A central claim in John Searle’s social ontology is that institutional entities exist in virtue of being represented as existing. This claim is indeed so central that it occurs already in the first two sentences in The Construction of Social Reality (Searle 1995), Searle’s seminal account of social ontology. Thus, we find Searle declaring it a puzzle that there “are objective facts in the world that are only facts by human agreement. In a sense there are things that exist only because we believe them to exist” (Searle 1995, p. 1). Searle’s work in social ontology can profitably be considered as an attempt to solve this puzzle.
On Searle’s account all institutional facts are representation-dependent facts. Thus, the fact that a piece of metal is money, the fact that a car is owned by a particular person and the fact that certain persons can be members of parliament obtain because we believe that they obtain. They exist in virtue of being represented as existing by a certain collective. In this they are to be distinguished from brute facts, which exist regardless of whether they are represented as existing or not.
Searle’s social ontology is explicitly designed to account for the presumed representation-dependence of institutional entities. Even so, it has been argued that some institutional facts do not depend on being represented as obtaining. Thus, we find Barry Smith (2001, 2003) arguing that some institutional facts obtain regardless of whether anyone believes that they obtain or not.
In this paper we shall examine one of Smith’s arguments to the effect that some institutional facts obtain even though they are not represented as obtaining. Smith argues that false beliefs about institutional entities can be prevalent in a society. Thus, for example, a piece of paper can be represented as a dollar, even though it is in fact a forgery, whereas a real dollar can be represented as a forged dollar. In the first case there is a representation of an institutional fact which does not obtain. In the second case there is an institutional fact that obtains even though it is not represented as obtaining.
In the next section I shall present Searle’s theory of social ontology and Smith’s argument to the effect that certain legal facts are not representation-dependent. I will not consider the full range of Smith’s arguments, but only those which pertain to a very restricted class of legal facts. In section three I shall argue that in order to account for Smith’s critique, the representationalist must work with a slightly richer ontology than Searle does. In particular, we must distinguish between institutional statuses and institutional functions. In section four I argue that in the problematic cases, institutional statuses and institutional functions depend upon different types of representations. And the fifth section I consider a possible reply from Smith.
2 Searle’s Theory and Smith’s Problem
According to John Searle, institutional facts obtain because they are represented as obtaining. So the fact that Barack Obama is the president of the USA obtains because we collectively represent Obama as the president of the USA. The fact that a certain coin is a means of payment obtains because we collectively represent such coins as means of payment, and so on and so forth.
On Searle’s account institutional facts exist in virtue of the fact that a certain collective accept certain constitutive rules. These rules do not regulate behavior that can exist in the absence of the rules themselves. Thus, traffic rules are not constitutive rules since it is apparently possible to drive a car but violate all traffic rules. Constitutive rules create the possibility of certain forms of actions rather than regulate pre-existing forms of action and behavior. Take for example the rules of a game. The rules for the game of chess create the possibility of playing chess (Searle 2010, p. 9).
A constitutive rule that is widely accepted in a context assigns status functions to objects. The status function of an object is the function that the object has in a certain context. Searle mentions as an example a line of stones (Searle 1995, p. 40). A line of stones may not physically manage to hinder people from crossing it. But it may be assigned the function of being a border between two areas. A quite different example concerns coins. A coin has no intrinsic physical property in virtue of which it has a monetary function. Yet it can nevertheless function as a means of payment because it has been assigned a status function by a constitutive rule.
Status functions are the bearers of what Searle calls “deontic powers” (Searle 2010, p. 8). This means that they are the bearers of “rights, duties, obligations, requirements, permissions, authorizations, entitlements and so on” (Searle 2010, p. 9). Consequently, it is because they carry deontic powers that status functions enable a person to perform certain institutional actions. For example, as a citizen of Sweden I have the right to vote in Swedish elections. So the status function of being a citizen gives me the deontic power to vote.
Status functions are assigned to objects by a constitutive rule when this rule is widely accepted in a society. What is required for an object to have a status function is consequently a collective representation to the effect that it has that status function. According to Searle, this representation has the following structure:
(i) X counts as Y in C (Searle 2010, p. 10).
