned deliberation about ends with isolated “decisions” about means, modern liberalism envisions and (if only in an impoverished sense) achieves moral neutrality as a possible, indeed indispensable way of being in the world. What, in a notably Burkean turn of phrase, Hans-Georg Gadamer some time ago had flagged as the “Enlightenment’s prejudice against prejudice” thus involves a deep-seated resistance to contemplation, sustained deliberation, and any type of “strong evaluation” (as Charles Taylor calls it). Put in positive terms, beginning with Smith’s mimetic and quasi-behaviorist concept of agency, it appears not only safe but decidedly preferable for the modern self to live out its existence unfettered by supra-personal and normative commitments. A significant implication of this “naturalist reduction” (Taylor) involves the pervasive loss of awareness regarding the deep history and hermeneutic complexity of moral concepts. While this implication remains oblique in Smith’s Theory, it is on full display a generation later, such as in Thomas Paine’s characteristically blunt assertion that “every generation is, and must be, competent to all the purposes which its occasions require. It is the living, and not the dead, that are to be accommodated.”69

Paine highlights a major implication of modernity’s reduction of the Aristotelian category of “action” (praxis) to a strictly methodological challenge: viz., how to “implement” or “comply” with abstract propositions whose authority and cogency it takes to be substantially unrelated to their material realization. Thus modern liberalism no longer understands “practice” or “doing” as a distinctive realm of contingent experience. Rather, action serves the strictly instrumental purpose of securing a non-contingent state captured by Nietzsche’s acerbic image of the herd grazing in “security, safety, contentment,” reassured that “suffering has been abolished.”70 In pursuit of its speculative utopias, liberalism dissolves action into behavior, that is, into the coordinated, bureaucratic, or habituated implementation of a concept. Social and material processes thus come to be seen as but the outward manifestation of a logical ratio of means and ends. Such Zweckrationalität (as Max Weber was to christen it) may present itself as the dialectical play of “interests” gradually civilizing the passions in the writings of the Scottish economists or as the abstract “felicific calculus” of Jeremy Bentham’s utilitarianism and the harsh social policies seemingly licensed by it.71 Yet Smith’s Theory of Moral Sentiments also reveals the extent to which liberalism’s vision of putatively rational and individual agents defined by a single, pervasive, and hence fundamentally predictable template of “behavior” ultimately conspires to dissolve the boundaries of agency and individuality alike. For when transmuted into “labor” and “behavior,” respectively, the Aristotelian categories of “work” and “action” are aggregated and tabulated by the metrics of modern statistics, social psychology, and probabilistic reasoning—conceptual and methodological innovations that helped realize the Enlightenment project of stabilizing social life in a set of “average” values, while sequestering contingency as a mere fluctuation of data.72 Yet as it is captured through disciplinary lenses of its own devising, modern individuality happens upon a model of rational and competitive social processes essentially indifferent to questions of subjective intention and objective responsibility. Modernity’s quintessentially linear and progressive (rather than cyclical) conception of time paradoxically pivots on the non-knowledge of individual agency as to the deeper systemic and ideological processes in which it is caught up. Hence the emancipatory claims of liberalism’s neutral or “punctual” self make for an odd contrast with the concurrent emergence of systemic (dialectical) models of rationality according to which all historical process is necessarily opaque to the “natural consciousness” (Hegel) so vicariously advancing it.

1. Figure of Theater, 177. For another account of Smith’s Theory of Moral Sentiments, see Griswold, Adam Smith; for a brief and wide ranging genealogy of sentimentalism, see Chandler, “Politics of Sentiment.”

2. Eagleton, Trouble, 71, 75. Eagleton’s central objection that “sympathy cannot be entirely spontaneous, since it needs to weigh the merits of its object” is the premise for Lori Branch’s recent work on the paradox of spontaneous self-other relations in Shaftesbury and Wordsworth; see Rituals, esp. 91–134 and 175–209.

3. Marshall is among the few critics to point out how Smith’s spectatorial doubling of the self reactivates Shaftesbury’s “dramatic method” in the Soliloquy and his advice there to “divide yourself, or be two” (Figure of Theater, 176).

