Lukes - Searle And His Critics

Anthropological Theory Copyright © 2006 SAGE Publications (London, Thousand Oaks, CA and New Delhi) http://ant.sagepub.com Vol 6(1): 5–11 10.1177/1463499606061729

Searle and his critics Steven Lukes New York University, USA

Abstract The ensuing articles pose various challenges to Searle’s thesis concerning social reality. Some exhibit misunderstandings; others identify inadequacies in the formulation of his thesis and failures to address issues within the limits of his project, notably his inattention to unintended consequences. Searle’s project is to distinguish social from biological and physical reality, but that, it is argued, offers a restrictive account of what social scientists study, which extends well beyond linguistically-constituted institutions to include the ‘brute realities’ of social life and, most significantly, the interactions between the ‘institutional’ and the ‘brute’, for example between ‘institutional’ and ‘brute’ power. Searle’s critique of Durkheim’s social ontology is, in part, endorsed but also criticized for focusing on the latter’s methodological pronouncements rather than on the ontology implicit in his substantive work. What bearing, in general, does getting social ontology right have on substantive social scientific work? Some suggestions are offered concerning the substantive implications of Searle’s theory. Key Words Durkheim • institution • intentionality • objective • ontology • power • social facts • social reality • subjective

When philosophers engage with anthropology, interesting things can happen. Lucien Lévy-Bruhl developed his ideas about ‘pre-logical mentality’ in reaction to the anthropology of his time (Lévy-Bruhl, 1910). John Ladd (Ladd, 1957), the Edels (Edel and Edel, 1959) and Richard Brandt (Brandt, 1954) explored the ethics of various tribal societies in response to the cultural anthropology of theirs. Peter Winch’s essay ‘Understanding a Primitive Society’ (Winch, 1964), deploying Wittgensteinian ideas in relation to E.E. Evans-Pritchard’s studies of the Zande and the Nuer, generated a debate, that ranged deep and wide, which continues to this day (see Wilson, 1970; Hollis and Lukes, 1982; Risjord, 2000; Lukes, 2006).1 It gets really interesting when the anthropologists and social scientists respond, as Durkheim did to Lévy-Bruhl (Durkheim, 1913), rejecting the idea of the pre-logical, and as Ernest Gellner (Gellner, 1968) and Robin Horton (Horton, 1976), among others, did to Winch. This remarkable symposium, consisting of John Searle’s essay, the various responses to it and his replies to these, constitutes a further, striking proof of the value of such mutual engagements. It is a coherent and 5

ANTHROPOLOGICAL THEORY 6(1)

focused collection of contributions that are all intellectually serious, challenging and well-written. They exhibit differing styles of thought, backgrounds and ranges of reference, but all grapple with the same range of basic issues raised in Searle’s characteristically clearly-stated and structured argument. These issues are fundamental to all the social sciences, but they are issues with which, as Searle remarks, contemporary philosophers rarely concern themselves. Roy D’Andrade points to the question at the center of Searle’s theory by stating a paradox: that ‘one cannot think an institution away because it is more than a thought, yet an institution cannot exist unless people think it is real’ (p. 34). A paradox is an apparent contradiction and Searle’s theory, which has that rare virtue of being at once simple and deep, offers to resolve the paradox, by showing that the contradiction is merely apparent. Its aim is to give an account of social reality in general and institutional reality in particular. Animals, including human animals, are social, exhibiting collective intentionality, but human animals have institutions, created and sustained entirely in individual minds by collective intentionality and enabling and empowering them, through the constitutive power of language, to act on desire-independent reasons. To appreciate the force and reach of this abstractly stated thesis one must read The Construction of Social Reality (1995) and, more particularly, the essay published here, summarizing and developing that thesis. Institutional facts are facts about money, property, government, marriage and so on, but it is important to note that Searle uses ‘institution’ as a technical term of art, stressing that it presupposes language in its constitutive role, set out initially in his speech-act theory. What he seeks to do here is to address the question: What are the distinguishing features of human social reality? In doing so, he articulates a view, distinguishing ‘institutional reality’ from ‘brute reality’, that is, as the reader will discover, sharply distinct from several other views. His work is, as he comments in his response to D’Andrade, ‘an effort to answer the challenge of sociobiology’ with its ‘implicit message . . . that humans are not different from other social animals and that the terms in which we need to understand human social behavior are essentially biological and above all evolutionary’ (pp. 40–1). He claims to take language, and in particular its constitutive role, seriously, as opposed, for example, to the ‘rich tradition of linguistic anthropology’ and contemporary sociological theorists such as Bourdieu, Habermas and Foucault. He firmly asserts the view that ‘the world consists entirely of physical particles in fields of force (or whatever the ultimately correct physics tells us are the final building blocks of matter)’ (p. 13) and so rejects theories that postulate further realities. There is, he insists, only one (unified, causally integrated) world, not two, or, as Popper claimed, three, or, as Richard Shweder, advocating ‘romantic pluralism’, suggests, as many worlds as there are alternative world views. And, in response to Neil Gross, he rather fiercely distinguishes his view from Durkheim’s account of social facts, which, he writes, is ‘mistaken’ and ‘not remotely like’, indeed ‘quite the opposite of ’ his own. To that matter I will return. D’Andrade’s article is a thoughtful and thought-provoking attempt to think through and re-state in social scientists’ language (‘culture’, ‘values’, ‘norms’ and so on) Searle’s thesis and its implications. As Searle notes, his essay is exploratory and, at certain points, indecisive, but it captures the main thrust of Searle’s idea and sees its importance, not least for how we are to think about personality, culture and social structure. I agree with Searle that the suggestion that we should no longer think of these as ‘separate kinds of 6