X is in this context the object to which a status function is assigned. Y is the status function itself and C is the context in which X has the status function it is supposed to have. So in the case of a coin, X will be the piece of metal, Y will be the monetary value of the coin and C will be the contexts in which the coin has this monetary value. In the case of the president of the United States, X will at the moment of writing be Barack Obama, Y will be the status function of being a president, and C will presumably be the United States. 1
Now, in a series of papers (Smith 1992, 2001, 2003) Barry Smith has argued that Searle’s account is erroneous. One line of argument here is that Searle cannot distinguish between something counting as an institutional fact and something actually being an institutional fact. For according to Searle, institutional facts exist in virtue of something counting as having a status function in a certain context. But then, it might appear that if something counts as having a certain status function, it does as a matter of fact have this status function. So it follows that institutional facts cannot obtain unless they are represented as existing. But this is contested by Smith. 2
Let us take a look at some of Smith’s examples. Suppose that an expert forger has flooded Albania with fake dollars and that these count as genuine dollars in Albania. But they do not count as genuine dollars in the United States. If they were checked in an American bank, they would be discovered as forgeries. The American context has priority over the Albanian context in this case. So in the American context a given piece of paper is revealed as what it really is, namely a forgery. In the Albanian context on the other hand the piece of paper is counted as a real dollar. But it is not a real dollar, but merely a forgery (Smith 2003, p. 293).
Smith’s example shows that a collective representation to the effect that an institutional fact obtains is not sufficient for this fact to obtain. But we may easily add another twist to the story which shows that a collective representation to the effect that an institutional fact obtains is not necessary for this fact to obtain either. Consider the case where a real dollar finds itself in Albania. This piece of dollar is however, slightly damaged. So it is not treated as a dollar in Albania, but rather as a forgery. So no one represents the paper as a dollar. Yet nevertheless it is a real dollar. So the institutional fact obtains even though no one represents it as obtaining. A representation to the effect that an institutional fact obtains is consequently not necessary for this fact to obtain either.
Consider a slightly different example. Someone happens to be in possession of a certain property, which he does not own. Nevertheless, his community accepts him as the rightful owner. He has also destroyed or amended all relevant documentation regarding the ownership of the property so that his purported ownership is accepted by the authorities and his fellow citizens. We might even assume that everyone in the society believes him to be the owner. Smith points out that in that case he will count as the owner. But he will not be the owner. Everyone in the society has wrong beliefs regarding the proper ownership (Smith 2003, p. 295). So once again the example shows that a collective acceptance of a certain institutional fact is not sufficient for this fact to obtain.
The example can be extended further. If the possessor is not the owner of the property then presumably someone else owns the property. And we might easily assume that this person is not represented by anyone as being the owner of the property. Indeed, she might not even represent herself as being the owner of the property. Yet she would nevertheless be the owner of the property. But then institutional facts about ownership can obtain even though they are not represented as obtaining.
If this argument is correct, collective acceptance of an institutional fact is neither a necessary nor a sufficient condition for this fact to obtain. Institutional facts are not dependent upon representations.
3 Institutional Statuses, Deontic Relations and Institutional Functions
Searle has a reply ready for Smith. According to Searle, we would not be able to make the distinctions between real and forged dollars, unless we accepted that dollars are representation-dependent entities. So Searle grants that forged dollars are not real dollars. But the reason why forged dollars are not real dollars is that there are collectively accepted criteria for what is a dollar. And the counterfeit dollars do not count as dollars according to these criteria. So we need status functions and constitutive rules to explain the cases after all (Searle 2003, p. 301–302).
Here however, we encounter a problem. Searle seems to say in his reply that forged dollars are not real dollars because we represent conditions for what dollars are, and an entity is a real dollar only if it meets these criteria. So if a piece of paper meets the conditions for being a real dollar, it has the status function of being a real dollar. But, nevertheless, this cannot be the entire story. For a dollar can apparently satisfy the conditions outlined by this representation, and still be useless to its owner. This is after all what happened in the fictional Albanian scenario.
