Gross - Comment On Searle

Anthropological Theory Copyright © 2006 SAGE Publications (London, Thousand Oaks, CA and New Delhi) http://ant.sagepub.com Vol 6(1): 45–56 10.1177/1463499606061734

Comment on Searle Neil Gross Harvard University, USA

Abstract While appreciative of Searle’s contributions, this commentary questions whether the core insights of The Construction of Social Reality advance all that much beyond the position staked out more than a century ago by Emile Durkheim. Durkheim, to be sure, was no analytic philosopher, but both Durkheim and Searle recognize the existence of something like collective representations, view a key feature of such representations as being that of imposing new statuses, powers, and meanings on objects, and accept that social institutions are composed of individual actors who act on the basis of this collective imposition of status functions. For this reason, some of the criticisms that may be levelled against Durkheim’s approach to social ontology – for example, that it ignores situations where collective representations are intrinsically coercive in nature – may also apply to Searle. Key Words collective representations • Durkheim • institutions • narrative • power • social facts

For most of its history, sociology has maintained an ambiguous relationship with philosophy. Founded on the insight that human life is shaped, patterned, and constrained by virtue of the fact that human beings live their lives in groups of widely varying scale, sociology has often tried to distance itself from philosophical inquiry, which it has conceived to be, with some obvious exceptions, locked into an essentially pre-sociological and atomistic worldview that fails to attend to sociality and to build such an attention into its consideration of, among other matters, ethics, political theory, or epistemology. At the same time that sociology has made these moves, however, it has actively incorporated into its theorization and even empirical research insights associated with various philosophical schools, including American pragmatism and phenomenology (Gross, forthcoming). Only rarely, though, have sociologists drawn on recent Anglo-American analytic philosophy. And few analytic philosophers, for their part, have shown any interest in engaging the ideas of sociologists, with the exception of philosophers of social science, a relatively marginalized bunch in American academic philosophy, and those who have worked to put out the epistemological fires lit by sociologists of scientific knowledge (see Ashman and Baringer, 2001; Zammito, 2004). 45

ANTHROPOLOGICAL THEORY 6(1)

Under these circumstances, to have a philosopher of John Searle’s stature turn to a question that would seem to lie directly at the intersection of the two fields – namely, what is the ontological basis for the existence of institutional facts? – is a rather remarkable development, one that could, it might be hoped, prompt further dialogue between analytic philosophy and sociology. My fear, however, is that when other sociologists read The Construction of Social Reality (Searle, 1995), they may have the same dialoguehalting reaction that I did: the sense that the core claims of the book do not advance all that much beyond those which Emile Durkheim staked out a century ago. To be sure, there is some benefit of clarity to be gained by re-expressing what are essentially Durkheimian ideas in an analytic idiom. My concerns about the book relate not to some alleged failure on Searle’s part to acknowledge his Durkheimian roots, but rather to the fact that Searle’s unacknowledged and unreconstructed Durkheimianism does not leave him well positioned to inform a number of important strands of sociological theory today or to build on all the work done in Durheim’s wake. Differences between Durkheim and Searle there surely are; my claim is that the similarities are more important than the differences. I

As I understand it, Searle’s basic concern is this: We now know that the basic ‘building blocks’ of matter consist of particles suspended ‘in fields of force . . . and . . . organized into systems’ (Searle, 1995: 6). All of material reality is composed of such particles, and human beings, as living organisms, are no exception. It is also the case, however, that certain features of human life, namely social institutions, have a kind of objectivity to them despite the fact that they are in no obvious way composed of particles. What accounts for this objectivity, and, more generally, what is the ontological foundation for the existence of social facts? To answer these questions, Searle offers six basic arguments, which I am simplifying here and also reordering. First, he points out that a distinguishing feature of the human species is its capacity for language use, which centers, in his view, on the ability to represent to oneself by means of a symbol some object or process in the world. Such symbols typically acquire their meanings as part of larger linguistic systems. Second, humans have the capacity to form collective intentions, which involves ‘doing (wanting, believing and so on) something together’ (1995: 24–5). Collective intentions, according to Searle, are not mere aggregations of individual intentions, but rather revolve around situations where actors understand themselves to be part of a ‘we’. Third, through the use of language and collective intentions, humans can assign functions to objects that those objects do not have simply because of their intrinsic physical characteristics. For example, we can collectively assign to diamonds the function of symbolizing for women who wear them on their fingers the state of being married, even though there is nothing intrinsic in the atomic composition of diamonds that would suggest this usage. Fourth, the assignment of function takes the form of a ‘constitutive rule’ according to which ‘X counts as Y in context C’ (1995: 55), as in a diamond counts as a symbol of being married when it is worn in a ring on one’s finger. Fifth, all social institutions are composed of constitutive rules that are iterated upward through the ‘impos[ition] of status functions on entities that have already had status functions imposed upon them’ (1995: 80). Typically, ‘the creation of a status function’ in this iterated fashion ‘is a 46