4. Habermas, Structural Transformation, 103; for a recent account of political and economic theory, see Mitchell, Sympathy, esp. 76–93.

5. Hume, Essays, 262–263.

6. Hirschman, Passions, 17; in so mediating between reason and passion, the new concept of socio-economic interestedness “was seen to partake in effect of the better nature of each, as the passion of self-love upgraded and contained by reason, and as reason given direction and force by that passion” (ibid., 43). On the emergence of financial and industrial capitalism, see K. Polanyi, Great Transformation, 33–42 and 56–76; Pocock, Virtue, 103–124, and Machiavellian Moment, 462–505; Colley, Britons, 55–100; for basic statistical information on this shift, see Porter, English Society, 185–213.

7. Machiavellian Moment, 456–457.

8. Schneewind, Invention, 380.

9. TMS, 117; this neo-Stoic project is echoed time and again in The Wealth of Nations, as in Smith’s familiar assurance that “the desire of bettering our condition, a desire which, though generally calm and dispassionate, comes with us from the womb and never leaves us till we go into the grave” (Wealth of Nations, III.2.28; italics mine).

10. Machiavellian Moment, 456–457; Hirschman, Passions, 32.

11. Pocock, Machiavellian Moment, 457. Though far more restrained than Burke’s Reflections, Part VI of TMS (added in 1790) shows Smith’s clear preference for the fluidity and contingency that Pocock here describes as a “spirit of system” such as might seek to establish an “ideal plan of government . . . all at once, and in spite of all opposition” (TMS, 234); for a summary of the two alternative scenarios developed by post-Harringtonian, Augustan political theory, see Pocock, Machiavellian Moment, 458–460, as well as Pocock, Virtue, 103–124.

12. On the gradual contraction of the word “interest” and, especially, its French counterpart (intérêt) to a strictly economic understanding, see Hirschman, Passions, 31–48; by the time of Hume’s essay “Of Interest” (1754) that shift had been all but completed.

13. On the changing scene of artistic production in the eighteenth century, see Brewer, Pleasures, 201–324; Barrell, Theory of Painting, esp. 1–162.

14. For a detailed account of Hellenistic philosophy as a non-specialist, therapeutic enterprise, see Nussbaum, Therapy, esp. 316–438.

15. Eagleton, Trouble, 43; Mitchell, Sympathy, 82, 87.

16. “There is,” Nussbaum argues, “in Greek thought about the emotions, from Plato and Aristotle straight on through Epicurus, an agreement that the emotions are not simply blind surges of affect, stirrings or sensations that are unidentified, and distinguished from one another, by their felt quality alone.” Indeed, “it was not an item of unargued dogma for the Stoics that the soul has just one part . . . was a conclusion, and a conclusion of arguments in moral psychology” (Therapy, 369, 373). Neither Zeno nor Chrysippus recognized (as some late Stoics would argue) a separate, emotional part of the soul; see Nussbaum’s detailed account, 366–386), as well as my discussion above of “judgment” in Aristotle and the Stoics, 88–107.

17. Seneca, De Ira (“On Anger”), 2.3. Similarly, Epictetus remarks that “it is not things themselves that disturb men, but their judgments about things” (Long and Sedley, Hellenistic Philosophers, 418).

18. On Stoic accounts of judgment in relation to a disordered, weak will and to desire, see above, 99–107. Amélie Rorty links akrasia to “background dispositional patterns of perceptual, cognitive, affective, and behavioral habits” and argues that akrasia need not be “self-centered,” is not anchored “in a specific desire,” and “need not involve any manifest conscious belief” (“Social and Political Sources,” 649, 650); see also her discussion of the concept in the context of Aristotle’s ethics (“Akrasia and Pleasure”). The once common rendering of akrasia as “weakness of will” has recently been called into question; see Holton, Willing, Wanting, Waiting, 80–96; and Rorty, “Social and Political Sources.”