The problem is that on Searle’s account an entity has a status function if it satisfies certain collectively accepted criteria. But it can satisfy these criteria and still lack the powers normally associated with these status functions. Searle claims that status functions are the bearers of deontic powers. But in our example, the powers have become dissociated from the status functions. The real owner of the property does not have the power to perform the actions her status as owner would entitle her to. Someone else has those powers without being entitled to it.
If my analysis is correct, Searle’s analysis is on the right track, but it is too coarse-grained. On Searle’s account status functions are the bearers of deontic powers. Someone has a certain deontic power because she has a specific status function. Indeed, Searle goes so far as suggesting that there is an essential connection between status functions and deontic powers:
Institutional reality is a system of status functions, and those status functions always involve deontic powers. For example, the person who occupies an office near mine in Berkeley is the chair of the philosophy department. But the status function of being chair of the department imposes rights and obligations that the occupant did not otherwise have. In such ways there is an essential connection between status function and deontic power. (Searle 2008, p. 31)
But if the present account is correct, the relationship between powers and status functions are much less intimate than Searle supposes. For a status function and the power it normally carries, can, as it were, become disassociated from each other.
Now, in order to solve our puzzle, I propose that we distinguish between institutional statuses, deontic relations and institutional functions. This distinction gives us the resources to explain one of the peculiarities of Smith’s examples. For in these cases the institutional statuses have been separated from the institutional functions. A forged dollar is not a real dollar. But it has the functions normally associated with dollars. It would be misleading to say that it has the deontic powers associated with dollars, for a forged dollar does not give you the moral or legal right to use it as a means of payment. But if it is erroneously accepted as a real dollar, it nevertheless allows you to perform certain monetary transactions, even though you lack the legal rights to perform these. Conversely, a real dollar that is not represented as a real dollar, is a real dollar. But it lacks the powers normally associated with a real dollar. We shall express this in terms of entities having or lacking an institutional function.
What characterizes these cases is that an entity with an institutional status has become separated from the institutional function it is deontically related to. A piece of paper that has the institutional status of being a dollar is for example deontically related to monetary power. The possessor of the dollar has a legal right to use it in economic transactions. But in the Albanian case, the possessor of the real dollar lacks this power. The case of the forged dollar is different. Here, the possessor of the forged dollar has an institutional power to use it in economic transactions, even though the legal right of using it as a means of payment in economic transactions is lacking.
The case of the real dollar that is represented as a fabrication illustrates the fact that someone can have a certain institutional status but lack the institutional power it is legally entitled to. The case of the forged dollar illustrates the converse situation. Someone may have the power to perform institutional actions involving money, while not being legally entitled to it.
We should note that the same examples can be given with respect to legal obligations as well. For example, I have the obligation to pay taxes. But the tax collecting authorities may err when they calculate my tax rate. So I could be forced to pay more than I am obliged to pay. In this case, I fulfill an institutional role that I am not legally obliged to fulfill. But the converse can also occur. I might be taxed for a lesser sum than the law stipulates that I am obliged to pay. In this case, I have an obligation to fulfill an institutional role that I do no fulfill.
So we should depart from Searle’s account and accept that an institutional status is something distinct from an institutional function. Institutional statuses are deontically related to institutional functions. Entities with certain statuses carry rights to institutional powers or obligations to perform institutional roles. But the fact that they have these rights and obligations, does not entail that they have the corresponding functions. 3
The distinction between the institutional status of an entity and its functions makes sense for a second reason as well. A status is deontically related to a function. But which particular function it is deontically related to may change over time. Thus, the monarchs of Europe in the 17th century were legally entitled to a quite different set of powers and legally obliged to fulfill a quite different set of functions than most of the monarchs in Europe today.
So the institutional function of a bearer of an institutional status can change over time. Moreover, the institutional function can change from the bearer of one institutional status to the bearer of another institutional status. Thus, for example, the power to dissolve the government may once have been the prerogative of the monarch, but later came to be the prerogative of the parliament.
But what then is an institutional status? I would say that an institutional status just is the kind of institutional entity something is. There are consequently several different kinds of institutional statuses. Prima face, I would say that there are at least three. Thus, for example, there is first of all the status pertaining to institutions like corporations, governments or courts of law. Secondly, there are the statuses assigned to persons that are in various capacities related to institutions – members of parliament, chief executives, justices and citizens. And there is finally the status assigned to various artifacts – documents, money and other entities. The list is not supposed to be exhaustive; there may be more kinds of institutional statuses.
