John Searle is in a café in Paris. The waiter arrives. ‘Un demi,’ Searle asks, ‘Munich, à pression, s’il vous plaît.’ The waiter brings the beer. Searle drinks it, puts a few francs on the table, and leaves. ‘An innocent scene’, he agrees, ‘but its metaphysical complexity is truly staggering, and its complexity would have taken Kant’s breath away if he had ever bothered to think about such things.’ Kant didn’t think about such things because, at the time, philosophers were obsessed with knowledge. ‘Much later,’ Searle observes, ‘for a brief, glorious moment, they were obsessed with language. Now this philosopher at least is obsessed with certain general structures of human culture.’
Sociologists have been trying to think about this matter for more than a hundred years. The first generation, however, people like Emile Durkheim, Georg Simmel and Max Weber, lacked what Searle regards as the necessary equipment with which to tackle it. They had no theories of the kind that Searle has spent his own career spelling out and defending in books on speech acts, expression and meaning, intentionality, The Rediscovery of the Mind and the exaggerated claims of cognitive science about minds and brains. Recent generations of sociologists have been more fortunate. But they’ve not been as clear as one might wish, and have made mistakes. They have tended to think that since institutions are socially ‘constructed’, they can only exist by some sort of agreement. If this is the case, they have thought, institutions can’t in the philosophical sense of the word be real, can’t exist independently of any agreement to present them as the institutions they are.
Searle’s characterisation is correct. Ethno-methodologists, for example, a species of sociologist that first appeared in Southern California in the Sixties, used to instruct their students to go out into the social world and turn it upside down. They were to beg from beggars, arrest policemen, instruct salesgirls to try on the clothes they were employed to sell. They were to perform acts which even in Southern California in the Sixties would, in the reactions they produced, show that social life rested on nothing more substantial than convention. An instructor at Santa Barbara used to demonstrate the point. He would come into class in his wet-suit, still goggled and dripping from the ocean.
The attitude has been extended to all inquiry. Our conceptions are all we have. There is no point of view that is not someone’s point of view. All views are conventional, there is no firm ‘foundation’ for any, and none is privileged. Searle dislikes what this has licensed. ‘In my observation,’ he says, ‘the rejection of realism, and the denial of ontological objectivity,’ of the view that the world is as it is and not otherwise, independently of any representation of it, ‘is an essential component of the attacks on epistemic objectivity, rationality, truth, and intelligence in contemporary intellectual life.’ This brisk, bold and extremely clear little book, the distillation of various lectures in America and Europe over the past three years, is his response. Realism, he wants to show, is right, even for social life. Moreover, our statements about things, if they’re true, correspond to the way things are. Ontologically and epistemically, as an argument about how to see the social world and about what we’re doing when we talk about it, this is unfashionable. Searle does not flinch. ‘Our aim is to assimilate social reality to the basic ontology of physics, chemistry and biology.’
We start, he says, with entities, pieces of printed paper, say, or magnetic traces on a disk. We impose a function or functions on these entities, making them mediums of exchange, indications of wealth, and so on. And we do this in a particular context, a context of ‘constitutive rules’, which makes exchanging and accumulating the paper or assigning the traces count as exchanging and accumulating money. At bottom, the entities are physical. ‘There are no institutional facts without brute facts.’ In more complicated institutions, of course, which is most of the institutions there are, the bottom is a long way down; there are many layers in between. Assigning the function of money to a physical object or a set of physical objects in a particular context or set of contexts allows us to assign further functions in further contexts whose constitutive rules in turn serve as objects to which we assign further functions in further contexts. There is money; the accumulation and exchange of money is assigned to banks and other financial institutions; further functions, like that of maintaining national currencies in central banks, are assigned to these institutions in further contexts; and so on. Institutions go deep, and usually stretch back through time.
That objects can perform a function in a con text, Searle insists, is a matter of rule. Those Russians, for instance, who a few years ago began to use packets of Marlboros instead of rouble notes, were introducing a new object to which to assign the functions of exchange and accumulation. In so doing, they merely changed the convention by which they assigned the function. They didn’t change the function itself or the rules for performing it. But when in June this year the Russian and Chinese Governments announced that they would abandon barter in their dealings with each other, in favour of using money, they were making a change in their rules of commercial exchange. The ethnomethodologists’ mistake was not to have seen this difference. They didn’t send their students out to disrupt the rules of begging, arresting and selling, or themselves defy the rules for teaching sociology. They merely changed the conventions for applying those rules.
