The Construction of Social Reality: An Exchange

Barry Smith and John Searle

American Journal of Economics and Sociology, 62: 2 (2003), 285-309

PDF Version


I. Barry Smith: The Ontology of Social Reality

A Two-Levelled Ontology

In The Construction of Social Reality, John Searle argues for a two-level ontology along the following lines. Facts on the lower level - which he calls brute facts - can exist independently of human beings and their institutions. Facts on the upper level, which he calls institutional facts, depend on human institutions and above all on an associated 'collective intentionality'. The existence of the Planet Earth is a brute fact, the existence of Utah is an institutional fact.

As Searle confesses, there is a sort of magic involved when 'we impose rights, responsibilities, obligations, duties, privileges, entitlements, penalties, authorizations, permissions ... in order to regulate relations between people'

[There is a] continuous line that goes from molecules and mountains to screwdrivers, levers, and beautiful sunsets, and then to legislatures, money, and nation-states. The central span on the bridge from physics to society is collective intentionality, and the decisive movement on that bridge in the creation of social reality is the collective intentional imposition of function on entities that cannot perform these functions without that imposition. (p. 41)

I am concerned here with Searle's account of what social objects are, whereby I use the term "object" in the widest possible sense to include both individual things (such as California driving licenses), powers (such as powers of a Supreme Court Judge), and relations (such as relations of ownership or authority over). Searle himself spends much of Construction explaining how social objects come into being. This question is not at issue here. Searle tells us what social objects are by giving us an account of the way the two levels are linked together, via the formula X counts as Y in context C. His ontology of social reality thus rests on three components

1. certain physical objects

2. certain cognitive acts or states in virtue of which such physical objects acquire certain special sorts of functions

3. these functions themselves

4. contexts in which the given cognitive acts or states are effective.

We shall need to investigate more closely in the sequel what sorts of entities are involved under each of these headings.

Consider, for example, a dollar bill. Here X is some physical object - a piece of paper with green printing on it. Y is the dollar bill, a social object. C is, for example, a bank in Miami.

The formula is couched in the object mode: X and Y are objects. In other places Searle prefers a fact mode. An institutional fact, he tells us, is a brute fact plus an assignment of function:

What is true of money is true of chess games, elections and universities. All these can take different forms, but for each there must be some physical realization. This suggests what I think is true, that social facts in general, and institutional facts especially, are hierarchically structured. Institutional facts exist, so to speak, on top of brute physical facts. Often, the brute facts will not be manifested as physical objects but as sounds coming out of peoples' mouths or as marks on paper - or even thoughts in their heads. (p. 35)

I believe that, expressed in the fact mode, Searle's account is correct but incomplete; it provides only a first, and almost trivial, part of an account of what social reality is. When expressed in the object mode it is more nearly complete, but incorrect. The object mode expression is more nearly complete, (a) because it gives a more extensive account of the phenomena of social reality (economic, legal, political, social phenomena of a wide range of types); but it is more complete also (b) because it tells us what such phenomena are they are objects (values of 'Y' in the formula), which result from the imposition upon other objects (values of 'X' in the formula, ultimately physical objects such as pieces of paper with green ink on them) of special sorts of functions. The problem is that there are many sorts of social objects which are fully comparable to Searle's favorite examples of such objects but which do not satisfy the formula because there is no corresponding value for the term 'X'.

Consider the money in my bank account, as recorded in the bank's computers. Searle in the following passage suggests that the social object in question fits his schema perfectly well (though he slips, revealingly, into the fact mode): there has, he insists,

all sorts of things can be money, but there has to be some physical realization, some brute fact - even if it is only a bit of paper or a blip on a computer disk - on which we can impose our institutional form of status function. Thus there are no institutional facts without brute facts. (p. 56)

But does a blip on a computer disk really count as money? Try using it to buy something with. Or does it not much rather represent money (in the way that it might also represent dollar bills or bars of gold in a safe)? Searle seems here to confuse records pertaining to the existence of something with that something itself. If I am correct about this, then the domain of money is, when measured in terms of the X counts as Y in C formula, a gappy affair as far as its physical underpinnings are concerned. Some money is the product of imposition of status functions. (He waves a dollar bill.) Not all money is so.

What are Searle's options, here?

1. He can deny that money in a bank account is money. That it is merely as if there is money there. Money will be brought forth, in appropriate amounts, whenever I go to the bank and request it; for the moment, though, there are only records (computer blips), which determine constraints on such bringing forth of money.  Searle cannot, I believe, accept alternative 1., since this would represent a departure from the realist theory of social objects to which he has otherwise remained faithful. (Moreover, alternative 1. may be a stepping stone to a theory which Searle would surely reject, to the effect that all talk of social objects is merely a façon de paler about other things.)

2. Searle can accept that there are two sorts of money (and two sorts of social objects in general), one of which satisfies the X counts as Y in C formula, the other of which demands a different sort of account (which Searle then still owes us).