It is important to note that these institutional statuses differ in kind. Members of parliament are ordinary persons that happen to bear an institutional status. But unlike members of parliament, courts of law and corporations cannot exist independently of their institutional status. They are their institutional status.
An institutional function is the kind of function an institutional entity has in an overarching institutional framework. Citizens of some states have the power to vote in elections in that state and the function of paying taxes. Parliaments have the power to enact laws. Governments have the function of executing legislation passed by parliament, and so on and so forth.
Deontic relations finally are quite simply the normative relations to institutional functions that an entity bears in virtue of having a certain status. I have opted to describe institutional functions in terms of powers and roles. So the deontic relations between an institutional status and an institutional function can either be a right to a power or an obligation to perform a role. It is important to note that these are legal rights and obligations. Whether they also are moral rights and obligations is a quite separate matter.
There are plenty of examples of deontic relations between statuses and functions. Citizens of democratic states for example are entitled to the power to vote because the status of being a citizen is deontically related to the power to vote. Parliaments have the right to legislate because the status of being a parliament is deontically related to the power to legislate, and so on and so forth. The powers of citizens and the powers of parliaments are obviously just as different as the rights of persons and the rights of institutions. But both cases importantly involve the capacity to perform certain institutional actions in certain situations. If, on the other hand, an entity has a legal obligation, it is required to perform an institutional role. Citizens have the role of paying taxes and governments have the role of executing legislation passed by parliament.
4 Different Institutional Entities, Different Representations
I have distinguished between institutional statuses, institutional functions and deontic relations holding between statuses and functions. Facts involving all three of our institutional entities are representation-dependent. But they depend upon different kinds of representations. Or so I shall argue in this section of the paper.
In order to see this we need first of all to distinguish between representations de re and de dicto. Facts involving an entity having an institutional function always depend upon representations de re. But the fact that an entity has a specific institutional status normally depends upon representations de dicto. Facts involving deontic relations also depend upon representations de dicto.
Now, it is common in philosophy of mind and language to distinguish between de re and de dicto. But the distinction is often cashed out in different terms in different contexts. In the present paper we shall say that a representation of a particular object is de re if the representation refers directly and not through satisfying some condition or other. A representation of a particular object is to the contrary de dicto if the object is referred to through a description. In the latter case, we can say that an object is referred to because it satisfies the conditions specified in the description. A representation de dicto refers to those objects, if any, which has the properties or features described. A representation de re on the other hand refers to an object regardless of the properties it may have. It should be noted that a representation de dicto need not be a definite description.
I now wish to suggest that the fact that an object has a certain institutional status normally depend upon representations de dicto. The kind of representations upon which institutional statuses frequently depend will have the following structure:
(ii) An object which is a K is an S in C.
The C-term has the same function as in Searle’s original account. The K-term and the S-term are however, new. The K-term specifies a certain condition for objects to meet, whereas the S-term refers to an institutional status. The proposition expressed by (ii) then predicates the property of being an S to all objects satisfying condition K. So what we have done here is quite simply to formulate the criteria mentioned by Searle in his response to Smith. We have also noted that this does not require a de re representation of the entities satisfying this condition. 4
In the case of being a citizen of the United States for example, (ii) might be cashed out in – the following way:
(a) Anyone who is born in the United States is a citizen of the United States.
The sentence (a) expresses a sufficient (but not a necessary) condition for being a citizen of the United States. The proposition expressed by (a) predicates the status of being a citizen of the United States to all persons born in the United States. It does this regardless of whether the person is de re represented as a citizen or not.
It is however, to be noted that I do not claim that all facts regarding institutional statuses are dependent upon representations of type (ii). Thus, some facts presumably depend upon de re representations. If there is a coup d’etat in a particular country, the new president (or prime minister) is presumably not president because he satisfies a general condition, but because he is widely de re represented as being a president. Thus, for example, general Pinochet became President of Chile in 1974 not because he satisfied any general condition stipulated by a legal code for being president, but because his predecessor was killed and Pinochet’s claim that he was the successor was widely represented as correct in Chile.