19. For a powerful (and witty) statement about philosophy as a remedy against suffering—rather than some sterile and inconsequential parsing of propositions—see Seneca, Moral Epistles, no. 48. Similarly, Musonius insists on the practical nature of philosophy as a regimen of mental health undertaken for the sake of action: “Whatever arguments [sophists] undertake, I say that these should be undertaken for the sake of deeds. Just as a medical argument is no use unless it brings human bodies to health, so too, if someone grasps or teaches an argument as a philosopher, that argument is no use, unless it conduces to the excellence of the human soul” (qtd. in Nussbaum, Therapy, 324).

20. On Augustine’s “rehabilitation of the affections” from their apparent proscription in Stoic thought, see Wetzel, Augustine and the Limits of Virtue, 98–111; Verbeke, “Augustin et le stoïcisme”; and Colish, Stoic Tradition, 142–238. On Stoic and Augustinian elements in Descartes’s philosophy, see Hanby, Augustine and Modernity, 137–144.

21. Though alert to the theatricality of Smith’s overall argument, David Marshall views “Smith’s endorsement of Stoic ideas . . . as the result of an antitheatrical sensibility” (Figure of Theater, 184); my argument, by contrast, is that Smith’s Stoicism is neither anti-theatrical nor properly Stoic in nature; rather, we are presented with a simulation of Stoic apatheia and a theatrically affected anti-theatricalism.

22. TMS, 24; with specific attention to the theatricality of emotion, Smith reiterates this point in Part I: “When we attend to the representation of a tragedy, we struggle against sympathetic sorrow which the entertainment inspires as long as we can, and we give way to it at last only when we can no longer avoid it: we even then endeavour to cover our concern from the company. If we shed any tears, we carefully conceal them” (TMS, 46).

23. How Novels Think, 14–15.

24. For accounts of this paradigm, see Anderson, Imagined Communities, 1–36; Taylor, Modern Social Imaginaries, 49–107.

25. Augustine’s emotional link between sympathy and narcissism ultimately points back to his underlying preoccupation with curiositas. “Curiosity” is always a vice for Augustine, and it is repeatedly linked to voyeurism by way of 1 John 2:16, a passage quoted or alluded to numerous times throughout the Confessions: “For all that is in the world is the concupiscence of the flesh and the concupiscence of the eyes and the pride of life, which is not of the Father but is of the world” (quoniam omne quod est in mundo concupiscentia carnis et concupiscentia oculorum est et superbia vitae quae non est ex Patre sed ex mundo est). On Book 3, Chapter 2.2, see James O’Donnell’s commentary on the Confessions at www.stoa.org/hippo/; on curiosity in Augustine, see Griffiths, Intellectual Appetite, 1–29. On the particular passage in Confessions, Book 3, see also the essay by Breyfogle in Paffenroth and Kennedy, Augustine’s Confessions, 35–52. Augustine later returns to the question of “compassion” (misericordia) as part of his discussion of Stoicism; “and what is compassion but a kind of fellow feeling in our hearts for the misery of another which compels us to help him if we can? This impulse is the servant of right reason when compassion is displayed in such a way as to preserve righteousness” (Quid est autem misericordia, nisi alienae miseriae quaedam in nostro corde compassio, qua utique, si possumus, subvenire compellimur? Servit autem motus iste rationi, quando ita praebetur misericordia, ut justitia conservetur). City of God, 10.5 (p. 365).

26. Marshall, Figure of Theater, 173. Hume had previously mused on the “unaccountable pleasure, which the spectators of a well-written tragedy receive from sorrow, terror, anxiety, and other passions on stage” (“Of Tragedy,” in Essays, 216). A radical Augustinian himself, Blake echoes the point in Songs of Experience: “Pity would be no more, / If we did not make somebody Poor: / And Mercy no more could be, / If all were as happy as we” (“The Human Abstract,” CPP, 27).

27. On the derivation of Smith’s sympathetic ties from early modern cosmology and natural philosophy, esp. the appeal to a sympathia naturae vita in discussions of magnetism, medicine, gravitational physics, and the chemical elements, see Vogl, Kalkül und Leidenschaft, 87.