3. In the following passages, Searle suggests a third alternative:

Social objectrs are always … constituted by social acts; and, in a sense, the object is just the continuous possibility of the activity. A twenty dollar bill, for example, is a standing possibility of paying for something. (p. 36)

What we think of as social objects, such as governments, money, and universities, are in fact just placeholder for patterns of activities. I hope it is clear that the whole operation of agentive functions and collective intentionality is a matter of ongoing activities and the creation of the possibility of more ongoing activities. (p. 57)

But here, again, he seems to come threateningly close to a sceptical theory of social objects, according to which there are not social objects (like California drivers licenses) after all, but only (somewhat vaguely) 'patterns of activities'. Certainly there are patterns of activities associated with California drivers licenses, but it is bad ontology to confuse the one with the other.

The problem I have identified is not confined to the case of money in a (fractional reserve) bank. The same problem arises, perhaps still more blatantly, in the case of property rights, debts, claims, obligations, and other like relational phenomena in the social world. Searle promises an account of such relational entities at the very beginning of his book (expressions referring to such entities are here picked out in bold):

Consider a simple scene like the following. I go into a café in Paris and sit in a chair at a table. The waiter comes and I utter a fragment of a French sentence. I say, "un demi, Munich, à pression, s'il vous plaît." The waiter brings the beer and I drink it. I leave some money on the table and leave. … [Notice] that the scene as described has a huge, invisible ontology: the waiter did not actually own the beer he gave me, but he is employed by the restaurant which owned it. The restaurant is required to post a list of the prices of all the boissons, and even if I never see such a list, I am required to pay only the listed price. The owner of the restaurant is licensed by the French government to operate it. As such, he is subject to a thousand rules and regulations I know nothing about. I am entitled to be there in the first place only because I am a citizen of the United States, the bearer of a valid passport, and I have entered France legally. (p. 3)

Searle might argue that my property right in relation, say, to a given parcel of land, is accounted for as follows: there is a certain physical item, the deeds to the property in my safe, which count as the property right in certain contexts. Here again, however, it seems that the deeds merely record or register the existence of the property right. An IOU note, similarly, records the existence of a debt, but it does not count as the debt. Moreover, even if a piece of paper, in a given case, truly does serve as the physical underpinning for the debt in the sense of Searle's formula, there are many other cases where debts exist with no paper record at all. Searle would say, perhaps, that the physical underpinning here is provided by blips (memory traces, beliefs) in peoples' brains; but once again, it seems ontologically wrong to state that blips in brains may count as debts in certain contexts. (And it seems wrong, also, to suppose that, by destroying such blips we would thereby succeed in destroying the debt.)

Relational social objects can exist even in the absence of all pieces of paper and in the absence of all blips (in brains or computers) and of records of any form. Imagine, for example, that we have before us a valuable Dutch painting. The painting is now owned by a certain family, say the family Lucca, a fact that we can now establish on the basis of duly accredited documents. But these documents were only issued yesterday, after an exhaustive 10 year investigation into the Lucca family's claims. This investigation revealed that there were, some 100 years ago, 4 people in a room in Amsterdam, one of whom was the then accredited owner of the painting, the other of whom were representatives of the Lucca family. 90 years later the painting is found in cellars of the Lucca family in Genoa. The latter's rightful ownership of the painting is then established, ontologically speaking, on a negative basis, i.e. it is established on the basis of the absence of any competing claims. And if the Lucca family is the rightful owner of the painting now, then they were the rightful owners also during the intervening 90 years during which there were no documents to serve as the underpinnings of this fact in the sense of Searle's formula.

Other tricky cases for Searle's formula include works of music. Here again, the score does not count as the work; the score is, rather, analogous to bank records or to a deed of sale. And the performance, too, does not count as the work (not least for the reason that the work exists, as we commonly suppose, even when it is not being performed).

Searle's account of real estate, in contrast, seems at first sight to be more fully in harmony with the X counts as Y in C formula:

Consider for example a primitive tribe that initially builds a wall around its territory. ... suppose the wall gradually evolves from being a physical barrier to being a symbolic barrier. Imagine that the wall gradually decays so that the only thing left is a line of stones. But imagine that the inhabitants and their neighbors continue to recognize the line of stones as marking the boundary of the territory in such a way that it affects their behavior. ... The line of stones now has a function that is not performed in virtue of sheer physics but in virtue of collective intentionality. ... The line of stones performs the same function as a physical barrier but it does not do so in virtue of its physical construction, but because it has been collectively assigned a new status, the status of a boundary marker. (p. 40)

But consider the border of Colorado. This is an abstract mathematical line. Thus it is not determined by any physical features on the ground in Colorado (not by anything there, where the border is, which could count as the border). Rather, we may suppose, there are certain lines drawn on a map in Washington D.C. and the map in Washington is then analogous to a record of transactions in the bank (or to the standard meter rule in Paris). The border corresponds to no physical objects (to no molecules in the rock, to no continuous line of paint or signposts). It is a fiat border, not a physical border. Air-traffic corridors and the various other administratively determined regions of air space are likewise such as to have not physical, but rather fiat, boundaries. They are abstract volumes of space, again corresponding to no underlying physical reality.

To this Searle might respond that it is the volume of space which serves as value of X in the X counts as Y formula. But which volume? Relativity theory tells us that there is nothing in physical reality like a volume of space (what we call a volume of space is itself an abstract construction, a fiat object, carved in complicated fashion, out of a whole called space time). Thus the volume of space, as an entity carved out from this larger whole, is thus much more like a value of Y than it is like a value of X, and - problematically for Searle's theory - it is like a value of Y for which there is no pertinent, independent value of X.

