The creation of institutional reality, special theory of relativity, and mere Cambridge change

Abstract

Saying so can make it so, J. L. Austin taught us long ago. Famously, John Searle has developed this Austinian insight in an account of the construction of institutional reality. Searle maintains that so-called Status Function Declarations, allegedly having a “double direction of fit” (i.e. a world-to-word and a word-to-world direction of fit), synchronically create worldly institutional facts, corresponding to the propositional content of the declarations. I argue that Searle’s account of the making of institutional reality is in tension with the special theory of relativity—irrespective of whether the account is interpreted as involving causal generation or non-causal grounding of worldly institutional facts—and should be replaced by a more modest theory which interprets the results of Status Function Declarations in terms of mere Cambridge change and institutional truth. I end the paper by indicating the import of this more modest theory for theorizing about the causal potency of institutional phenomena generated by declarations.

1 Introduction

John Searle famously maintains that all of institutional reality is created by “Status Function Declarations” (Searle 2010: pp. 13, 93, 99). Status Function Declarations fall into the category “extralinguistic declarations”, a category of speech acts that Searle distinguishes from mere “linguistic declarations” (Searle 1989: pp. 549, 554–555, 2010: pp. 110–114). Linguistic declarations create speech acts (such as promises, statements and orders), but extralinguistic declarations create extralinguistic facts and entities in the world. Extralinguistic declarations are, conceptually, of two kinds: supernatural declarations and Status Function Declarations (Searle 1975/1979: pp. 18, 1989: pp. 549–550, 2010: p. 100). Although we cannot, as human beings, successfully perform supernatural declarations, we can, according to Searle, change the world and create institutional facts and objects through Status Function Declarations.

In Searle’s own words:

Naively the best way to think of the distinction [between linguistic and extralinguistic declarations] is this: A declaration is a speech act whose point is to create a new fact corresponding to the propositional content. Sometimes those new facts are themselves speech acts such as promises, statements, orders, etc. These I am calling linguistic declarations. Sometimes the new facts are not further speech acts, but wars, marriages, adjournments, light, property transfers, etc. These I am calling extralinguistic declarations. […] When God says, “Let there be light!”, that I take it is a declaration. […] It makes it the case by fiat that light exists. […] We ordinary humans do not have the ability to perform supernatural declarations, but we do have a quasi-magical power nonetheless of bringing about changes in the world through our utterances. (Searle 1989: p. 549) We can create boundaries, kings, and corporations by saying something equivalent to “Let this be a boundary!” “Let the oldest son be the king!” “Let there be a corporation!” […] in the nonlinguistic Status Function Declaration, we do more than represent: we create. (Searle 2010: pp. 100, 114)

Moreover, according to Searle, Status Function Declarations generate their worldly changes synchronically—that is, simultaneously with the performance or application of the relevant Status Function Declaration (Searle 1989: pp. 556–557, 2010: p. 99). This should be contrasted with how directives and commissives, such as orders and promises, generate their worldly changes: such speech acts, if successful, have their characteristic worldly effects at later points in time (e.g. Searle 1975/1979: pp. 14, 19).

Searle is sometimes criticized on grounds that it is doubtful that all of institutional reality is created through declarations: much of institutional reality seems rather to evolve gradually and unintentionally, without the performance and acceptance of declarations (see e.g. Blackburn 2010; Epstein 2014; Lawson 2016). In this paper, I ignore this kind of criticism and focus instead on the kinds of example where Status Function Declarations intuitively do play an essential role.¹ I argue that it is deeply problematic to speak of Status Function Declarations, such as verdicts, appointments, gradings, and incorporations, as literally “changing the world”—that is, as creating worldly “extralinguistic facts” (typically of the form, a’s having institutional property or status function Y) that “correspond” to the propositional content of the declarations, as claimed by Searle. The Searlean view is intuitively problematic already within a commonsensical Newtonian framework—what could the relevant mechanism be (cf. Effingham 2010: p. 253)?—but it is decisively troublesome within a relativistic setting. Famously, instantaneous causation between spatially separated events is precluded in relativistic (Minkowskian) spacetime, meaning that a causal interpretation of Status Function Declarations (see e.g. Searle 1983: pp. 171–173) is ruled out; and synchronic grounding of an institutional fact by a Status Function Declaration (cf. Searle 1989, 2010: p. 4; Schaffer forthcoming) will, in a suitable reference frame in relativistic spacetime, involve backward grounding of the institutional fact, meaning that a grounding account of Status Function Declarations is in trouble too. To avoid such problems, it is preferable, I shall argue, to characterize the “results” of Status Function Declarations in terms of so-called mere Cambridge changes and institutional truths. This more modest conceptualization avoids giving a false picture of what happens when we perform Status Function Declarations and allows for simultaneous “generation” of institutional phenomena across reference frames, even within a relativistic framework. The conceptualization also has ramifications for the way we theorize the causal standing of the putative institutional phenomena.