LUKES Searle and his critics

things’ but, as D’Andrade puts it, as ‘words for different organizing processes’ is challenging and worth further reflection. The remaining articles pose a series of challenges to which Searle responds, with exemplary care and clarity. Among these, some are due to misunderstandings, some allege that there are deficiencies in his formulations and some point to his failure, within the limits of his own project, to address certain issues. Among the misunderstandings, which Searle succeeds, in my view, in rectifying, are these. In saying that institutional facts have a ‘logical structure’ he means (contra Friedman), not that they are not illogical or irrational, but that they have a propositional content and structure (such as X counts as Y in context C: thus these actions count as declaring war in these circumstances). He is not, in the accepted meaning, a ‘materialist’ (Friedman), even an inconsistent one (Shweder), or a ‘reductionist’ (Gross), reducing ‘narratives’ to ‘constitutive rules’. His ontological project does not exclude giving historical explanations (Friedman and Gross); he describes ‘logical structures, not their histories or their relations to larger social contexts’ (reply to Friedman). And Searle responds to Shweder’s detailed and eloquent ‘romantic pluralist’ critique by insisting, correctly, that he can, indeed that one must, distinguish institutional facts that involve illusions and falsehoods (e.g. that being a witch gives you supernatural powers) from those that do not and that his position does not require him to accept the perspectivedependence of all social knowledge claims. He also correctly denies that his ‘external realism’ is a metaphysical ‘leap of faith’, arguing that it must, indeed, be presupposed in order to identify the spirit- and ghost-infested belief systems in which anthropologists such as Shweder are interested. And, responding to Gross, he refutes the objection that his notion of collective intentionality excludes habitual acceptance of relations of domination, while admitting the central importance of explaining how this occurs – a topic on which I, for one, would rejoice in reading his reflections. A challenge which, in my view, reveals an inadequacy in Searle’s formulation of his view is Friedman’s doubt about his use of the notion of ‘observer-dependence’ to characterize the ‘intentionality’ and ‘ontological subjectivity’ of institutional facts. In response, Searle says that he means by ‘observer’ to indicate ‘all of the forms of intentionality that human beings have in dealing with their environment’. Indeed, he goes on to say that he is referring to ‘participants in the institution, and not the observer from outside’. But this is genuinely confusing, and it would have been clearer to speak of ‘participant-dependence’. (In using, possessing, borrowing money and so on, we do not ‘observe’ it.) I also think that Friedman has secured a hit with his all too brief observation about ‘non-intentional systemic realities’ – of which Friedman cites business cycles as an example. In citing the unexpected effects of players being right- or left-handed pitchers and batters, Searle has not taken the measure of this objection, though from his reply it looks as though he acknowledges its importance. As things stand, unintended consequences play no role in Searle’s argumentation and that must be a weakness, given his ontological project. Social institutions – economic, political, judicial, penal, and so on – are non-transparent, or opaque, to the extent that individuals who ‘accept’ them neither expect nor understand how they function and what their effects are. As Friedman argues, they have ramifying consequences for all of us. It is social scientists’ chief task to reveal these. Surely, therefore, opacity must be a crucial aspect for which a theory of social ontology must account. Furthermore, several of the contributors seem uneasy with 7