Social Objects and Their Contextual Foundations

There are problems, too, arising from the role of contexts Searle's X counts as Y in context C formula. Consider necessary truths, such as

electric charge is either positive or negative,

space is three-dimensional,

nothing can be red and green all over,

every promise gives rise to a mutually correlated claim and obligation.

Philosophers of language in the middle of the century often defended contextual theories of the necessity involved in such necessary truths. Broadly, they held that necessary truths are in every case truths of logic. The propositions listed do not appear on their face to be logical truths. However, they each follow logically from the axioms of a certain pertinent scientific theory. Thus they can be converted into truths which are logical within the context of a theory. All necessity, they held, can in this way be reduced to logical necessity.

Such contextual theories of necessity face three sorts of problems:

(1) There are cross-contextual necessary truths - for example

green is not a cardinal number,

you can't smoke a phoneme,

electricity does not have an ethnicity.

These are truths which, because they transcend context, or because they involve a mixture of contexts, cannot adequately be dealt with on the basis of the contextual account. Defenders of the latter thus resorted to ad hoc banishments of such propositions into the realm of meaninglessness; to assert that green is not a cardinal number, they said, is to be guilty of a 'category mistake.'

(2) The pertinent axioms of scientific theories are themselves, at least in some cases, necessary truths of the problematic sort, and it cannot be an enlightening account of the necessity of p that it follows logically from p.

(3) There are necessity truths governing scientific theories themselves, truths which cannot (or cannot easily) be reduced to truths of logic. The defenders of a contextual theory of necessity are thus required to supply an account of these truths (an account of the contexts which play so central a role in the theory) which can be shown to be consistent with their reductionist program.

Searle's contextual theory of social objects, now, is subject to analogous objections, which can be formulated as follows

(1) There are social objects Y which transcend contexts in the strong sense that there is no over-arching context C in relation to which the counting as formula can be applied. We shall discover, in fact, that there is a rich variety of such cross-contextual social objects, and that some of them - whether they be in Kosovo, in the West Bank, in Nagorno-Karabhak - are of quite peculiar importance to the study of social and political ontology.

(2) There are certain fundamental types of social objects which are analogous to axioms of a scientific theory in the sense that they themselves are so basic in the edifice of social ontology that they cannot themselves be seen as products of cognitive acts in anything like the way suggested by the counting as formula.

(3) Included in this set of basic social objects are contexts themselves, the very objects at the heart of the Searle an theory. Contexts, too, are social objects. The defenders of this theory are thus required to supply an account of the ontology of contexts which can be shown to be consistent with its reductionist program.

Cross-Contextual Social Objects

Four types of cases can be distinguished:

Case 1 is of the form

X counts as Y in context C and X counts as Y1 in context C1, where neither C nor C1 has priority

Consider an area of territory X on the Indo-Chinese border, an area which India claims as Indian and China claims as Chinese. X counts as Indian territory in India-friendly contexts, and as Chinese territory in China-friendly contexts. What is the correct account of the ontology of this piece of territory, qua social object? If we adopt a neutral, scientific perspective we might say (truly)this is a social object which is conceived by India as Indian and by China as Chinese. It may even be the case that neither side has a legitimate claim to the territory in question, so that this territory is a social object for which only the external context-free description can do justice to it as a social object. It is a social object whose nature is exhausted by no single context, and which thus breaks the bounds of the Searle an formula.

Case 2 is of the form

X counts as Y in context C and X counts as Y1 in context C1, where C1 has priority

Suppose a sophisticated Italian forger has flooded Albania with fake dollar bills. There are many green pieces of paper which count as dollar bills in context C (here Albania), but the given pieces of paper are not dollar bills (as would become clear immediately were any of them to be presented for payment in an American bank). In virtue of what does the American bank context have priority over Albanian contexts? One tempting answer is this: that in the American bank context the given social object is revealed for what it is (rather than for what it merely counts as in some given context). Then, however, the intrinsic nature of the social object has priority over what gets imputed to the relevant underlying physical pieces of money in a given context. Once again, therefore, the Searlean theory is in trouble.

Case 3 is of the form

X counts as Y in context C and X1 counts as Y in context C1, where neither C nor C1 has priority

Suppose Y is the border between India and China in a given disputed area. Different X's are offered as candidates to be (or to count as) Y in different contexts. Again, Searle's theory proves to be inadequate as a theory of social objects like Y. He might respond that there are here two social objects, both of which are counted, in their pertinent contexts, as falling under the description 'is the border between India and China in area such-and-such'. The problem with this account is that it fails to do justice to the distinction (which Searle seeks to undermine) between counting as Y and really being Y. For people who really know about social reality in this region know that, while X counts asY to those on the Indian side, and X' counts as Y to those on the Chinese side, both groups are wrong. For by our hypothesis of equal priority of the two contexts, the qualified social ontologist/geographer/theorist of international relations is in a position to know either that the border is somewhere else, or he might know that there is no border in that region or that there is no border at all in the given region.