The structure of the paper is as follows: I begin by describing Searle’s theories of declarations and Status Function Declarations (Sect. 2). (Readers well acquainted with these theories may skip to Sect. 3.) I then criticize Searle’s theory of Status Function Declarations—first, when construed as involving causation, subsequently, when interpreted in terms of grounding—drawing on the special theory of relativity (STR), (Sect. 3). Following this, I develop an alternative account in which the notions of mere Cambridge change and institutional truth play a key role (Sect. 4). I conclude with some observations about the implications of the alternative account (Sect. 5).

2 Declarations and status function declarations

Searle originally stated his general theory of declarations in his (1975/1979), where a taxonomy of illocutionary acts is developed.² He explains:

It is the defining characteristic of this class [of illocutionary acts] that the successful performance of one of its members brings about the correspondence between the propositional content and reality, successful performance guarantees that the propositional content corresponds to the world. […] Declarations bring about some alteration in the status or condition of the referred to object or objects solely in virtue of the fact that the declaration has been successfully performed. (Searle 1975/1979: pp. 16–17)

He symbolizes the structure of declarations thus (ibid.: 19):

$$ D\!\uparrow\!\downarrow\!\emptyset\left( p \right) $$

“D” stands for declarational illocutionary force or point; the arrows indicate that declarations have both words-to-world and world-to-words direction of fit (as explained below); the null symbol indicates that there is no sincerity condition (this seems intuitively wrong, but for our purposes, this aspect can be ignored³); and “p” stands for the propositional content of the declaration.

Searle explains what is involved in the idea that declarations have a “double direction of fit” by contrasting declarations with other types of illocutionary act. Thus, in Making the Social World (2010) he writes:

Some speech acts [e.g. statements] function by purporting to represent how things are in the world. […] I think of these speech acts as hovering over the world and pointing down to it, as fitting or as failing to fit the world, as having what I call the word-to-world direction of fit. I represent these with the downwards arrow ↓. […]

But there are lots of speech acts that are not in the business of trying to tell us how things are in the world. They are trying to change the world to match the content of the speech act. […] The order is aimed at causing obedience; the promise is aimed at causing fulfillment. […] I say of these cases that they have the world-to-word direction of fit. […] I represent the upward or world-to-word direction of fit with an upward arrow ↑. […]

There is a fascinating class of speech acts that combine the word-to-world ↓ and the world-to-word ↑ direction of fit, which have both directions of fit simultaneously in a single speech act ↑↓. These are cases where we change reality to match the propositional content of the speech act and thus achieve world-to-word direction of fit. But, and this is the amazing part, we succeed in doing so because we represent the reality as being so changed. More than three decades ago, I baptized these as “Declarations.” They change the world by declaring that a state of affairs exists and thus bringing that state of affairs into existence. (Searle 2010: pp. 11–12)

As we saw in Sect. 1, some declarations bring speech acts of other kinds into existence.⁴ Others bring extralinguistic phenomena into existence: these phenomena might (at least logically) be physical (created by “supernatural declarations”) or institutional (created by “Status Function Declarations”). In the rest of the paper, I focus on Status Function Declarations.

In his (2010: p. 93), Searle maintains that the “general form” of Status Function Declarations is:

We (or I) make it the case by Declaration that the Y status function exists.

A status function is a function (i.e. “a cause that serves a purpose”, Searle 2006: p. 17, 2010: p. 59) with two special but interrelated features:

First, […] they require collective intentionality, both for their initial creation and for their continued existence. And second, they are functions that a person or other entity has, not in virtue of physical structure, […] but in virtue of collective imposition and recognition of a status. (Searle 2010: p. 59, see also 1995: pp. 40–51, 2010: pp. 7, 102–104)

Status functions, then, are additional to the physical powers or abilities of objects, and imposed on objects through Status Function Declarations. Their continued existence requires their ongoing acceptance or recognition by a plurality of individuals in the society in question (Searle does not specify how large the fraction of accepting individuals needs to be).