GROSS Comment on Searle

matter of conferring some new power’ (1995: 95, emphasis in original) on people, objects, or events. Social institutions, thus understood, are able to maintain themselves as long as ‘the individuals involved and a sufficient number of members of the relevant community . . . continue to recognize and accept [their] existence’ (1995: 117). This is not to say, however, that institutions exist only insofar as people consciously know and follow all of the constitutive rules that compose them. Searle’s sixth and final major argument – which in his view links him up in certain respects with theorists of practice like Wittgenstein or Bourdieu – is that human beings who interact with institutions know their constitutive rules only tacitly, by drawing on that aspect of human cognitive ability he calls the ‘Background’. By virtue of the operation of the Background, ‘the person who behaves in a skillful way within an institution behaves as if he were following the rules, but not because he is following the rules unconsciously nor because his behavior is caused by an undifferentiated mechanism that happens to look as if it were rule structured, but rather because the mechanism has evolved precisely so that it will be sensitive to the rules ’ (1995: 146, emphasis in original). To say that a social institution has an objective existence is thus to draw an unbroken line between it and the undeniably objective fact that human beings are biological entities with certain innate capacities. II

Although he was writing in a completely different intellectual and institutional context, Durkheim’s core concerns were not so far from Searle’s own. Where Searle’s interest is in the question of how ‘phenomena like consciousness or social institutions that are not in any obvious way physical or chemical’ (1995: xi) can exist, Durkheim’s was in establishing that social facts represent a sui generis ontological realm built upon but in no sense reducible to the psychology of individuals. It is commonplace to observe two phases in his efforts along these lines: an early phase, associated above all with The Division of Labor (Durkheim, 1964a [1893]), in which social facts are described as having a quasi-material existence in the form of a conscience collective which, though external to the individual, nevertheless exerts a powerful constraint upon her while being structured primarily by what Durkheim called the ‘volume’ and ‘density’ of social relations; and a later phase, signaled by The Elementary Forms (1995 [1912]), in which interest shifts to understanding a much broader spectrum of collective representations and doing so by taking account of a larger number of explanatory variables. Although I would not deny that Durkheim’s thought underwent change over time, I am doubtful of the claim that there is a radical discontinuity between these two periods, or that Durkheim’s assertion of the reality of social facts rested on some dubious metaphysics. When the early Durkheim asserted the reality of the conscience collective, or when Durkheim of the middle period insisted that social facts are things, he was not positing the existence of some kind of collective mind, as his critics charged, or giving in to some other illegitimate ontologizing tendency, but simply pointing out that members of social groups often hold certain sentiments and beliefs in common and act toward one another on the basis of those sentiments and beliefs. To the extent that large numbers of them do so, these beliefs may form the basis for the existence of such social institutions as morality, law, the family, and the state, which, built up out of the actions of myriad individuals, then become objective features of the environment that everyone in the society 47

ANTHROPOLOGICAL THEORY 6(1)