Case 4. is of the form:

X counts as Y in context C and X1 counts as Y in context C1, where C has priority

This case arises where one or other of the two sides enjoys priority (so that those on the Indian side are right to hold X as the border, and those on the Chinese side are wrong; the one context overwhelms the other). Consider the following example. The Berlin Wall, during the period of German separation, counted as the border between West Berlin and the surrounding territory of the German Democratic Republic. This was the case in almost all contexts, including contexts involving customs, administrative, economic matters, and so on. The actual border, however (i.e. the border as recognized by certain responsible international authorities, including those of the GDR), lay some meters away from the wall itself; it was separated from the wall by a buffer zone, which belonged officially to the GDR but was accessible only from the side of West Berlin, and by a death zone on the GDR side. (A similar buffer zone of some 100 meters separated the actual border between East and West Germany from the beginning of the barbed-wire fence/mine-field/tank-trap complex which, again in almost all contexts, counted as the border.)

Further problems arise as a result of the fact that for a time, the border between East and West Germany was not recognized by West Germany at all. Thus one and the same X counted as Y in one set of contexts, while in another context all claims to this effect were negated.

Can Searle argue that the majority, including the majority of institutional contexts, is in error here? If it is in error, why is it not in error when it ascribes ultimate authority to the pertinent international authorities (whose status is, after all, dependent upon their counting these authorities as having the powers they have)?

Suppose that I inveigle myself into the position where everyone believes that I am the owner of a certain property (perhaps I have killed the original owner and buried her beneath a thick slab of concrete in the cellar of her house). Suppose also that I have been able to destroy or to amend the relevant registration documents, so that my claim to this property cannot (or cannot easily) be contested. Then I count as the owner; but I am not the owner.

This is not an isolated or made-up example. There have been periods in history in which some violations were perpetrated systematically by government authorities, and it may be that, as a result of such violations, the only available historical information is that provided by the violators. As Bell and Geissel point out in their discussion of the case of German unification (http//, between 1945 and 1949 the old titles (as these had existed before the expropriation) entered in the land registration records were painted black or the pages were removed. New titles were written in by the expropriating East German administration, which had little respect for the rights of private property.

The problem here is that we have a whole panoply of social objects - pieces of property, ownership rights and relations, putative owners - but no contexts in which their status as social objects can be adequately accounted for according to the Searlean formula. Of course Searle might argue that there is a relevant overarching context, which is that provided by the current German government, which is striving to resolve a gigantic number of competing claims to rectification. But as Giessel and Ball write

Even the general solution currently under consideration in Germany has many administrative problems. The largest is to figure out who is entitled to which property. Often the real owner cannot be found and different people claim land or real estate as their own. Consider, for example, the East German government sale of a house of someone who left East Germany to move to West Germany. The current owner bought the house and holds a title. The refugee who fled to West Germany bought the house originally from the Nazi government in the thirties. He holds a title too. The same house is sometimes claimed by Jews or their descendants who saved their lives by leaving Germany in the thirties. They never received rectification from the East German government and have now, under the rule of West Germany, justified claims. Thus, there are three parties all of whom hold titles to the house.

These cases are not unusual. For example, in Kleinmachnow, a little town near Berlin in East Germany, 80% of the private houses are claimed by West Germans. In Schildow, East Germany, Mr. Nuscheler has filed a claim for his property there. He fled East Germany in 1977 and never sold his house. But his house was sold later by the local government to an officer of the East German army, Mr. Rosenthal. The registration record (Gundbuch) where Mr. Nuscheler was noted as the owner of the house mysteriously disappeared. Today, both of them hold titles for the same house. The case - 3 years after reunification - is still unresolved. On the East German island of Ruegen in the Baltic Sea, the local office for resolving open property questions reported that up to twelve people claim the same real estate as their own.

The phrase "counts as" is, after all, normally used precisely in order to draw attention to the fact that performances may count as (say) making a promise even where other conditions necessary to promising fail to be met (as when, according to the practices of a certain auction house or race track, lifting ones finger, even unintentionally, counts as making a promise). One might, of course, rule out this connotation by speaking instead of what "counts correctly as a promise", but this, it seems, could mean nothing other than "is a promise", and it is just locutions such as this which Searle is seeking to analyze by means of his counts as formula.

We can raise also the following epistemological criticism of the X counts as Y formula. How are we to give meaning to a phrase such as "counts as a promise" or "counts as a greeting" if we do not already understand terms like "promise" or "greeting" independently of the formula? How, in general, are we to make sense of talk of what counts as an X in the absence of any prior understand of what an X (in itself) might be? How could I ever come to know that such and such counts as a promise, unless I was independently familiar with promising itself? And what good would this knowledge be, even if it could be achieved? For if I know that something counts as X, and yet do not know what this 'X' signifies (really signifies), then surely I know nothing at all. (Smith 1993)

Contexts as Social Objects

Recall our two other objections to contextual theories of social objects

(2) There are certain fundamental types of social objects which are analogous to axioms of a scientific theory in the sense that they themselves are so basic in the edifice of social ontology that they cannot themselves be seen as products of cognitive acts in anything like the way suggested by the counting as formula.

(3) Included in this set of basic social objects are contexts themselves, the very objects at the heart of the Searlean theory. Contexts, too, are social objects. The defenders of this theory are thus required to supply an account of the ontology of contexts which can be shown to be consistent with its reductionist program.

Consider the following social fact

Context C overwhelms (has priority over) context C1.