Not all practices, of course, are rule-governed to the same degree. There are differences between exchanging, accumulating, begging, arresting, selling and teaching sociology. When one adds, say, having sex and shooting people, these become plain. For Searle, they fall along the continuum that runs from the wholly institutional to the barely social. His formula (of which, he insists, he does not want to make a fetish) is ‘X counts as Y in C,’ where X is the fact, Y the function and C the context of constitutive rules. What makes an act institutional, he argues, and not just social, is that the function assigned to the fact carries with it further functions. These are typically further rights and responsibilities and are indicated in how we describe the function. I can have sex, but this doesn’t make me a husband. I can organise with others to shoot to kill, but this doesn’t mean that I’m engaged in war. ‘There is,’ Searle suggests, ‘exactly one primitive logical operation by which institutional reality is created and constituted.’ What makes a practice an institution is that we consciously accept that a person or a group of persons has the continuing power to perform it.
So far, one might think, so good: admirably clear, but unremarkable. At least until that last sentence. We accept? We always do so consciously? And we always accept? Searle’s aim, to recall, ‘is to assimilate social reality to the basic ontology of physics, chemistry and biology’. ‘The central span on the bridge from physics to society,’ he insists, ‘is collective intentionality, and the decisive movement on that bridge in the creation of social reality is the collective intentional imposition of functions on entities that cannot perform those functions without that imposition.’ Many philosophers and social theorists will have trouble with this suggestion. Whether ends are given ‘exogenously’, as in the presently popular models of ‘rational choice’ (models of rational foolishness, as Amartya Sen has described them), or reflected on in the course of practical reasoning, they are, most will insist, the ends of individuals. It’s only individuals that can have intentions. Most philosophers, Searle concedes, have thought ‘that anybody who recognises collective intentionality as a primitive form of mental life must be committed to the idea that there exists some Hegelian world spirit, a collective consciousness, or something equally implausible.’ But we’re not required to posit such things. ‘The crucial element in collective intentionality is a sense of doing (wanting, believing etc) something together, and the individual intentionality that each person has is derived from the collective intentionality they share.’ The capacity for collective intentionality is ‘biologically innate’, as evident in us as it is in a pack of hyenas attacking a gnu. What humans distinctively do is impose a Y function on the X element and in so doing, usually through language, symbolise that function. ‘The capacity to attach a sense, a symbolic function, to an object that does not have that sense intrinsically is the precondition not only of language but of all institutional reality.’
Even if we accept this, however, and agree that we derive our intentions from a collective intention, it surely can’t be the case that in every practice we encounter we’re fully conscious of the rules? Searle agrees. To deal with the problem, he introduces what in his previous writing he’s called ‘the background’. This, he says, is what the later Wittgenstein was talking about, what the French sociologist Pierre Bourdieu has tried to capture in his notion of the ‘habitus’. It’s the taken-for-granted world. Some have thought of it as unconscious. That won’t do. Conscious mental states minus the consciousness is an empty riddle. The behaviourist account is unsatisfactory also. That simply eliminates intentionality altogether. Searle prefers to say that we’re good at picking up the rules because we’re sensitive to their structure. We haven’t learnt them. That’s impossible. There are far too many, and they’re far too complicated. Nor, as some psychologists and sociologists, visibly flailing, used to say, have we in some more obscure way ‘internalised’ them as the rules they are. If we’re ‘at home’ in our institutions and social practices, it’s because as a species we’ve evolved to develop a set of capacities and abilities that makes us so.
In this way, to this extent, we’re conscious of the rules. We accept that someone has, or has had, the power to impose a function Y on an object X in a context C, and that this function continues. Or we don’t. For we’re often not ‘at home’. We often dispute the rules, contesting the function that someone’s ascribed to something and suggesting another. If others are not at home either, they may respond and the practice may change, even end. Institutions, whole nation-states, even whole empires, as with the former Soviet Union, can collapse, and can sometimes do so very suddenly. When they do, new authorities will ascribe the functions they used to perform to another object, or borrow ascriptions from elsewhere, as the Russians have done, or perhaps invent new ones.
Searle’s ultimately biological explanation of how it is that we have institutions at all and work them may be correct. But it doesn’t refute the suggestion that as the particular institutions they are, they rest ultimately on nothing more than an agreement that they should function as they do. And it this is so, it would seem impossible to say that they exist independently of our representations of them. Searle, however, does want to say this.
Realism is easy for bits of paper and hyenas. Things of this sort are as they are independently of any account of them, independently of any account being given of them at all. This is the sense of realism that G.E. Moore tried to sustain when he held his hands up and invited the audience to look. It’s the sense that trades in the belief that to be real, objects must be objects, most comfortably, as J.L. Austin put it before we all became bemused by particles, ‘middle-sized dry goods’. To the extent that institutions at bottom do rest on such objects, in the way that money rests on certain metals, pieces of paper or magnetic traces, they too are real in this sense. But what makes an institution the institution it is is the way in which it’s symbolically represented. What can be real about that?