must navigate, and may function according to an emergent logic that cannot be reduced to that of individual psychology. The key notion here is that of collective representations, which, admittedly, Durkheim did not fully develop until after he published The Rules (1964b [1895]). Although by no means a Kantian, either in his views about the bases for human knowledge or with respect to his theory of morality, Durkheim was deeply influenced by Kant’s critique of empiricism as refracted through the interpretive lenses of figures on the local French philosophical scene (see Jones, 2001; Schmaus, 2004). Following the lead set by these scholars, Durkheim held that facts cannot be made sense of unless interpreted in light of the various categories of understanding like time, space, and causality, and also through a repertoire of less fundamental concepts. Together, categories and other concepts make possible representations of reality, and it is only through such representations that reality can be known. Durkheim would eventually come to the conclusion that social experience determines the content of collective representations of the categories themselves (Schmaus, 2004), thus disputing the notion that such representations derive from the manifestation of the categories in individual psychology. But certainly it is the case that by the time he wrote The Rules, he was already of the view that in addition to the categories, concepts are necessary for representation, and that many of the concepts people hold are not inductively derived from facts, but rather learned through social interaction (the point about representation Durkheim had made already in his 1883–4 lectures, and I take him to be commenting on the social origins of linguistic concepts when he says, in The Rules, that among social facts are ‘the system of signs that I employ to express my thoughts’ [1964b: 51]). The concepts held in common by members of a social group help to give rise to distinctive patterns of representation among its members, not least representations of the group itself and of its major social institutions (see the discussion in Pickering, 2000). Although manifesting themselves in each individual, these representations are collective because they are ‘a product of shared existence, of actions and reactions called into play between the consciousness of individuals’ (Durkheim, 1964b [1895]: 56). Collective representations involve beliefs, sentiments, and practices that are shared by those in a social group, function to establish the group’s boundaries, identity, and sense of solidarity, and typically outlast any particular individual. Durkheim’s claim was that it is collective representations that form the basis for the existence of institutions. This approach to social ontology did not change appreciably in The Elementary Forms, even while Durkheim’s substantive interest there shifted to a more subtle conceptualization of cultural variation and its social-organizational and processual correlates, and away from the notion of constraint. Like Searle, Durkheim held that a key feature of collective representations is that they establish purposes for objects that those objects would not otherwise have. Consider his analysis of Australian aboriginal religion (Durkheim, 1995 [1912]). On Durkheim’s understanding, religions are composed of rites and beliefs. Religious rites always address objects, and what makes those rites religious is that they take place within a system of belief that establishes an absolute divide between objects that are sacred and those that are profane, alongside a cultural logic that offers up rules for their interrelations. ‘When a certain number of sacred things have [such] relations of coordination and subordination with one another’, Durkheim writes, ‘so as to form a system that has a certain coherence and does not belong to any other system of the 48

GROSS Comment on Searle

same sort, then the beliefs and rites, taken together, constitute a religion’ (1995 [1912]: 38). But as soon as Durkheim offers this definition, he qualifies it on the grounds that, stated thus, it fails to distinguish between religion and magic. In order to establish this distinction, it is necessary to make the further point that religious rites and beliefs – but not those pertaining to magic – compose collective representations: ‘Religious beliefs proper are always shared by a definite group that professes them and that practices the corresponding rites. Not only are they individually accepted by all members of that group, but they also belong to the group and unify it. The individuals who comprise the group feel joined to one another by the fact of common faith’ (1995 [1912]: 41). This is to say, in Searle’s language, that religious beliefs involve collective intentions. And these collective intentions work by investing objects and people with functions and powers they would not otherwise have. Durkheim’s most memorable example of this from The Elementary Forms concerned churingas, ‘pieces of wood or bits of polished stone’ that are ‘held by the string from which they are suspended’ and ‘rapidly whirled in the air so as to produce’ a ‘humming’ that ‘has ritual meaning and accompanies all religious ceremonies of any importance’ (1995 [1912]: 118). Here Durkheim points out the obvious: that ‘in themselves, the churingas are merely objects of wood or stone like so many others’ (1995 [1912]: 121). What makes them sacred is the system of collective representations that invests them with meaning. Because they are stamped with ‘the totemic emblem’ (1995 [1912]: 125), churingas come to stand symbolically for the group itself. Durkheim did not use the notion of iteration, but, like Searle, held that social institutions such as religions are composed of large numbers of beliefs and practices that involve the cultural construction of objects as having meaning of a certain sort, and that build upon a sedimented bed of other similar constructions. To be sure, there are important differences between Durkheim and Searle. One obvious one is that Searle’s understanding of language is far more sophisticated than Durkheim’s own. I do not believe it is correct to say, as Searle does in his essay for this issue, that Durkheim, alongside Weber, Schutz, and Simmel, ‘all presuppose the existence of language and then, given language, ask about the nature of society’ (p. 14). What Searle appears to mean by this is that they, along with contemporary theorists, fail to see the essentially constitutive role of language. Language does not function just to categorize and thus give us power, à la Bourdieu, and it does not function just, or even primarily, to enable us to reach rational agreement, à la Habermas. It has much more basic and fundamental functions . . . (pp. 14–15, emphasis in original) Durkheim and Schutz (to say nothing of Bourdieu) would have completely agreed with this point. All paid explicit attention to the role played by language in helping to constitute social life. Apropos of social contract theorists like Rousseau, who Searle says fell into the same trap of not taking language’s constitutive role seriously, Searle asserts that ‘if you share a common language and are already involved in conversations in that common language, you already have a social contract’ (p. 14), apparently unaware that Durkheim often pointed to language as the premier example of a social fact, and was critical of Rousseau for failing to recognize that individuality, much less social institutions like the state, could never have come into existence were humans not born into a pre-existing system of social relationships that, among other things, provided them 49