28. Marshall, Figure of Theater, 176; Eagleton, Trouble, 48.

29. Vogl, Kalkül und Leidenschaft, 80; likewise, Marshall notes how the transferential operation of sympathy “more than doubles the theatrical positions Smith sees enacted in sympathy by compelling us to become spectators to our spectators and thereby spectators to ourselves” (Figure of Theater, 173). Similarly, for Soni, even as sympathy, which “promised to serve as a bridge between self and other, betrays its promise and leaves the self embroiled with its own emotions, which it imagines to have come from the other” (Mourning Happiness, 309). Here my account differs in that it views sympathy in Smith as garden-variety transference and projection, simply because the self has no substantive identity prior to or independent of such projection. There is only a virtual self, and its putative inner states are of an equally virtual (or behaviorist) nature, as remains to be shown.

30. Soni, Mourning Happiness, 305, 311.

31. Hume to Smith (28 July 1759), in Letters, 1:313.

32. Soni, Mourning Happiness, 300–301.

33. Letter to Smith (14 June 1759), qtd. in TMS, 26.

34. July 1759 (no. xxi); though unsigned, the review is attributed to William Rose; qtd. in TMS, 27.

35. Burke, letter and review as quoted in TMS, 28–29.

36. Journal encyclopédique (October 1760), qtd. in TMS, 29.

37. R. Mitchell, Sympathy, 89. Similarly, Eagleton notes that in Smith’s Lacanian framework, “there is no Other of the Other—no meta-language which would allow us to investigate our intersubjective meanings from a vantage-point beyond them, . . . rather as for Adam Smith there is no ground to our world beyond ‘the concurring sentiments of mankind’” (Trouble, 75–76).

38. On Goethe’s (early phenomenological) conception of seeing in his botanical writings, see Pfau, “All is Leaf.” On changing models of visual perception in the eighteenth century, see Crary, Techniques, esp. 1–24 and (on Goethe) 67–96.

39. Vogl succinctly captures numerous tensions, by remarking how Smith’s global principle of sympathy is located at a “precarious threshold”: “Sympathy encompasses involuntary responses, and yet its specificity originates in an act of consciousness; it construes as a self-relation what, in fact, becomes manifest in a relation of self to other; it produces realities whose mode of origination is indistinguishable from deception [Täuschungen]; it separates the world of functional processes from the world that legitimates these processes, and yet it postulates their interdependency; it locates sociality as an empirical object sui generis, and yet it delineates the grounds of legitimacy for that social world; it generates a field in which judgments and natural (physiological, psychological) forces are entwined; and [sympathy] purports to furnish a principle equally capable of formulating criteria for affections and actions” (Kalkül und Leidenschaft, 90; trans. mine).

40. See Jakobson, “Linguistics and Poetics,” esp. his distinction between conative, emotive, referential, and meta-lingual functions (Language in Literature, 69–73).

41. Strauss, Natural Right and History, 11.

42. On the Stoic view of the cosmos as a living being (zôion) “ensouled and rational” (Diogenes Laertius), a unified, interconnected, and living One, see White, “Stoic Natural Philosophy” in Inwood, ed., The Stoics, 124–152; Sellars (Stoicism, 81–106) views Stoic physics as something inaccessible to the categories of either reductive naturalism or theism, a thoroughly material and empiricist “cosmo-biology” that yet eschews nominalism, and a rationalist cosmogony that steers clear of Platonic forms; on the one hand, they espouse “a theory of rigid causal determinism, but on the other hand, qua religious pantheists, they hold a doctrine of divine providence” (99–100).

43. On the sociality of emotion in Keats, see Ricks, Keats and Embarrassment, 1–49; and Pfau, Romantic Moods, 309–339; on embarrassment as a sociological concept, see Goffman, “Embarrassment,” which contains some uncanny echoes of Smith’s Theory of Moral Sentiments.

44. Goffman, “Embarrassment,” 268.

45. Mitchell, Sympathy, 82–83.

46. Lacan, Seminar, I, 83. See also Eagleton, who notes that “for Smith, as for Lacan, our actions are always at some level a message directed to the Other. It is just that in Lacan’s view this dialogue can never be reduced to the imaginary reciprocities of a Smith, for whom each of us thrives under the benignant eye of a collective other” (Trouble, 75).