Is this, too, a product of counting as in some further, higher-order context C2? If so, in virtue of what is it the case that

Context C2 overwhelms (has priority over) contexts C and C1?

If not, then again there are social facts which fall outside the scope of Searle's counting as formula.

The threatened vicious regress here is not a peripheral matter but cuts to the heart of Searle's theory since its earliest inception. There are, Searle tells us already in some of his very earliest writings, two different kinds of rules or conventions:

Some rules regulate antecedently existing forms of behaviour. For example, the rules of polite table behaviour regulate eating, but eating exists independently of these rules. Some rules, on the other hand, do not merely regulate an antecedently existing activity called playing chess; they, as it were, create the possibility of or define that activity. The activity of playing chess is constituted by action in accordance with these rules. Chess has no existence apart from these rules. (1969a, p. 131)

The same can be said also, from Searle's point of view, of the institution of promising

The institutions of marriage, money, and promising are like the institutions of baseball and chess in that they are systems of such constitutive rules or conventions. (loc. cit.)

Institutional facts are distinguished in this way, that they presuppose deliberate constitutive arrangements of the given sort. Many forms of obligations, commitments, rights and responsibilities are, Searle holds, a matter of institutional facts in this sense (and Searle holds that the oughtness of obligation follows as a matter of necessity from the isness which is the making of a promise). For Searle, this is a definitional matter the making of a promise is for him by definition a case of acting according to certain conventional rules and in these rules the notion of obligation is involved in the relevant sense.

Constitutive rules are for Searle purely conventional, as are the sorts of objects to which they give rise endowment mortgages, lien bonds, football team-managers, and so on. The corresponding institutional concepts are introduced into the language via non-circular definitions in terms of concepts which are unproblematically more basic. Clearly, however, we must by these means eventually arrive at basic institutional concepts, which is to sayinstitutional concepts not capable of being further defined on the institutional level. Context, presumably, is a concept of this sort, as also is counting as; others might beownership, rule, obligation, benefit, exchange, utterance, uptake, understanding, agreement, preference, sincerity, and so on. Similar basic institutional concepts are required also e.g. in the realm of games - concepts such as winning, losing, playing, breaking a rule, and so on. Moreover there are basic institutional concepts which must be involved wherever constitutive rules are formulated and adopted in the realm of the positive law, concepts such as command, decision, authority, consent, acknowledgement, jurisdiction, and so on.

Consider, now, the truths holding of such basic institutional concepts, including negative truths such as a context is not a waiving of a claim; a relation of authority is not a preference; sincerity is not obligation, and so on. The question for Searle is: are such truths purely conventional in the sense defined above? Clearly not: for the very formulation and adoption of constitutive rules presupposes concepts of the given sort. Are they, then, merely analytic? Certainly they are not truths of logic; and since we are dealing here with basic institutional concepts there are, ex hypothesi, no definitions, here, which could be eliminated in favor of more basic institutional concepts such a way as to exhibit the underlying structure of the truths in question as one of logical necessity. Can we, then, suppose that all such concepts can be defined in non-circular ways in terms of non-institutional concepts on the level of brute facts? Not at all, for then all institutional concepts would turn out to be thus definable, an outcome which Searle quite rightly rules out (1969, p. 56.) The only alternative which remains, therefore, is for Searle to accept that the given truths express irreducible material necessities, that is, that they express necessary relations between certain uninventable sui generis categories - an outcome which must surely, for Searle, be unpalatable. That he has not faced the necessity of drawing this conclusion follows from the fact that he has always already presupposed a rule-positing society, without ever asking how this society and its rule-positing practices and contexts came about).


Searle, John R. 1969 Speech Acts. An Essay in the Philosophy of Language, CambridgeCambridge University Press.

Searle, John R. 1969a "How to Derive 'Ought' from 'Is'", in W. D. Hudson, ed., The Is/Ought Question, LondonMacmillan, 120-134.

Searle, John R. The Construction of Social Reality, New YorkFree Press, 1995.

Smith, Barry 1993 "An Essay on Material Necessity", P. Hanson and B. Hunter, eds., Return of the A Priori (Canadian Journal of Philosophy, Supplementary Volume 18), (1993), 301-322.

II. John Searle: Reply to Barry Smith

I am very grateful to Barry Smith for his detailed comments on my book The Construction of Social Reality. I think in the end he makes many useful points, but I also believe that he misunderstands me in certain very profound ways. I believe his misunderstandings derive from the fact that he approaches this topic with a set of concerns that are fundamentally different from mine, and in consequence, he tends to take my views as attempts to answer his questions rather than attempts to answer my questions. I wrote a book which for the most part is not about social objects, or even about social facts, but it is mostly devoted to the analysis of institutional facts. I claim these are matters of status function, and that these status functions in general involve deontic powers. If you read Smith's article to me, you would not get the idea that I am writing about status functions and deontic powers. Furthermore, my entire approach is deliberately and self-consciously naturalistic. That is, I see the human ability to create money, property, government, and marriage as an extension of more basic biological phenomena such as the ability of human beings to engage in cooperative behavior, and their innate capacity for linguistic symbolism. My concern, in short, is with institutional reality, which is a special case of social reality. It is a matter of status functions, it is about the deontic powers acruing to status functions, and is utterly naturalistic. You will not get an impression that this is what is going on from reading his article.