ANTHROPOLOGICAL THEORY 6(1)

with the linguistic and conceptual materials necessary to form representations of such things as individuality and the state. But certainly it cannot be denied that Durkheim’s theory of language was underdeveloped. In his earliest lecture course, for example, given in 1883–4, Durkheim expressed the highly conventional view that language is ‘a system of signs’ by means of which people ‘communicate with one another’ (Durkheim, 2004 [1883–4]: 221). Durkheim argued that while in principle ‘thought could exist in the absence of a system of signs’, it ‘would be distorted and impoverished and would require tremendous mental labor’ (2004: 225). Thus, in Durkheim’s view, not only social institutions but thought itself is rooted in language. But such an understanding – one that Durkheim never really moved beyond in his subsequent writing – is no match for Searle’s much better developed philosophy of language, centered on the theory of speech acts, which he invokes in The Construction of Social Reality in making the point that status functions are often imposed on objects by means of performative utterances in which an actor who is socially authorized to make a declaration concerning an object’s status function – for example, authorized to say ‘I now pronounce you husband and wife’, and to have such a saying count as a doing – does so. Second, while Searle is interested in accounting for the objectivity of social institutions, he is much concerned to preserve the distinction between ‘brute’ and ‘institutional’ facts; the former ‘exist independently of any human institutions’, where the latter ‘can exist only within human institutions’ (1995: 27). Durkheim was not as careful as this. Despite his quite reasonable view that social institutions and social facts more generally are built up out of the actions of individuals informed by collective representations, his desire to legitimate sociology in an institutional climate where science meant above all the physical and natural sciences led him sometimes to elide the distinction, declaring simply that social facts are as ‘real’ as any others. Insofar as Searle helps us keep this distinction clear – and leaving aside the empirical point that social relations, processes, and institutions actually do play a role in how scientists come to understand the facticity even of brute facts (Shapin, 1995) – his book represents an advance beyond Durkheim. Finally, Durkheim’s approach to social institutions never paid much explicit attention to anything like the Background. I make this point even though I am convinced by Charles Camic’s (1986) claim that many of the classical sociological theorists were deeply interested in the significance of habit for social life, a significance that later commentators like Talcott Parsons (1937) would downplay, not only out of an interest in showing that action could be subsumed under the framework of the ‘unit-act’, which presumed a motivational basis for all action and its regulation by rule-like norms and valuations, but also because the notion of habit was associated with behaviorist psychology, from which discipline builders in American sociology sought to distance themselves. Insofar as Durkheim was one of those who took habit seriously, he can be read – contrary to Parsons’ depiction of him – as one of the initiators of the turn toward practice and the interest in tacit knowledge that characterizes the theory field today (see Schatzki et al., 2001). This is an ironic reading given that his later work helped to inaugurate precisely the kind of structural anthropology against which theorists of practice like Bourdieu (1977, 1990) reacted. Still, this set of themes remained implicit in Durkheim, whereas Searle brings them into the light of day. Notwithstanding these differences, however, the similarities between Searle and Durkheim are striking: both center their accounts on the notion of representation, both 50