47. Goffman, “Embarrassment,” 270–271.

48. Vogl, Kalkül und Leidenschaft, 89; Soni also notes the attenuation of action in Smith’s Theory, even as “the unhappiness of the other is an impulsion to moral action.” For Smith limits sympathy, and the action it enjoins, strictly to what can be seen: “what cannot be seen does not enter the field of responsibilities,” with the result that “Smith, despite himself, ends up minimizing the value of action” (Mourning Happiness, 315).

49. TMS, 22; italics mine. Smith’s argument clearly furnished the conceptual template for Wordsworth’s “Preface” to Lyrical Ballads (1800), in particular, the transmutation of “the spontaneous overflow of feelings” into second-order passions: “while he describes and imitates passions, his situation is altogether slavish and mechanical, compared with the freedom and power of real and substantial action and suffering. So that it will be the wish of the Poet to bring his feelings near to those of the persons whose feelings he describes, nay, for short spaces of time perhaps, to let himself slip into an entire delusion, and even confound and identify his own feelings with theirs; modifying only the language which is thus suggested to him by a consideration that he describes for a particular purpose, that of giving pleasure. Here, then, he will apply the principle of selection which has been already insisted upon. He will depend upon this for removing what would otherwise be painful or disgusting in the passion; he will feel that there is no necessity to trick out or to elevate nature” (Wordsworth, Lyrical Ballads, 751). For a fuller discussion, see Pfau, Wordsworth’s Profession, 246–260.

50. Watson, “Psychology,” 163, 160; subsequent references are given parenthetically.

51. For a recent, comprehensive account of that tradition, see C. Taylor’s Sources; for another, albeit more guarded version of that story, see Schneewind’s Invention.

52. “Behaviorism,” in Stanford Encyclopedia of Philosophy (at www.plato.stanford.edu/entries/behaviorism/), accessed 14 December 2010.

53. Watson, “Psychology,” 158.

54. Ibid., 174n.

55. Arguably, it is the rise of cognitive-science models—which by now have metastasized across the entire disciplinary spectrum of the humanities, several social sciences, as well as law, economics, and religion—that appears to be the most obvious heir to the behaviorist enterprise. A case in point would be the improbable appeal of the cognitivist model for mental causation, such as Libet’s “readiness-potential” or Wegener’s related reductionist view of consciousness and will as an “illusion” or an epiphenomenal “feeling of doing”; see Mele, Effective Intentions, esp. 31–90; Tallis, I Am, 287–324.

56. Watson, “Psychology,” 174n.

57. Though substantially correct, the etymology is incomplete. The Latin habitus is derived from the Greek hexis (itself a derivative of the verb-form ekhein = to have). It bears noting that another Latin term for “habit,” consuetudo, operates prominently in Augustine—esp. in De Vera Religione, the Confessions, and Opus Imperfectum contra Julianum. Yet unlike the neutral habitus, Augustine’s consuetudo tends to denote a “bad habit.” For a detailed account, see Prendiville (“Idea of Habit”), who also points out significant parallels between Plotinus’s theory of purification (Enneads, 3.6.5) and Augustine’s project of overcoming the soul’s habitual fixation on embodied, sensory states (consuetudo corporum).

58. Quaedam autem sunt adventitiae, que ab extrinseco efficiuntur, et possunt amitti. The point is more fully developed in ST, Ia IIae Q 53 (“On the Corruption of Habits”).

59. Nietzsche, Beyond Good and Evil, no. 62; the phrase constitutes something of a paradox since by Nietzsche’s own account (in this and various other books) it is precisely the attribute of animals to be determined, viz., by the guidance of instincts that have long evolved in response to determinate environmental factors.