By declaring a status function into existence, we create an institutional fact or state of affairs (e.g. Searle 2010: p. 93). Crucially, for Searle, a fact or state of affairs is not merely a true proposition or statement (Searle 1995: pp. 205–206, 220). He rejects such a Fregean view of facts on the ground that “facts function causally in a way that true statements do not” (1995: p. 205). Facts are rather worldly things to which true statements correspond (Searle 1995: p. 202); they are entities that make statements true:

… the fact stated needs to be distinguished from the statement of it […] the whole point of having the notion of “fact” is to have a notion for that which stands outside the statement but which makes it true, or in virtue of which it is true, if it is true. (Searle 1995: pp. 2, 211, emphasis in original; see also Austin 1950/1979: pp. 121–124)

In some places Searle is content to describe such facts loosely as “conditions in the world” (Searle 1995: p. 211). In other places, he seems to accept a more specific and traditional understanding of facts or states of affairs (cf. e.g. Armstrong 1997): facts are “entities in the world that have a propositional structure” (e.g. Searle 2014: p. 19); that is, a fact is (I read Searle as saying) a structured entity consisting of an object with a certain feature or property (or a constellation of objects standing in certain relations to each other). On yet other occasions, Searle uses the term “fact” to denote various things, such as events, actions, speech acts, properties, and objects (e.g. Searle 1989: passim, 2010: passim, 2010: passim). Although these conceptions of factuality differ, it is clear that Searle wants to use “fact” to pick out a worldly—and potentially causal—entity (or condition or structured complex), as opposed to an abstract linguistic entity, such as a true proposition.

Importantly, institutional facts are to be contrasted with brute facts (Searle 1995: pp. 2, 27). Brute facts, such as Earth’s being 93 million miles from the sun, exist independently of any human institution.⁵ Institutional facts, such as Donald Trump’s being president of the U.S., require institutions for their existence—where institutions are systems of “constitutive rules” (1969: p. 51, 1995: p. 28), or as Searle prefers to put it in his (2010: pp. 10, 13): systems of Status Function Declarations.⁶ The essential point is that, “all institutional facts are created by the same logical operation: the creation of reality by representing it as existing” (Searle 2010: p. 93).

The overall idea concerning institutional reality, then, is that statements about institutional reality are made true by institutional facts (Searle 1995: p. 221); and institutional facts are brought into existence by Status Function Declarations (Searle 2010: p. 13).⁷

Before I move on I should point out the following. For Searle, there is a strong connection between status functions and so-called deontic powers, i.e., rights, duties, permissions, obligations, etc. (1995: pp. 100–101, 2010: pp. 8–9). Searle sometimes puts this by saying that “status functions carry deontic powers” (2010: p. 8). For example, a person who acquires the status function of being president of the Unites States thereby acquires certain rights and duties. But though there is a strong connection between status functions and deontic powers, they should not, I think, be identified on Searle’s scheme of things. For example, the status function of being money may be imposed on certain pieces of paper, but the associated right to use those pieces of paper as payment belongs to the owner of the money, not the paper items (1995: pp. 95–96, 101, 2010: pp. 101–102). Thus, status functions do not reduce to deontic powers, since status functions and their associated deontic powers may very well be had by distinct entities.

It is also worth stressing that, for Searle, institutional facts, status functions and institutional objects do not exist in people’s minds (pace e.g. Johansson 2003: p. 247). Institutional phenomena somehow depend on mental phenomena of the kind involved in collective acceptance, or recognition, of Status Function Declarations, but they are not identical with those phenomena. As Searle puts it: “we need to introduce a class of phenomena [money, property, government, etc.] that are not actually located in our minds but are dependent on our attitudes.” (2010: p. 17, my emphasis) (Admittedly, there is a puzzling anti-realist strand in Searle’s writing about institutional reality which is somewhat hard to square with the views and formulations summarized and quoted above. For example, at various places he says that institutional facts are “ontologically subjective” (e.g. 2010: p. 18). I will return to this seemingly irrealist aspect of Searle’s theory at the end of Sect. 4.)