GROSS Comment on Searle

recognize the existence of something like collective representations, both view a key feature of such representations as being that of imposing new statuses, powers, and meanings on objects, and both accept that social institutions are composed of individual actors who act on the basis of this collective imposition of status functions. But the question remains: Why is it a critique of Searle to point out his similarities to Durkheim, at least with respect to Durkheim’s approach to social ontology? This may seem all the more puzzling in light of the recent revival of interest in Durkheim, which has involved not simply a new wave of historical and interpretive scholarship (e.g. Jones, 1999; Schmaus, 2004; Strenski, 1997), but also efforts to fold Durkheimian ideas back into contemporary sociological theory (e.g. Alexander, 1988; Emirbayer, 1996; Joas, 1996). If Searle is a Durkheimian at heart, wouldn’t this indicate his utility for sociology? I will give two reasons why I think Searle’s approach has less utility than might at first be imagined. III

The first reason concerns the relationship between culture and power. It is axiomatic for sociologists of culture today, building up from a long line of theorization that traces its roots back to Marx and Weber, that dominant social groups typically rise to power and maintain their dominance not simply through the use of force, but also through the deployment of cultural resources that establish their legitimacy (e.g. Lamont and Fournier, 1992). This, it is argued, is particularly the case in modern societies, where citizens have grown less willing over the last half century to tolerate the internal use of violence against those who make claims for a greater share of societal power and resources. This tendency, foreseen by both Norbert Elias and Michel Foucault, makes it harder for elites to employ force on their own behalf, at least domestically. At the same time, consistent with Parsonsian projections, late modernity has witnessed a widespread embrace of the principle of meritocracy as the only basis for legitimate domination, resulting in a situation where dominant classes and status-groups, and those who wish to rise to power, must rely on culture as never before to justify their positions and aspirations. The processes by which they do so, however, have been rendered much more complex than in previous eras by the increasing autonomization of fields of cultural production, shifting technologies of cultural transmission, the growing permeability of the boundaries between high and popular culture, and heightened reflexivity with respect to representations on the part of a wide variety of social actors. The challenge for sociologists of culture has been to develop analytic frameworks capable of comprehending this fluid situation – focusing perhaps especially on the question of how, in light of this complexity, dominant groups still manage to use major social institutions to inculcate beliefs that legitimate their domination – without reducing all of culture to its instrumental use or otherwise flattening out important variation in cultural dynamics. Searle acknowledges and in fact makes central to his argument the Weberian insight that social order typically cannot be maintained by force alone. But where Weber’s concept of legitimate domination, much like Marxian notions of ideology, recognized that when subordinate groups accept the ideas that establish the legitimacy of their subordination they are still being dominated, Searle’s resort to the notion of collective intentionality seems to recognize no such thing. Collective intentionality, according to Searle, is a matter of ‘genuine cooperative behavior’ (1995: 23). To be sure, it does not 51

ANTHROPOLOGICAL THEORY 6(1)

rest on the assumption that every individual in society will agree as to the value of the constitutive rule in question, but rather on acceptance of the idea that the rule represents the way that ‘we’ as a society do things. For this reason, collective intentionality is not supposed to characterize only situations of consensus, and Searle gives as an example of such intentionality situations of social conflict, such as prize-fights, in which, despite contestation and even violence, social actors still agree in principle to follow certain conventions and collective definitions of the situation. The problem with this notion, from the standpoint of an awareness of the dynamics of legitimate domination, is that there are many instances when actors collectively agree to constitutive rules without realizing that their agreement has the effect of helping to subordinate them. Take the case of the English language as spoken in the USA. A variety of actors may agree that the conventions and rules that together compose ‘standard’ English are the conventions that we as Americans are going to follow, and this may also be the case for some who do not personally follow those conventions all the time. It seems clear, however, that some actors who agree to this may thereby be helping to perpetuate their own domination or the domination of others insofar as the practice of speaking standard English is unequally distributed across race and class, and serves as a cultural marker by which opportunities are afforded to certain people and denied to others. I wonder whether the notion of intentionality, understood in the specific way it is by Searle, is really appropriate for describing the assent that occurs in such situations. As I understand it, collective intentionality implies consent. Searle seems to say as much when, in a later article on political power (Searle, 2003), he draws a distinction between status functions that are accepted by people only because of the threat of violence, and those that are accepted as valid without any such direct threat. But there would seem to be many instances where individuals agree to accept certain constitutive rules as valid only because processes of obfuscation and misdirection and even habit have prevented them from recognizing that their lives might be much better off if a different set of institutions were in place. There may be consent in such instances, but it is very thin. Searle’s favorite example of a social institution based on collective intentionality is money – this, he says, involves the collective intention to impose the constitutive rule that ‘such and such bits of paper count as money’ (1995: 44). But one does not have to be a Marxist to worry about the possibility that many people who tacitly assent to this rule – peasants who might prefer more traditional modes of exchange, or workers who, if given the choice, might prefer a system that lends itself less well to commodification – are, when they so assent, having power exercised over them and being unwittingly dominated. From this point of view, which I associate with Ralf Dahrendorf (1968), the problem with an essentially Durkheimian social ontology is not the old saw that Durkheim’s theory of society fails to take adequate account of social conflict as an engine of historical change, but rather that the notion of collective representations assumes beliefs held in common by virtue of at least an implicit desire to consent to common norms, where in fact this consent is often coerced. Contemporary sociologists, sensitive to questions of power, are, for this reason, likely to be suspicious of Searle’s formulation. This is certainly not to say that social institutions are lacking in reality, or that there do not exist genuine situations of collective intentionality – only that the ultimate foundation for much of social reality, if one need be identified, likely rests not on collective intentionality alone, but as well 52