Descriptive representations of the kind specified in (ii) characterize legal systems. Laws and legal rules are normally formulated in impersonal terms. They rarely specify which particular individual object has a certain status. Rather, they specify the conditions an object must meet in order to have a specific institutional status. Thus, for example, whereas laws and legal rules normally do not specify who owns what property, they do specify the conditions for something to be a property and for someone to be the owner of something. If someone satisfies the condition for being an owner, then she is legally the owner of the property. She has the status of being the owner. In order for this to be the case, she need not be de re represented by anyone as the owner of the property.
Something similar is true of money as well. The law specifies which conditions an entity must meet in order to be money (normally, it must have some physical properties and the right kind of causal prehistory). Coins meeting these conditions are money, regardless of whether they are de re represented as money or not.
If we apply the present account to our two problematic cases we can see that the real dollar that is counted as a forgery, is de re counted as a forgery. But whether or not something is a dollar, does not depend upon whether it is de re represented as a dollar. It depends to the contrary upon whether it satisfies the general criteria for being a dollar. And since it satisfies this condition, it is in reality a real dollar. Conversely, the forged dollar may well be de re represented as a real dollar, but it does not satisfy the conditions specifying which properties real dollars have. So it does not have the status of being a dollar.
A similar account can be given of ownership. The fact that someone is de re represented as owner, does not automatically entail that this person is the real owner. Ownership is presumably a status that depends upon satisfying some general condition for ownership. But satisfying that condition does not automatically entail that one is de re represented as the owner.
If this is correct, a de re representation to the effect that someone has a particular status is (with some exceptions) neither a necessary nor a sufficient condition for the entity to actually have this status. Whether or not an individual is de re represented as having a status is neither necessary nor sufficient for actually having that status. In a similar way, the fact that an individual has a certain status, does not entail that this individual is de re represented as having that status.
It is important to bear in mind that in modern bureaucratic states, the condition specified by the K-term may well include that one is the object of a de re representation. In order for a pair to be married for example, it is required that the spouses and a representative of the legal order all recognize that those particular individuals enter a marriage in a certain ceremony. This may well be a highly complicated condition. For the K-term in these cases must presumably also specify what is to count as a proper representative of the legal order, and this condition in turn may also involve being the object for various descriptive or de re representations.
The fact that an object has a certain institutional function however, is grounded in a representation of a less complicated structure. I suggest that it has the following structure.
(iii) X is F in C.
In this case, the X-term and the C-term have the same function as in Searle’s original account. The X-term refers de re. The F-term refers to the function X has in C, namely what powers X has or which role in the overarching institutional framework it performs. If X for example refers to a piece of paper, F may refer to a certain purchasing power. Whereas institutional statuses depend upon representations de dicto, it is difficult to see how institutional functions could depend upon anything but de re representations. This at any rate seems to be the case for the statuses under consideration; I cannot see how an entity could be used as a means of payment unless it was de re represented as having that function.
A dollar that is not de re recognized as a dollar is obviously set apart from other pieces of paper which neither are, nor are recognized as, dollars. But unless a paper is de re represented as having the function of being a means of payment, it cannot have that function.
If we apply this account to the problems described above, we can see how institutional functions can become detached from institutional statuses. Having a certain function depends upon being de re represented as having that function. But, clearly, the forged dollar is de re represented as having a certain monetary function, whereas the real dollar which is believed to be a forgery is not thus de re represented. So the forged dollar has a certain monetary function in the Albanian context that the real dollar lacks. A similar story could obviously be told about ownership.
So far I have presented an explanation of how two different kinds of institutional fact depend on representations. The first analysis gave an explanation of how the fact that an entity has an institutional status depends on representations. The second analysis gave an explanation of how the fact that an entity has an institutional function depends on representations. But there is a third kind of fact that needs to be explained. For, as we have seen, institutional statuses bear rights and obligations to institutional functions. They are deontically related to institutional functions. And we also need to explain how there can be facts about deontic relations between statuses and functions.