Finally, it should be noted that spelled-out declarations are typically present-tensed, and often they contain indexicals like “now”.⁸ According to Searle, the present tense—or what he sometimes calls the “dramatic present” or the “present present” (1989: 556)—indicates that the creation of the represented fact, i.e. the event of creation, is simultaneous with the speech act:

This tense is used to mark events which are, so to speak, to be construed as instantaneous with the utterance. […] In these cases, the sentence describes an event that is simultaneous with its utterance […] Now, because the performative utterance is both self-referential and executive, the present present is ideally suited to it. “I promise to come and see you” [or “We make it the case by Declaration that object X now has the status function Y in C”] marks an event [the making of a promise, the creation of an institutional fact] which is right then and there, simultaneous with the utterance, because the event is achieved by way of making the utterance. (Searle 1989: pp. 556–557, my emphasis)

This synchronicity thesis is arguably entailed by the “double direction of fit” idea. If the worldly institutional fact were created at some point after the completion of the declaration, we would apparently have a case of world-to-word direction of fit, ↑; but according to Searle, in cases of declarations, we have ↑↓.

I will turn now to examine the consequences of taking seriously the idea that we can synchronically create extralinguistic, institutional facts through Status Function Declarations, and thereby literally change the world.⁹

3 Bringing reality into correspondence with propositional content through status function declarations? The lessons from STR

Obviously, if someone utters something, the world thereby changes in the trivial sense that the very act of uttering is a process consisting of physical changes (the lips move, and so on). Clearly, however, this is not the kind of change Searle is talking about when he says that we can change the world by making Status Function Declarations. There is nothing “amazing” (2010: p. 12), “quasi-magical” (1989: p. 549) or “god-like” (1983: p. 176) about such changes. Instead the idea is, as we have seen, that the world changes because an institutional fact (which, remember, is non-identical with a true proposition and with people’s mental states) begins to exist in the world when a Status Function Declaration is made—a fact which corresponds to the propositional content of the speech act and which serves as a truth-maker of the declaration, yielding the success both of the declaration’s world-to-word ↑ direction of fit and its word-to-world ↓ direction of fit.¹⁰ Thus, Searle appears to be asserting that beyond the physical and mental changes involved in lip movements and shared recognition, something more is created. It is the creation of this additional feature or thing—specifically, a “status function”, such as being money, or an “institutional object”, such as a corporation—which constitutes the relevant change or event in the world, and which allows us to refer, quite literally, to the creation of an institutional fact.

Let us take Searle at his word, and ask: Can the claim that institutional facts are created, or brought into existence, in the way he describes be reconciled with our best scientific theories of how the macroscopic world works¹¹—in particular, is it compatible with the tenets of the special theory of relativity (STR)? I shall argue that if the relevant change or creation is not local, i.e. does not have the same spacetime location as the declaration itself, the answer is No.¹²

Consider the marking of exam papers. When an authorized examiner assigns a grade to a student’s script (and thereby to the student), in accordance with the marking scheme, say by signing a certain physical document or by entering information into a computer system, that act can be interpreted as an extra-linguistic declaration of the following form:

I declare: NN is hereby assigned grade Y in subject S.

The examiner’s declaration certainly makes it true that the student has acquired the relevant grade in the relevant subject. And Searle would in all likelihood regard the grading as an imposition of a “status function”, since the grade (we may suppose) gives the student new deontic powers, such as being entitled to enroll on an advanced course in the same subject (cf. Searle’s discussions of convicted criminals, 1975/1979: p. 20, 1995: p. 50).¹³

Suppose now that the acquisition of the grade involves a genuine change in the world (beyond the agreed physical and mental changes): an instance of an institutional feature or property, grade Y (a “status function”), is created. It is instantiated by the relevant student, a, and in this way a worldly institutional fact—a’s having grade Y—corresponding to the propositional content of the declaration is created. On this construal, it is natural to suppose that the change involving a is an effect of the declaration, and that the grading and acquisition of the grade are a cause-and-effect pair (I will later discuss the possibility that they are related by an ontological dependence or grounding relation). Indeed, in his (1983) Searle explicitly defends a causal interpretation of declarations:

…what is [a person] trying to do when he makes a declaration? He is trying to cause something to be the case by representing it as being the case. More precisely, he is trying to cause a change in the world […] in the case of directives, commissives, and declarations, the utterance, if satisfied, will in various ways function causally in the production of the state of affairs it represents. (Searle 1983: pp. 172–173, emphasis in original)¹⁴