I can summarize my misgivings by saying that there are three misconceptions that structure his paper. First he thinks that I am trying to analyze the nature of what he calls "social objects". He thinks that my analysis is that a social object is created by laying a function on top of a physical object, and that these social objects only exist in a social context. His second mistake is that he thinks that my formula "x counts as y in c" is intended as part of a definition of social objects; that is, he thinks it is intended to give us necessary and sufficient conditions for the application of social concepts, and these necessary and sufficient conditions are stated in terms of the ordinary use of the expression "counts as". Third, because he neglects the naturalism of my account, I think he thinks that I am trying to answer his various questions about necessity. I am not. He compares this problem to the old positivist attempts to reduce analytic propositions to logical truths. I thought the positivist enterprise was pointless from the beginning - it would be just as valid to reduce logical truths to ordinary analytic propositions: the reason that it is logically true that all unmarried men are unmarried is that it is a substitution instance of the much more fundamental and basic analytic truth, all bachelors are unmarried.

For me the formula "x counts as y in c" is intended as a useful mnemonic to remind us that institutional facts only exist because people are prepared to regard things or treat them as having a certain status and with that status a function that they cannot perform solely in virtue of their physical structure. The creation of institutional facts requires that people be able to count something as something more than its physical structure indicates. The formula is a simple summary of a complex thought. But this useful mnemonic is not intended as a definition of "social objects" or even of institutional facts, and indeed when I actually develop a formalism for the assignment of status functions, it is no longer essential to the analysis. If it is misleading, you could state the thesis of the whole book without it. I have found it immensely useful because it captures a crucial element of, namely that status functions depend on the attitudes of the participants in the social institution in question.

He objects to this that sometimes people might count something as something and still be mistaken. They might count a counterfeit dollar bill as a real dollar bill, even though it is not really a dollar bill. They might count something as the boundary of a territory even though it is not the real boundary of the territory. But I think he ought to ask himself the question, what fact about it makes it a real dollar bill as opposed to a counterfeit, and what fact about the real boundary of the territory makes it the real boundary, as opposed to the one that people think is the real boundary. And my thesis is that if you examine these questions carefully, you will find that the "counts as" phenomenon comes back in your face, even after you have distinguished between what people mistakenly count as so and so and what really is so and so. The reason for that is, to say that something is the real boundary even though it is not counted as the real boundary, or to say that something is a real dollar bill even though it is not counted as a dollar bill, in both cases is to say that there are certain preferred criteria according to which we count certain things as certain other things, and it is those preferred criteria which should prevail over the actual practices in question. But in both cases, both the case where the counterfeit is erroneously counted as a dollar bill, and the case where the real dollar bill is correctly counted as a dollar bill, the "counts as" formula, or some equivalent, is crucial to understanding the phenomenon in question, because the "counts as" formula captures the essential feature of status functions, namely that the function cannot be performed without a certain set of attitudes on the part of the participants in the institution. If the "counts as" formula is misleading, then we can simply get rid of it, and state it all in terms of the imposition of status function, but the notion of status function is not a term in ordinary speech, and I thought it helpful, indeed I have found it over the years immensely helpful, to be able to state the fundamental idea in colloquial terms, and I am not too worried about the fact that there are colloquial uses of these expressions that do express the complex idea I am trying to summarize with this formula.

The notion of a social object seems to me at best misleading, because it suggests that there is a class of social objects as distinct from a class of non-social objects. But if you suppose that there are two classes of objects, social and non-social, you immediately get contradictions of the following sort: In my hand I hold an object. This one and the same object is both a piece of paper and a dollar bill. As a piece of paper it is a non-social object, as a dollar bill it is a social object. So which is it? The answer, of course, is that it is both. But to say that is to say that we do not have a separate class of objects that we can identify with the notion of social object. Rather, what we have to say is that something is a social object only under certain descriptions and not others, and then we are forced to ask the crucial question, what is it that these descriptions describe?

Again, when I am alone in my room, that room contains at least the following "social objects". A citizen of the United States, an employee of the state of California, a licensed driver, and a tax payer. So how many objects are in the room? There is exactly one: me.

In so far as we do have a coherent notion of social object, it is derived from the notion of social and institutional facts. Thus there is only one object which is both a piece of paper and a dollar bill, but the fact that it is a piece of paper is not the same fact that it is a dollar bill, even though they are both facts about one and the same object. A typical question from my analysis is what is the relationship between the fact that this is a piece of paper and the fact that it is a dollar bill, how, so to speak, does humanity get from the facts about paper to the facts about dollar bills? I think you cannot ask or answer that question coherently if you start off with the idea that you are investigating the ontology of social objects.

The point can be put with a bit more precision using the resources of contemporary philosophy. The open sentence "x is a social object" is not extensional with respect to substitutability, thus it does not determine a class. This is not a harmless logical feature, because we tend to hear the notion "social object" on analogy with, for example, "objects made of iron". But "objects made of iron" form a distinct class. There is a distinction between objects made of iron and objects not made of iron. But there is not in that way a distinction between the class of social objects and the class of non-social objects, because one and the same thing can be a social object relative to one description, and a non-social object relative to another description.