ANTHROPOLOGICAL THEORY 6(1)

his use of the term ‘acceptance’, as in the formula ‘We accept (S has power [S does A])’. What exactly is acceptance and shared acceptance? What are the criteria for identifying its presence? How do we distinguish genuine or voluntary ‘acceptance’ from acquiescence induced by power relations? This deep and difficult question relates to that raised by Gross referred to earlier. These questions concerning non-intended consequences and power raise a more general issue which none of the contributors ventures to press. ‘Institutions’ (whether in Searle’s sense or a wider sense) do not exhaust the realm of the social. Consider, for example, what Durkheim called ‘social facts of an “anatomic” or “morphological” nature’: ‘the number and nature of the elementary parts which constitute society, the way in which they are articulated, the degree of coalescence they have attained, the distribution of population over the earth’s surface, the extent and nature of the network of communications, the design of dwellings, etc.’ (Durkheim, 1982: 57).2 Searle’s title is ‘Social ontology: Some basic principles’ and his book is entitled The Construction of Social Reality, and this might suggest that his topic encompasses the whole domain of social facts that constitute the object of the social sciences, such as anthropology. But, as I have sought to make clear, his interest as a philosopher is in ascertaining what is uniquely characteristic of human beings and thus distinctive of human social reality, that is, what distinguishes it from the biological and the physical, and his answer is: linguistically constituted institutions. But that is not to say that social scientists’ object of inquiry and explanation is similarly restricted. On the contrary, their concern must be with both the ‘institutional’ and the ‘brute’ realities of social life and, most particularly, with the zone of interaction between them. Consider, for example, political power. Institutionalized power consists, as Searle writes, of deontic power. This embraces rights, duties, obligations, commitments, authorizations, requirements, permissions and privileges. But ‘brute’ power does not always coincide with these, with what the constitutive rules require. Bargaining power and also the power to induce acquiescence to disadvantage or subordination are obvious instances of this. Which constitutive rules get instituted and accepted can be determined by who wins a struggle for brute power. Furthermore, the very adherence of citizens to the constitutive rules can be a problem; people will have ‘brute interests’ which may or may not lead them so to adhere. When this occurs, ‘institutional interests’ can be forged to encourage such adherence, as when the socialist parties were induced to enter into electoral competition, or when the American constitutional founders divided power into separate branches, so that these would develop their own separate interests. In short, as Ignacio Sánchez-Cuenca has argued, we need a theory ‘capable of explaining how brute and institutional power interact in concrete cases’ (Sánchez-Cuenca, 2003: 90).3 I can see no reason why Searle need deny any of this, though he does sometimes write as though the non- and extra-institutional dimensions of the social were of no significance (especially when choosing his titles). His rather homely examples, typically from games, university life and contemporary politics, do not raise the key issue of legitimacy. And in a separate essay on ‘Social Ontology and Political Power’ he displays a similarly restricted, institutionally consensual view of politics, writing that ‘all political power is deontic power’, indeed that ‘the essence [sic] of political power is deontic power’ (Searle, 2003: 204, 207, emphasis in original). If what I have argued here is cogent, then this is at best highly misleading. 8