GROSS Comment on Searle

on something like the capacity of human beings to grow accustomed – if only for limited stretches of time – to domination by powerful others. IV

The second reason why I think Searle is not in a great position to engage in dialogue with what I see as some of the most promising developments in contemporary sociology turns on the three-way relationship between social institutions, culture, and constitutive rules. Searle argues that institutions exist only insofar as individuals interact with one another on the basis of common beliefs that involve the imposition of status functions on objects that would not otherwise have them. Likewise, a key claim among those sociological theorists who have written in recent years on the nature of institutions and social structures is that, far from being the obdurate stuff that has primacy relative to culture in shaping social action, institutions and social structures are actually constituted out of culture, at the same time that they react back upon it. This is a central argument of the ‘new institutionalism’ in the study of organizations (Powell and DiMaggio, 1991), and of William Sewell’s (1992) attempt to develop a theory of the ‘duality of structure’ by synthesizing and critically reworking the contributions of Bourdieu and Anthony Giddens. That Searle’s approach rests on the cognate assumption that ‘institutional reality is constructed through linguistic representation’ has been noted by Roger Friedland and John Mohr (2004: 7) in their recent edited volume on the ‘cultural turn’ in American sociology. But if it is correct that institutions and social structures are constituted out of culture, the question becomes how to analyze the cultural material on the basis of which this occurs, and that raises the further question of the ultimate nature of that material. It is at this point that differences between Searle and contemporary theory begin to arise, for one of the insights of a number of important approaches is that these cultural materials include forms that have heretofore been the exclusive purview of those working in the humanities. When Paul Ricoeur (1991) declared that social action and culture more generally could be analyzed as if they were texts, giving expression to a sensibility that, as Jeffrey Alexander (2003) points out, was already present in the work of thinkers like Kenneth Burke or Clifford Geertz or Hayden White, he may have inadvertently given succor to Baudrillardian flights of fancy, but he also opened up the possibility that sociological analysts of culture, whether interested in institutions or not, might begin to study the significance for social life of such things as narrative, genre, metaphor, and trope, building a bridge between the sociology of culture and literary theory. To date, exponents of the new institutionalism in the study of organizations have not been at the forefront of this bridge-building exercise. For them, culture at the individual level means primarily cognitive scripts: ‘Not norms and values but taken-for-granted scripts, rules, and classifications are the stuff of which institutions are made’ (Powell and DiMaggio, 1991: 15). Such scripts, as DiMaggio (1997) has pointed out, can be understood as involving procedural rules that actors mobilize to help them make sense of situations by thinking about things in particular ways. Those social analysts who believe that culture is ultimately a matter of such scripts will be able to appreciate Searle’s book, for the scripts they have in mind when studying institutions do seem to involve constitutive rules in Searle’s sense of the term. But those of us who think that institutions are also composed of individuals drawing on metaphors, narratives, tropes, and genres might wonder 53