However, since the acquisition of the grade occurs simultaneously with the grading, or its official completion, the relevant relata must be related by simultaneous causation if they are causally related.¹⁵ Worse, since the student and the act of grading are likely to be located in different places, the relevant causation will involve events that are both simultaneous and separated in space.¹⁶ But that is ruled out by STR: it says that no causal process can propagate outside the future light cone of an originating event—that is, no causal process can propagate faster than the speed of light, which is a constant and finite magnitude (299 792 458 m/s) across all reference frames (e.g. Taylor and Wheeler 1992: pp. 171, 181–182).¹⁷ Thus, if Status Function Declarations are supposed to bring about their simultaneous worldly changes by causal processes, the theory is straightforwardly at odds with STR (assuming non-locality).¹⁸

A causal interpretation of synchronic declarational creation of institutional facts thus seems doomed. And, in his (1989), Searle appears to have changed his mind about the generative means, although he does not explain why. He now says:

In order intentionally to produce changes in the world through our actions, normally our bodily movements have to set off a chain of ordinary physical causation. […] But there is an important class of actions where intention, bodily movement and desired effect are not related by physical causation in this way. If somebody says, “The meeting is adjourned,” “I pronounce you husband and wife,” “War is declared,” or “You’re fired,” he may succeed in changing the world in the ways specified in these utterances just by performing the relevant speech acts. […] the difference between pounding a nail and adjourning a meeting is that in the case of adjourning the meeting the intention to perform the action, as manifested in the appropriate bodily movement (in this case the appropriate utterance) performed by a person duly authorized, and recognized by the audience, is constitutive of bringing about the desired change. (Searle 1989: pp. 547–548, my emphasis)

Searle here seems to be suggesting that declarations—or the intentions behind them, and the collective recognition of them¹⁹—and their associated changes are related, not by causation, but by a relation of constitution. However, since he also denies, as we saw in Sect. 2, that institutional phenomena are located in people’s minds (2010: p. 17), this “constitution relation” cannot be conceived of in terms of identity. Rather, it seems that Searle is here alluding to a non-causal, asymmetric ontological dependence relation—in contemporary terms, to a grounding relation (see e.g. Schaffer 2009, forthcoming)—obtaining between non-identical phenomena (x grounds y is simply the converse of y depends ontologically on x).²⁰ An account in terms of grounding is, at least prima facie, more appropriate, since, in contrast to causation, grounding is standardly taken to be a paradigmatically synchronic relation (see e.g. Bernstein 2016).

Our question, then, is whether an account of Searlean declarational generation of institutional facts framed in terms of asymmetric ontological dependence or grounding is compatible with STR. I shall argue that the answer to this question is in the negative.

Apart from the fact that it is unclear what the mechanisms, if any, of grounding are supposed to be, a grounding account²¹ of declarational generation of institutional facts still faces the issue of simultaneity and spatial separateness: the relevant Status Function Declaration and the change in the world (the event of creation) are supposed to be simultaneous, and if the entity which acquires the status function is located at some place differing from the place of the Status Function Declaration, then, trivially, the declaration and the change are spatially separated. Thus, this kind of case is in an important respect different from the paradigmatic examples of grounding discussed in the grounding literature—wholes grounded in their parts, mental states grounded in brain states, tropes grounded in substances, sets grounded in their members—where the grounded phenomena are, typically, taken to be co-located with their grounds. Now, according to STR, it is a relative matter whether or not events which are spatially separated and exist in each other’s “absolute elsewhere”—i.e. outside the lightcone of the other event in question—are simultaneous: events that are simultaneous in one reference frame are not simultaneous in another frame if the latter is moving at high speed relative to the former (e.g. Taylor and Wheeler 1992: pp. 62–63, 181). This is a problem for a grounding account of Searlean generation of worldly institutional facts by Status Function Declarations, because unless the account treats the relevant grounding as something that is in some reference frames directed backward in time—which is highly counterintuitive and violates Searle’s simultaneity requirement²²—it will have to insist that the relevant events are simultaneous in all reference frames, in violation of STR.

Consider Fig. 1. In it, a student, a, and an examiner, b, who are spatially separated but (approximately) stationary relative to each other, are depicted. During the time interval shown they agree on the simultaneity of events: they slice up spacetime into the same set of simultaneity planes (or times or hyperplanes of spacetime). The grading, e, and the acquisition of the grade, e´, happen at the same time according to them, namely t₁.²³ However, in the diagram, a third person, c, moving uniformly at a very high speed relative to a and b, towards them along their x-axis, is also depicted (I consider a futuristic sci-fi scenario with the purpose of making a philosophical point). Person c’s planes of simultaneity are at an angle to a’s and b’s planes of simultaneity. In particular, according to c, event e´ occurs before event e! But if e´ depends on e, then e’s grounding or generation of e´ is directed backward in time in c’s reference frame. Such a scenario, involving backward grounding of a real, worldly institutional property, will, I take it, be generally rejected by social ontologists²⁴: how could a person or object, relative to a certain frame of reference, acquire a worldly institutional property before the relevant declaration has occurred? Such backward grounding seems almost as weird as backward causation. Certainly, scenarios like this will be rejected by anyone who, like Searle, is already committed to the simultaneity requirement.²⁵