LUKES Searle and his critics

One sub-theme running throughout these texts is what we might call the Searle–Durkheim debate, or rather confrontation. This arises from Gross’s challenge, which turns out to be far more of an attack on Searle than its author evidently intended. To Searle’s considerable discomfort, Gross claims that his work re-expresses Durkheimian ideas in an analytic idiom (the other contributors make similar claims). In response, Searle mounts a 10-point refutation of Durkheim (for the sake of brevity). Some of this is, in my view, compelling as a critique of The Rules of Sociological Method and Durkheim’s essay on ‘Individual and Collective Representations’. Especially helpful is his dissection of Durkheim’s favored analogies in claiming the sui generis reality of social facts, as set out in the latter. Searle effectively shows just why the analogies of chemical alloys, of living matter and its chemical ‘substratum’ and of mental life and the activities of neurons are unsatisfactory. He effectively shows that Durkheim runs together four different senses of ‘coercion’ when explaining the constraining character of social facts.4 He also complains about Durkheim’s view that social facts are ‘things’. He cites Durkheim insisting that they be studied objectively as external things, rather than representations in the mind. But what Durkheim is here insisting upon is precisely the need to study them as data, rather than relying on preconceived and speculative ‘facts’. Moreover, Durkheim’s slogan ‘consider social facts as things’ needs to be understood in its context. Impressed by the successes of German social science, Durkheim was urging that education in Third Republic France should be ‘à l’école des choses’ to induce in students a proper sense of the complexity of their social and political world, the better to predict and control it. Searle is, of course, right that Durkheim lacked his distinctions between observerrelative and observer-independent phenomena and between the ontological sense of the objective/subjective distinction and the epistemic sense, though, as already stated, I think it clearer to say that ontological subjectivity is a case of participant-, not observerrelativity. Clearly, Durkheim was after epistemic objectivity. Searle writes that Durkheim failed ‘to see that we can have an epistemically objective science of a domain that is ontologically subjective’ (p. 63, emphasis in original). This I doubt. Searle claims that Durkheim ‘thinks that collective representations are not in the minds of individual actors’ (p. 66), yet he himself quotes Durkheim writing that the conscience commune ‘is only realized in individuals ’ (p. 61, my emphasis). There are many such assertions by Durkheim, alongside others which suggest a sui generis reality and even a collective mind and personality. The point is that Durkheim was, in making these apparently contradictory statements, expressing the very paradox noted by D’Andrade: that from the perspective of individual agents, social facts appear external and constraining, while from the analyst’s they (or, more specifically, the institutional facts in Searle’s sense) are only real because people think they are. My suggestion is that Durkheim did indeed see what Searle says he failed to see, but that he lacked the analytical apparatus to work it out. Moreover, much of Durkheim’s substantive sociology, especially his later writings, consists in the analysis of the functioning of collective representations. Searle is right that he ‘was not in a position to state the precise details of collective representations because he lacked the logical apparatus’ (p. 62). This raises a more general point. I do think that Searle’s critique of Durkheim lacks judicious balance or, better still, fairness. Though The Rules, in my view, provides a poor guide to his practice, and although I agree with Searle that his argument suffers from its innocence of subsequent advances in analytic philosophy, and although I fully agree that Durkheim fails to incorporate the enabling aspect of social facts into his purely 9

ANTHROPOLOGICAL THEORY 6(1)

constraint-focused account there, nevertheless Searle altogether fails to appreciate the explanatory power of his sociological vision, exemplified (as Gross rightly suggests) above all in The Elementary Forms of Religious Life. In fact, I think that Searle would come up with a much less dismissive view of Durkheim if he were to reflect on the ontology implicit in his substantive work as opposed to his inadequate reflections on his own ontology and methodology. This raises the last point I wish to make, in the form of a question. What bearing does this discussion of social ontology have on substantive social scientific work? (Do good practicing social scientists have to be good philosophers – or, more precisely, do they have to be good at giving an accurate and philosophically adequate account of their own practice? Very often, when they try, they are rather bad at it.) What can we explain better, and in what ways would our explanations be better, if we accept Searle’s ontological theory? Which kinds of explanatory approach should we avoid, and which should we pursue? Here are some tentative suggestions. First, following D’Andrade’s suggestion, we can avoid treating culture and personality as ‘different kinds of stuff ’ but view them as different ways of organizing their component elements. Second, in similar spirit, Searle’s work can induce a certain skepticism about the application of economics-based models to other spheres of social life, such as politics. Within economic life, motivation is typically assumed to be desire-dependent; hence the central role in economic theory of the notion of ‘preference’. As Searle himself has written, ‘not all political motivation is selfinterested or prudential’ and, comparing economic and political systems, he observes that, though ‘the logical structures are similar, the systems of rational motivation are interestingly different’ (Searle, 2003: 207). This is obviously no less true of other non-economic spheres of social life. Indeed, perhaps Searle’s theory, with its focus on the role of ‘desireindependent reasons’, suggests a plausible basis on which to criticize rational choice theories which often, perhaps inherently, involve a kind of ‘brute reductionism’ of institutional interests to brute desires (see Lukes and Haglund, 2005). And finally, and more positively, as I have suggested, we can apply Searle’s distinction by attending, as he does not, to the interactions between the institutional and the brute realities of social life. Notes

1 Two fine recent examples are Moody-Adams (1997) and Cook (1999). 2 This passage, indeed the whole paragraph from which it comes, was inexplicably omitted from the 1938 translation of The Rules, which was the only one available until 1982. 3 I am indebted to this fine article for the examples cited. 4 He is not the first to notice this. See Lukes (1973: 12–13). References