ANTHROPOLOGICAL THEORY 6(1)

whether Searle’s attempt to reduce institution-composing beliefs down to constitutive rules results in a loss of analytic power. For example, Margaret Somers has convincingly argued that the institution of the modern nation-state is closely bound up with a ‘political culture structure’ in which a central role is played by the ‘metanarrative of AngloAmerican citizenship theory’ (Somers, 1995: 235, 232), a story that holds that, beginning in the 17th century, people joined together in a realm of civil society outside the arena of formal politics in order to resist the power of the administrative state. It is not incidental to Somers’ account that this story is a narrative: As a narrative, Anglo-American citizenship theory is organized as a structure of relationships and assumptions that connect and configure over time and space; thus, the narrative takes on the mantle of historicity and assumes the form of a story with a valid causal and explanatory plot. Moreover, as a cultural structure, the narrative is constituted by dynamics and presuppositions that can be understood largely in terms of earlier temporal and spatial conditions, which give it the status of a historical as well as a cultural object. In this, it takes on the character of a public narrative. It provides explanations and accounts for private and public happenings in terms of a normative symbolic schema that explains the past in terms of the present, and prescribes actions that will dictate the future. (Somers, 1995: 237) I take Charles Taylor’s (2004) recent interest in various ‘modern social imaginaries’ to be expressive of a concern with a similar set of conceptual issues, though with different content. It is not clear to me that narratives of the kind that interest Somers and Taylor can be reduced down to a series of constitutive rules taking the form ‘X counts as Y in context C’, just as it is not clear how the metaphors, genres, and tropes that also play a significant role in our understanding of institutions, and hence in constituting those institutions as real, can be reduced down to rules of this sort. These and other cultural forms would seem to be sui generis in nature; not a matter of Xs counting as Ys, but of characters proceeding along in plot lines, certain features of objects being picked out by juxtaposition with other objects, particular conventions for storytelling having a deep resonance for cultural producers and consumers, and so on. If Friedland and Mohr are right that US sociology today is taking a cultural turn, and if many of those making this turn take such hermeneutic considerations seriously, there is reason to doubt whether Searle’s understanding of the nature of constitutive rules will seem adequate to many sociologists. They might readily agree that social institutions and structures are constituted partly out of constitutive rules. But they would insist that institutions and structures are constituted as well out of a variety of other cultural forms, whose existence cannot be easily made sense of with Searle’s approach. In this regard, it is once again Searle’s immersion in a Durkheimian framework that may have stood in the way of his constructing a theory with thicker links to contemporary sociology. For though the ‘strong program’ in cultural sociology, as conceived by Alexander (2003), is committed to the Durkheimian insight that cultural systems are often structured in terms of binary oppositions between sacred and profane elements, Durkheim himself was concerned primarily with the rules by which objects are given one or the other of these statuses, and showed little sensitivity to other hermeneutic matters. 54

GROSS Comment on Searle

V

In offering these criticisms, it is not my intention to suggest that sociologists will not profit from reading The Construction of Social Reality. They will, and many will find it stimulating and insightful. But I do not think the book will serve as a jumping off point for a sustained dialogue between sociology and analytic philosophy. Although its formulations may appeal to some sociologists, others will doubt that firm philosophical foundations for the discipline can be established on the basis of Searle’s particular understanding of the social. Searle may dispute this claim that he owes anything to Durkheim, but I suspect that most sociologists will appreciate the similarities. At the same time, they will recognize that Durkheim’s approach to social ontology is only the starting point for more than a century of rich sociological theorization of the nature of social reality, which Searle almost entirely ignores. Acknowledgements

For helpful comments on an earlier draft of this article I thank Roy D’Andrade, Warren Schmaus and Stephen Turner. References