So we need a third type of representation, namely one which demonstrates how entities with a certain status have a right to an institutional power or an obligation to perform an institutional role. This kind of representation is more or less always de dicto. I shall suggest that it has the following form:
(iv) All objects which are S ought to have F in C.
What this says is that a certain status corresponds to a certain function in the sense that the legal system prescribes that all S ought to have F. 5 Thus, if someone has a certain status, then this person ought also to have a corresponding function. If F is a power, then (iv) stipulates that all objects which are S, have a right to F. If F is an institutional role, then (iv) stipulates that all objects which are S, have an obligation to F.
I have argued that facts about whether an entity has an institutional status depend upon the collective acceptance of representations of type (ii). And facts about whether an entity has an institutional function depend upon the collective acceptance of representations of type (iii). Facts about deontic relations between statuses and functions likewise depend on the collective acceptance of representations of type (iv). The collective acceptance of all three of our representations creates new kinds of institutional facts. In this sense they are constitutive rules. They do not regulate preexisting types of behavior.
It is important to note that if someone accepts (iv) and de re represents an object O as being an S, it is rational for this person to de re represent O as having a right or an obligation to F. So if a collective accepts (iv), and de re represents O as being S, it is rational for the collective to de re represent O as having a right or an obligation to F. And this is also what occurs in states governed by the rule of law.
Representations of kind (iv) are essential in order for the legal system to work. It describes which function ought to be assigned to which objects. But it does not actually assign this function. Functions are assigned not on the basis of whether entities have a status that deontically relates to the function, but on whether entities are represented as having such a status.
In order for society to work in accordance with the rule of law, law officers and bureaucrats must consistently judge de re which entities fulfill the conditions for having a certain status. So in order to examine whether someone owns a property, someone has to pass a judgment as to whether the individual in question satisfies the conditions of ownership for that particular piece of judgment. These judgments are however, not to be conflated with legal statutes. Rather, they are applications of legal statutes to particular cases.
Of course, judgments as to whether an individual satisfies the conditions for having a particular status may or not may be erroneous. And this is what generates the problematic cases discussed by Smith. Thus, if a piece of paper is erroneously predicated as a real dollar, it does not thereby have the status of a real dollar. But since we believe that dollars have a certain monetary function, and since we believe that this particular piece of paper is a dollar, we de re represent the paper as having a certain monetary function. And since a de re representation to the effect that an object has a certain monetary function is sufficient for it to actually have this function, it thereby acquires a monetary function.
The case of the real dollar believed to be a forgery functions in a similar way. There is no de re representation of that dollar as a dollar. So it will not be de re represented as having any monetary function. And consequently it will lack a monetary function.
5 A Possible Counter Objection
I have argued that the problematic cases introduced by Smith are representation-dependent. Smith however, argues that the mentioned cases are problematic because they involve a certain priority of contexts. And the fact that one context has priority over another context is not a representation dependent fact. Thus, for example, in the case of the dollars, some pieces of paper count as forged dollars in the USA but as real dollars in Albania. But in this case the US context has priority over the Albanian context. Smith goes on to claim that the US context has priority over the Albanian context in the sense that in the context of the US banking system pieces of paper are revealed as what they are – to wit dollars or forged dollars – whereas they merely count as one thing or another in the Albanian banking system (Smith 2003, p. 293–294).
Can this be used as a counter objection to the notion that these institutional facts are representation-dependent? It seems to me that this is not the case. For whether a context is prior to another context or not, also seems to be a representation-dependent fact. Forged dollars are not legal entities according to the Albanian legal system. There is no law that stipulates that forged dollar is a legal monetary entity in Albania. Nor is there such a law in any other part of the world. And if by any chance there were a law that stipulated that papers looking like real dollars but fabricated in Albania were to be the legal monetary entity in Albania, then these papers would not be forgeries, but rather some kind of Albanian dollars.
What the Albanian example shows is that in some contexts in Albania, various persons use what they believe are real dollars according to the American legal system as monetary entities. This belief is however, false. But this fact is not stranger than the fact that some people in America might be in the possession of forged dollars which they use as means of payment. So the American context has priority over the Albanian context because everyone, Albanians and Americans alike, represent it as such.