Brandt, R. (1954) Hopi Ethics: A Theoretical Analysis. Chicago, IL: Chicago University Press. Cook, J.W. (1999) Morality and Cultural Differences. New York and Oxford: Oxford University Press. Durkheim E. (1913) ‘Review of Lévy-Bruhl 1910’, L’Année sociologique XII: 33–7. Durkheim, E. (1982) The Rules of Sociological Method and Selected Texts on Sociology and its Method (edited with an introduction by Steven Lukes, translated by W.D. Halls). Basingstoke: Macmillan. 10

LUKES Searle and his critics

Durkheim, E. (1995) The Elementary Forms of Religious Life. New York: Free Press. Edel, M. and A. Edel (1959) Anthropology and Ethics. Springfield, IL: Thomas. Gellner, E. (1968) ‘The New Idealism – Cause and Meaning in the Social Sciences’, in I. Lakatos and A. Musgrave (eds) Problems in the Philosophy of Science, pp. 377–406. Amsterdam: North Holland. (Reprinted in E. Gellner, 1973, Cause and Meaning in the Social Sciences. London: Routledge and Kegan Paul.) Hollis, M. and S. Lukes, eds (1982) Rationality and Relativism. Oxford: Blackwell. Horton, R. (1976) ‘Professor Winch on Safari’, Archives européennes de sociologie 17: 157–80. (Reprinted in R. Horton, 1993, Patterns of Thought in Africa and the West: Selected Theoretical Essays in Magic, Religion and Science. Cambridge: Cambridge University Press.) Ladd, J. (1957) The Structure of a Moral Code: A Philosophical Analysis of Ethical Discourse Applied to the Ethics of the Navaho Indians. Cambridge, MA: Harvard University Press. Lévy-Bruhl, L. (1910) Les Fonction mentales dans les sociétés inférieures. Paris: Alcan. (Translated as How Natives Think by L.A. Clare, 1926. London: Allen and Unwin.) Lukes, S. (1973) Emile Durkheim: His Life and Work. A Historical and Critical Study. London: Allen Lane Press. Lukes, S. (2006) ‘Apparently Irrational Beliefs’, in S. Turner and M. Risjord (eds) Philosophy of Anthropology and Sociology (in Handbook of the Philosophy of Science), pp. 274–90. London and Amsterdam: Elsevier. Lukes, S. and L. Haglund (2005) ‘Power and Luck’, Archives européennes de sociologie XLVI(1): 45–66. Moody-Adams, M.M. (1997) Fieldwork in Familiar Places: Morality, Culture and Philosophy. Cambridge, MA: Harvard University Press. Risjord, M. (2000) Woodcutters and Witchcraft: Rationality and Interpretive Change in the Social Sciences. Albany: State University of New York Press. Sánchez-Cuenca, I. (2003) ‘Power, Rules and Compliance’, in J.J. Maravall and A. Przeworski (eds) Democracy and the Rule of Law, pp. 62–93. Cambridge: Cambridge University Press. Searle, John R. (1995) The Construction of Social Reality. New York: The Free Press. Searle, J. (2003) ‘Social Ontology and Political Power’, in F.F. Schmitt (ed.) Socializing Metaphysics: The Nature of Social Reality, pp. 195–210. Lanham, MD: Rowman and Littlefield. Wilson, B.R., ed. (1970) Rationality. Oxford: Blackwell. Winch, P. (1964) ‘Understanding a Primitive Society’, American Philosophical Quarterly 1: 307–24. (Reprinted in B.R. Wilson, ed., 1970, Rationality. Oxford: Blackwell.) STEVEN LUKES is Professor of Sociology at New York University. He has previously held posts at Balliol College, Oxford, the European University Institute in Florence, the University of Siena and the London School of Economics. He is a Fellow of the British Academy and an editor of the European Journal of Sociology. His published works include Emile Durkheim: His Life and Work; Individualism, Marxism and Morality; Liberals and Cannibals: The Implications of Diversity; The Curious Enlightenment of Professor Caritat: A Comedy of Ideas and Power: A Radical View, which has just appeared in a much expanded second edition, published by Palgrave. He also co-edited Rationality and Relativism and The Category of the Person: Anthropology, Philosophy, History. [email: [email protected]]

11