Searle answers the question by pressing two distinctions. The first is between the common view that realism says how things are, that it makes claims about the stuff of the world, and the view that realism says merely that the way things are is logically independent of all human representations. The second is between the view that realism enables us actually to say how things are, that it has epistemological promise, and the view that it doesn’t, that it’s merely what we have to suppose if we’re to have beliefs and intelligible conversations. Searle agrees with the anti-realists that we can’t say what the world consists in and that we can never know that we know that we’re representing it as it is. At the same time, if we didn’t suppose that what we’re representing was logically independent of our representing it, we’d find it impossible to have beliefs and talk intelligibly to each other. This is so equally for hyenas, institutions and God. This kind and degree of realism is presupposed by public language. The argument, as Searle describes it, is a transcendental one.
It’s not, he wants to be clear, a proof of external realism. There almost certainly couldn’t be any such proof that didn’t beg the question. Yet a correspondence theory of truth, he insists, makes sense. This is not to say that the familiar argument against this theory doesn’t itself make sense. There’s no escape from self-defeat once we see that to say that a statement corresponds to a fact in the world requires us to state what that fact is. (This is why the deeply disappointing disquotational theory of truth, that ‘S is true’ if and only if S is true, is thought to be the only acceptable theory we can have.) It’s rather to say that we must re-align our conception of ‘true’. ‘True’, observes Searle, comes from ‘trust’, which in turn derives from the Indo-European root deru, ‘tree’, something that suggests uprightness ‘and reliability generally’. When we say that statements are ‘true’, therefore, we don’t mean that they’re ‘factually’ correct. (‘Fact’, notwithstanding his own image earlier in the book of ‘brute facts’, is a snare; facts require statements, and take us back to the tautology of disquotation.) We mean that true statements are trustworthy. ‘The way they represent things as being is the way things really are.’ And the best way in English to describe this relation is to say that they ‘correspond’ to the way things are.
This is philosophy in the best modern American manner. It doesn’t start by assuming that the intuitions with which humankind gets through its day are irremediably confused. It first asks whether it isn’t certain philosophers who have got themselves to the far side of the moon with no way back. At the same time, it shows that when the rest of us use notions like ‘real’, ‘true’ and ‘correspond’, there are certain things we can’t coherently intend to say. Likewise with Searle’s sociology. This doesn’t disturb our existing intuition that institutions and other enduring social practices are constructed and maintained (or not) by common agreement – more exactly, by a common agreement on who or what has the authority to assign a certain sort of ‘status-function’, as he calls it, to a certain sort of object. What it does do, which is Searle’s own first purpose, is to elucidate the general structure of this agreement.
Structural simplicity, however, is one thing. Unravelling the metaphysical complexity is another. If the choice is between assuming that humans are endowed with mysterious powers of an irreducibly non-natural kind and believing that their consciousness and what we think of as its creative capacities are the upshot of an ultimately natural evolution, it’s a choice that is, for some of us, clear. Too much theoretical sociology, and not a little of the philosophy that sociologists have relied on, like the later Wittgenstein’s, has been animated by what intellectual Eurosceptics might think of as a distinctively Central European (in its more recent manifestations, often surreptitiously Catholic) project to defend Geist against nature. Searle’s achievement, in this respect much like Nietzsche’s, is to have shown how we might reconcile the two. Nietzsche was far ahead of other 19th-century thinkers: indeed, we’re still catching up with him. But like the others, he lacked the equipment with which to pursue his intuitions. We do not. For us, or so one hopes, evolutionary biology has at last been disentangled from spiritual sentiment and the belief that all good explanations must without loss reduce what they explain to things of another kind. We now can see what a non-reductive explanation of our collective capacities might look like.
To see the shape of such an explanation, however, is not actually to have it. In Searle’s sociology, there are two crucial moves. One is from individual to collective intentionality. The other is from the implausible assumption that we’ve learnt or ‘internalised’ all the social rules we use to the suggestion that we’re causally predisposed to sense them. The first is little more than an inference from the fact that language is public and that we frame our intentions in it, together with the assumption that collective intentionality is innate. The second is a way of reformulating the incoherent idea of an unconscious consciousness in order to make consciousness safe for evolutionary biology. Each, one might say, merely shows us where we might look further. But this is no mean achievement. It might help to put us back on track after two hundred years of idealist diversion and behaviourist evasion.
Send Letters To:
London Review of Books,
28 Little Russell Street
London, WC1A 2HN
Please include name, address, and a telephone number.