60. Quod habitus quidem perfectio est, non tamen talis perfectio que sit terminus sui subiecti (ST, Ia IIae Q 52 A 1).

61. Descartes speaks of “the habit of holding on to old opinions,” the “habit of making ill-considered judgments,” and he envisions his own philosophical regimen as a way of counteracting the influence of “my habitual opinions . . . until the weight of preconceived opinion is counter-balanced and the distorting influence of habit no longer prevents my judgment from perceiving things correctly.” Notably, though, his neo-Stoic project of intellectual self-discipline is itself envisioned as a type of habituation: “by attentive and repeated meditation I am nevertheless able to . . . get into the habit of avoiding error” (Meditations, 23, 56, 15, 43). Kant’s outlook on habit is more nuanced, in part because he seeks to move beyond Hume’s construal of “habit” and “custom” as pervasive, and epistemologically illegitimate practices justifying his own skepticism; see esp. Kant’s discussion of Hume, Critique of Pure Reason (B 788–797). Still, Kant’s own “critical” rather than “skeptical” perspective on reason begins with a critique of habit. Having restated the central tenets of his transcendental method in the Prolegomena for a Future Metaphysics, Kant admonishes “the reader who retains his long habit [lange Gewohnheit] of taking experience for a merely empirical composition of perceptions . . . to heed well this difference between experience and the mere aggregate of perceptions” (Prolegomena, §26, 115). Arguably, Kant’s most revealing discussion of habit occurs in his late Religion within the Boundaries of Mere Reason, in a “General Remark” appended to Part I (Religion, 89–97).

62. Ravaisson, Of Habit, 29, 25, 37, 39.

63. Ibid., 43, 51–53.

64. Ibid., 53–55.

65. The roots of the English “behaviour” partially converge with the etymology of “habit” (Lat. habitus, from habere = to have). Thus the OED constructs the provenance of behavior as follows: “by form-analogy with HAVOUR, hauyoure, common 15th-16th c. form of the word which was originally Aver s.b. (q.v.), aveyr, also in 15th c. avoir; really in OF. aveir, avoir, in sense of ‘having, possession,’ but naturally affiliated in English to the verb have, and spelt haver, havour, haviour, etc. Hence, by analogy, have: havour, -iour: behave: behavour, -iour.” Smith is not mentioned, though the subsequent examples of historical usage tend to be clustered around eighteenth-century writers.

66. On the shift from historically defined “place” to abstract, bureaucratic and legal “space,” see Giddens, Consequences, 1–20. On formal-aesthetic echoes of that shift, see my discussion of early Romanticism’s transmutation of the ballad form into pastoral elegy in Wordsworth’s “Michael,” in Romantic Moods, 191–225.

67. TMS, 86; Ferguson, Solitude, 9. What Ferguson identifies as a peculiar feature of Burke’s Enquiry also haunts Smith’s Theory; in both cases, the drift of their arguments “makes the impossibility of sustaining claims for a unified and unitary self seem, paradoxically, to emerge precisely out of their basis in individual experience” (ibid., 9).

68. See Pfau, “Bildungsspiele.”

69. Gadamer, Truth and Method, 273; C. Taylor, Sources, 25; Paine, Rights of Man, 42.

70. Beyond Good and Evil, 41.

71. On Weber, see Milbank, Theology, 84–100; for Milbank, history can only be captured in the metastases of means/end rationality: “fully objective history (sociology) is primarily about economic rationality, formal bureaucracy, and Machiavellian politics. What lies outside these categories cannot be read as a certain distinctive pattern of symbolic action, but only negatively registered.” As such, Weber’s notion of Zweckrationalität is “teasingly ambiguous.” “At times . . . [it] is a mere matter of methodological convenience, at other times it is the dark business of our Western fate” (ibid., 87).

72. On Aristotle’s implicit idea of freedom as “the presupposition of the exercise of the virtues and the achievement of the good,” see MacIntyre, AV, 146–164 (quote from 159); on Aristotle’s concept of “action,” see especially Arendt, HC, 175–247. Echoing Arendt, Milbank notes how “praxis, in the old Aristotelian sense, referred to a dimension of action which was categorically ‘ethical’ because it could not be separated from a person’s essential being or character (ethos); it meant a doing which was also a being. It also implied action directed towards a particular end (telos), but an end immanent within the very means used to achieve it, the practice of ‘virtue’” (Theology, 161).