Smith’s discussion of the problem of different owners leads him to ask us to consider the problem of ownership of property as it arose in Germany after the unification. In that situation several different persons made competing claims for the same piece of property. Thus, for example, three or more persons may claim ownership of the same house. One person perhaps owned it in the twenties or early thirties but was forced to leave it and flee from the country when the Nazis took power. The house was then expropriated by the Nazi government and sold to a second person. She however, left for West Germany when Germany was partitioned. Under communist rule, a third person owned the house (Smith 2003, p. 295–296).
In this case there are three different persons (or their descendants) that count as the owner. But they count as the owner in different legal systems: One person counts as the owner in the context of the legal system of the Weimar Republic, one as the owner in the context of the legal system of Nazi Germany and one as the owner in the context of the legal system of East Germany. But who then is the owner?
Smith points out that the representationalist can argue that the owner is the one who owns the property in the overarching context provided by the current legal system (Smith 2003, p. 296). Smith however argues that there are epistemic problems with this argument. If I understand him correctly, the problem is that the present legal system may often enough be unable to determine who the real owner is. The issues may be legally irresolvable.
Now, it seems to me that this objection does not threaten the account I defend here. For as I have stressed, whether or not an individual entity actually satisfies the condition for having a status, is a separate question from whether this entity is de re represented as satisfying this condition. So whether or not a particular entity has a certain status may well be epistemologically indeterminate. But it may well be that there is a determinate fact of the matter whether the entity satisfies a certain condition or not. And this fact may well be dependent upon a representation de dicto. So the fact that the issue is epistemologically indeterminate does not preclude that the fact is representation-dependent.
This is not to say that Smith is wrong in arguing that there are other types of institutional facts which are not representation dependent. I think that other arguments of his are more promising. But his failure to show that these particular cases are representation-dependent gives us some insights into the variety of the representations that legal facts depend upon.
Andersson, Åsa (2007): Power and Social Ontology. Malmö: Bokbox.
Celano, Bruno (1999): “Collective Intentionality, Self-Referentiality and False Beliefs: Some Issues Concerning Institutional Facts”. In: Analyse & Kritik 20, p. 143–158.
Lagerspetz, Eerik (2006): “Institutional Facts, Performativity and False Beliefs”. In: Cognitive Systems Research 7, p. 298–306. [Crossref]
Moural, Josef (2002): “Searle’s Theory of Institutional Facts: A Program of Critical Revision”. In: Günther Grevendorf and Georg Meggle (Eds.): Speech Acts, Mind and Social Reality. Discussions with John R. Searle. Dordrecht: Kluwer, p. 271–286.
Roversi, Corrado (2010): “Constitutive Rules in Context”. In: Archiv für Rechts- und Sozialphilosophie 96, p. 223–238.
Searle, John R. (1995): The Construction of Social Reality. New York: The Free Press.
Searle, John R. (2001): “Reply to Barry Smith”. In: Philosophical Explorations 4, p. 75–77.
Searle, John R. (2003): “Reply to Barry Smith”. In: American Journal of Economics and Sociology 62, p. 299–309.
Searle, John R. (2008): “Social Ontology and Political Power”. In: Barry Smith, David M. Mark, Isaac Ehrlich (Eds.): The Mystery of Capital and the Construction of Social Reality. Chicago and La Salle: Open Court, p. 19–34.
Searle, John R. (2010): Making the Social World. The Structure of Human Civilization. Oxford: Oxford University Press.
Smith, Barry (1992): “An Essay on Material Necessity”. In: Canadian Journal of Philosophy supplementary volume 18, p. 301–322.
Smith, Barry (2001): “The Chinese Rune Argument”. In: Philosophical Explorations 4, p. 70–75.
Smith, Barry (2003): “The Ontology of Social Reality”. In: American Journal of Economics and Sociology 62, p. 285–299. [Crossref]
Thomasson, Amie (2003): “Foundations for a Social Ontology”. In: Protosociology 18–19, p. 269–290.
Zaibert, Leo (2004): “Toward Meta-Politics”. In: The Quarterly Journal of Austrian Economics 7, p. 113–128.