Alexander, Jeffrey (2003) The Meanings of Social Life: A Cultural Sociology. Oxford: Oxford University Press. Alexander, Jeffrey, ed. (1988) Durkheimian Sociology: Cultural Studies. Cambridge: Cambridge University Press. Ashman, Keith M. and Philip S. Baringer, eds (2001) After the Science Wars. London: Routledge. Bourdieu, Pierre (1977) Outline of a Theory of Practice. Stanford, CA: Stanford University Press. Bourdieu, Pierre (1990) The Logic of Practice. Cambridge: Polity. Camic, Charles (1986) ‘The Matter of Habit’, American Journal of Sociology 91: 1039–87. Dahrendorf, Ralf (1968) Essays in the Theory of Society. Stanford, CA: Stanford University Press. DiMaggio, Paul (1997) ‘Culture and Cognition’, Annual Review of Sociology 23: 263–88. Durkheim, Emile (1964a [1893]) The Division of Labor in Society. New York: Free Press. Durkheim, Emile (1964b [1895]) The Rules of Sociological Method. New York: Free Press. Durkheim, Emile (1995 [1912]) The Elementary Forms of Religious Life. New York: Free Press. Durkheim, Emile (2004 [1883–4]) Durkheim’s Philosophy Lectures: Notes from the Lycée de Sens Course, 1883–4, edited by Neil Gross and Robert Alun Jones. Cambridge: Cambridge University Press. Emirbayer, Mustafa (1996) ‘Useful Durkheim’, Sociological Theory 14: 109–30. Friedland, Roger and John Mohr (2004) ‘The Cultural Turn in American Sociology’, in Roger Friedland and John Mohr (eds) Matters of Culture: Cultural Sociology in Practice, pp. 1–68. Cambridge: Cambridge University Press. Gross, Neil (forthcoming) ‘Pragmatism, Phenomenology, and Twentieth-Century American Sociology’, in Craig Calhoun (ed.) Sociology in America: The American Sociological Association Centennial History. Chicago, IL: University of Chicago Press. Joas, Hans (1996) The Creativity of Action. Chicago, IL: University of Chicago Press. 55

ANTHROPOLOGICAL THEORY 6(1)

Jones, Robert Alun (1999) The Development of Durkheim’s Social Realism. Cambridge: Cambridge University Press. Jones, Susan Stedman (2001) Durkheim Reconsidered. Cambridge: Blackwell. Lamont, Michèle and Marcel Fournier, eds (1992) Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago, IL: University of Chicago Press. Parsons, Talcott (1937) The Structure of Social Action. New York: McGraw-Hill. Pickering, W.S.F., ed. (2000) Durkheim and Representations. London: Routledge. Powell, Walter W. and Paul DiMaggio, eds (1991) The New Institutionalism in Organizational Analysis. Chicago, IL: University of Chicago Press. Ricoeur, Paul (1991) From Text to Action: Essays in Hermeneutics, II. Chicago, IL: Northwestern University Press. Schatzki, Theodore R., K. Knorr-Cetina and Eike von Savigny, eds (2001) The Practice Turn in Contemporary Theory. London: Routledge. Schmaus, Warren (2004) Rethinking Durkheim and his Tradition. Cambridge: Cambridge University Press. Searle, John R. (2003) ‘Social Ontology and Political Power’, in Frederick F. Schmitt (ed.) Socializing Metaphysics: The Nature of Social Reality, pp. 195–210. Lanham, MD: Rowman & Littlefield. Searle, John R. (1995) The Construction of Social Reality. New York: Free Press. Sewell, William (1992) ‘A Theory of Structure: Duality, Agency, and Transformation’, American Journal of Sociology 98: 1–29. Shapin, Steven (1995) ‘Here and Everywhere: Sociology of Scientific Knowledge’, Annual Review of Sociology 21: 289–321. Somers, Margaret (1995) ‘Narrating and Naturalizing Civil Society and Citizenship Theory: The Place of Political Culture and the Public Sphere’, Sociological Theory 13: 229–74. Strenski, Ivan (1997) Durkheim and the Jews of France. Chicago, IL: University of Chicago Press. Taylor, Charles (2004) Modern Social Imaginaries. Durham, NC: Duke University Press. Zammito, John H. (2004) A Nice Derangement of Epistemes: Post-Positivism in the Study of Science from Quine to Latour. Chicago, IL: University of Chicago Press. NEIL GROSS is Assistant Professor of Sociology at Harvard University. He is co-editor and co-translator, with Robert Alun Jones, of Durkheim’s Philosophy Lectures: Notes from the Lycée de Sens Course, 1883–4 (Cambridge University Press, 2004). Recent pieces include ‘A General Theory of Scientific/Intellectual Movements’ (American Sociological Review, 2005, with Scott Frickel), ‘The Detraditionalization of Intimacy Reconsidered’ (Sociological Theory, 2005), and ‘Pragmatism, Phenomenology, and 20th-Century American Sociology’ (forthcoming in Sociology in America: The ASA Centennial History, Craig Calhoun, editor). He is completing a book tentatively titled Richard Rorty’s Pragmatism: The Social Origins of a Philosophy. Address: Department of Sociology, Harvard University, William James Hall, 33 Kirkland Street, Cambridge MA, 02138, USA. [email: [email protected]]

56