Furthermore, many of the phenomena that are absolutely crucial to my analysis of institutional reality are not in any ordinary sense objects at all. Consider my obligation to pay money to you which obligation I incurred when I made a promise to you last week. When I impose the status function on my utterance, one might, decide to construe the utterance as an object, at least in the sense that it was an event that occurred in space and time. But what kind of an object is an "obligation" that persists after the demise of its physical creation? And this is not an exceptional case on my account, because the ontology of institutional reality according to me amounts to sets of rights, obligations, duties, entitlements, honors, and deontic powers of various sorts. Smith thinks of all of these as "objects", but I believe that it is an obstacle to understanding their nature if you think of them as all objects in the sense in which chairs and tables are objects.

In so far as it is useful to talk about social objects at all, they are derivative from what I call social and institutional facts. So both methodologically and logically it seems to me the analysis proceeds better if we examine institutional facts rather than social objects.

So far I have made two criticisms of Smith's approach. First that he is mistaken in thinking that the formula "x counts as y in c" is intended to give us logically necessary and sufficient conditions or a traditional logical analysis using the ordinary language notion of "counts as". Second, that I am trying to analyze social objects rather than social and institutional facts. This second is not merely a misunderstanding, but I think it is actually a weakness of his approach. Third, I think it is impossible for him to take my naturalism seriously, because he seems to have a lot of metaphysical commitments that are not made fully explicit, but I do not share them. For me, we are all animals, biological beasts, we share with all sorts of other animals the capacity for collective intentionality, and with collective intentionality you get social facts automatically. For me a social fact is simply any case of collective intentionality involving two or more animals. Intentional facts are more interesting, because they involve a deontic component, and with that deontic component comes the requirement of language. Smith says that I have "presupposed a rule-positing society, without ever asking how this society and its rule-positing practices and contexts come about". This is not quite right. I presuppose a society of biological beasts capable of collective intentionality, and evolution gives me that for free. I also presuppose another capacity given to us by evolution, namely the capacity to symbolize. That is an interesting and essential capacity, and one I have attempted to analyze in some depth \** .FS Intentionality, John R. Searle .FE But now what I want to know is, how do we get from these basic biological capacities to cultural institutional phenomena such as money, property, marriage, and government.

With these points in mind it seems to me that many other of his misunderstandings follow quite naturally, and I will simply list them.

1. He supposes that I offer a two-level analysis, presumably of social and non-social objects. But that is not correct. I started off with a distinction between brute and institutional facts, but by the time I stated the general theory of social and institutional reality, there are at least seven different levels. See the chart on page 121 for these levels.

2. Because he thinks that social objects are the phenomena we are analyzing, he thinks that a social object can only exist as long as the corresponding non-social object exists. He then purports to find a lot of cases where social objects exist without a corresponding non-social object. That is, he supposes, mistakenly, that in order for the y term to exist, the x term must continue to exist.

But on my account all of that is confused. Remember, my analysis originally started with speech acts, and the whole purpose of a speech act such as promising is to create an obligation that will continue to exist after the original promise has been made. I promise something on Tuesday, and the act of uttering ceases on Tuesday, but the obligation of the promise continues to exist over Wednesday, Thursday, Friday, etc. And that is not just an odd feature of speech acts, it is characteristic of the deontic structure of institutional reality. So, think for example, of creating a corporation. Once the act of creation of the corporation is completed, the corporation exists. It need have no physical realization, it may be just a set of status functions.

3. Smith concedes, "Searle himself spends much of his time explaining how social objects come into being. This question is not at issue here." But that is precisely the issue if we are really talking about institutional facts and not about "social objects". The whole point of institutional facts is that once created they continue to exist as long as they are recognized. You do not need the x term once you have created the y status function. At least you do not need it for such abstract entities as obligations, responsibilities, rights, duties, and other deontic phenomena, and these are, or so I maintain, the heart of the ontology of institutional reality. I am not interested, to repeat, in the category of social objects, but I am interested in such questions as how are institutional facts created and maintained, indeed I devote quite a number of pages to those specific questions.

4. Because he does not see that the analysis is really about institutional facts and not about social objects, he fails to see the distinction between genuine institutional facts and all sorts of other social phenomena that do not fit the category of institutional facts. Thus he takes me to task because my account does not deal with works of music. But works of music are not matters of status function. A piece of music performs its function in virtue of its physical structure, in virtue of the sounds in question. A work of music by itself is not a matter of obligations, rights, duties, or other status functions being imposed on the set of sounds in question.

This misunderstanding, I believe, reveals a fundamental misunderstanding of my entire project and the analysis of institutional facts. The crucial notions for me involve rights, duties, obligations and various other sorts of powers. Institutional reality, broadly speaking, is about power, remembering of course that some of these are negative powers, and some attenuated powers and have evolved into mere honors or dishonors. But the structure of institutional reality is a structure of power. And that is not the case with works of music. Works of music are just acoustical phenomena on which we have imposed a function, but not a status function.

5. The same mistake about the x term, namely the mistake of supposing that the institutional fact, or as he calls it, the "social object", can only exist as long as the x term exists, is repeated when it comes to contexts. He seems to suppose that a social object, in order to continue to exist, requires some specific context in which it must exist. Furthermore he thinks it is a requirement of my analysis that the contexts in question need not themselves involve institutional facts. But neither of these points is correct. It is certainly not universally the case where status functions are concerned. Often a particular context is required in order to create a status function, but once created it then exists in a fashion which is context-free. He seems to suppose that a continuation of the context is essential for the continued existence of a "social object". That is sometimes the case but not always. In order for Bill Clinton to continue as President of the United States, the United States has to continue to exist. But there are lots of institutional facts that become totally context-free, because the context that was essential to their creation, creates them in such a way that does not depend on any further contextual features. Thus, for example, the New York Yankees won the 1998 World Series. In order for their movements to count as winning it, those movements had to take place in a certain context. But once they have won it, then they are the victors of the 1998 World Series for all time and for all contexts.

Furthermore, it is a mistake that I thought I had blocked in the actual text, to suppose that in order to create institutional facts in a context c, context c must itself be non-institutional. That is not at all the case. Indeed I give many examples of how the context is itself institutional. Thus the context in which saying certain words counts as getting married in the state of California requires the presence of a qualified official. But the fact that someone is a qualified official is itself an institutional fact of the sort that I am analyzing. This is not a difficulty of the analysis, it is one of its features, and I believe one of its strengths that it shows the interlocking structure of institutional reality.

In the course of Smith's paper he gives a fairly large number of examples that he thinks create difficulties for me. I am extremely puzzled by these examples, because it seems to me they create no difficulties for my analysis at all. There are cases where there is, for example, a dispute about the ownership of a piece of property or a painting. These arise all the time. Now, the point I am making is that in order for us to even have an analysis of the nature of the dispute we have to understand that what is in dispute is the assignment of status functions. That is, the difficulties he raises about the Nazi expropriation of property, or disputes about the ownership of a painting, or about the boundary line between two countries, are real life disputes among people competing for the right to assign status functions to objects. They are not problems for philosophical analysis of the ontology of institutional facts, they are real life problems to be settled by judges and lawyers, and in the end perhaps by armies and political movements. It is amazing to me that he seems to think that I am trying to provide an algorithm for resolving political and legal disputes about boundaries and property ownership. I am doing nothing of the sort. I am trying to describe the logical structure of the dispute itself, not to resolve it. The fact that there are such disputes is not an objection to my account, it is a further illustration of its strength.

On at least one point it seems to me he is quite right, and the account I gave in the book is mistaken. I say that one form that money takes is magnetic traces on computer disks, and another form is credit cards. Strictly speaking neither of these is money, rather, both are different representations of money. The credit card can be used in a way that is in many respects functionally equivalent to money, but even so it is not itself money. It is a fascinating project to work out the role of these different sorts of representations of institutional facts, and I hope at some point to do it.

Some phenomena that seem to me clearly brute phenomena Smith denies are so. Thus the fact that the bounday line between Colorado and Utah is a sequence of extensionless points, does not make it a non-brute phenomenon. It is brute in the same sense that the edge of any physical object is a brute phenomena. A physical object can have an edge not itself extended, even though the existence of the edge in question is not an institutional fact. Furthermore, what he says about airspace seems to me mistaken. It is not the volume of the air in question, but it is an actual area over a part of the earth which counts as an airspace to which a certain airline might be entitled.

6. In a sixth misunderstanding, Smith says "Constitutive rules are for Searle purely conventional". But that is not quite right. I am very careful to distinguish between the constitutive rules and the conventional realizations of those rules. It is, for example, a constitutive rule that because I have satisfied certain tests to establish driving ability, I am a licensed driver in the state of California. Being able to drive is not a "conventional" way of obtaining authorization to drive, there is nothing arbitrary about the fact that licensed drivers, like licensed physicians and licensed brain surgeons, are supposed to meet certain physical criteria. The conventional element enters in because we may have different ways of establishing the ability to drive. So, in California they might use a multiple choice test in addition to the driving test, and in other states they might not use multiple choice. But the fact that the authorization to drive is based on actual physical abilities of drivers is not a matter of convention. It is the whole purpose of the constitutive rule to have a non-conventional fact underlying the assignment of the status function.

Finally, Smith seems to think that I have a "reductionist program". Nothing of the sort. I am trying to analyze how the world works, and one feature of the world that interests me is the ontology of institutional reality, the logical structure of institutional facts. But there is nothing in any interesting sense reductionist about the project.

To summarize my response to Smith, then, I would like to reemphasize these three points. First of all he has a mistaken conception of the basic tool of the analysis. I find the formula "x counts as y in c" immensely useful because it gives us a way of articulating the distinction between those functions where the function is performed in virtue of an intrinsic physical feature of the object, and those functions which are performed in virtue of collective recognition of a status. If he finds the "x counts as y in context c" formula confusing, then my whole analysis can be stated without it, it is just much more long-winded, and indeed in the crucial chapter on the ontology of the deontic powers, I do not use this formula. The second misunderstanding is not just a misunderstanding but I believe it is a mistake on his part. I think that attempting to analyze social and institutional facts in terms of "social objects" is barking up the wrong tree.

But the third point is the most important. I think he fails to see the need for a naturalistic account of institutional facts. Sometimes when discussing the difference between what people count as an institutioal fact and what really is an institutional fact, he assumes that there must be some real observer-independent truth about institutional facts. There must, for example, be some fact totally independent of human observers about whose property the painting really is. If my account is right this is a fundamental mistake.