Fig. 1

Events e and e´ do not occur simultaneously according to observer c. Specifically, e´ occurs before e in c’s frame of reference. (y- and z-dimensions, and light-cone structures, are not displayed in the diagram)

To avoid (relative) backward (but also forward, see notes 22 and 25) grounding of worldly institutional facts, a defender of an ontological dependence account will have to insist, against STR, that e and e´ are simultaneous in all reference frames (i.e. relative to all observers, including a, b and c). A natural way of defending such a view would be to say that we inhabit a Newtonian spacetime with absolute simultaneity relations, and maintain (in line with the reasoning of Prior 1996/1998; Craig 2000, Ch. 5; Tooley 1997, Ch. 11; Craig and Smith 2008) that the relative simultaneity planes of STR are mere artefacts of certain contingent, operational definitions of temporal and spatial intervals, and account for the classical Michelson-Morley and the Kennedy-Thorndike experiments (which support the notion that the speed of light is a constant) in terms of absolute clock retardation and absolute shrinking of measuring rods, in the manner pioneered by Hendrik Lorentz (see his papers in Einstein et al. 1952).

Such a view is very unattractive. As is well known, to date no satisfactory explanation of why physical processes slow down (absolutely) and material objects contract (absolutely) when moving relative to the absolute rest frame, assuming such a frame exists, has been provided. Without such an explanation, the Newtonian/Lorentzian theory looks seriously ad hoc. By contrast, the purely relative time dilations and length contractions of STR follow from the postulates of the theory (see Sklar 1974, Ch. IV, B for discussion of the rival theories of Lorentz and Einstein).

A further problem, and one that applies specifically to the issue of institutional reality, is as follows. Defenders of a grounding account who insist that events such as e (i.e. declarations made by humans on earth) and e´ (the changes created by such declarations) are not only relatively but absolutely simultaneous are apparently committed to the notion that earthlings are absolutely at rest relative to absolute space (at least approximately, or for brief periods of time, so that Earth’s rotation around its axis can be ignored), although there is no way of empirically confirming that such a situation obtains, due to absolute time dilations and length contractions.²⁶ Earth’s being at rest relative to absolute space would seem to be an extraordinary coincidence in need of explanation and theoretical support. It cannot be simply assumed.²⁷

In short, the Newtonian understanding of space and time is both archaic and introduces added explanatory burdens and complications. With this in mind, I will assume from here on that the absolute simultaneity option is not on the table for Searleans. Moreover, the issue of the relevant generative mechanism would be left unresolved within such a framework.²⁸

The lesson of the discussion in this section, then, is not that STR is false. Nor, in my view, should we conclude that the relevant results of Status Function Declarations are effects or events that happen absolutely after the declarations, i.e. within the future light cone of the relevant declarations. For while that view would be compatible with STR, it would violate Searle’s simultaneity requirement—an intuitively plausible thesis which I think should be retained, in some form or another, if possible.^{29, 30} Thus, I think the most plausible conclusion of the foregoing discussion is that Status Function Declarations do not genuinely change anything in the world (the agreed physical and mental changes apart): such declarations bring nothing “extralinguistic” into existence; they do not create institutional properties, or institutional objects; they do not create worldly institutional facts. The acquisition of a grade by a student, for example, is not a genuine event, taking place at a certain location; it does not involve a genuine change in a subject.

4 Mere Cambridge change and institutional truths

If the acquisition of a grade is not a genuine event, what is it? Students do, after all, acquire grades—that can hardly be denied. I shall argue that acquisitions of (putative) institutional features or properties (“status functions”) should be thought of as so-called “mere Cambridge events” (cf. Kim 1974: p. 48). Such events involve “mere Cambridge change”: a monadic predicate becomes applicable (or ceases to be applicable) to an object in virtue of changes that happen elsewhere. The “creation” of full institutional objects, I shall argue, should be understood in similar terms: existential, institutional statements or propositions become true (if present-tensed), but the truth-makers for such truths do not involve ontic institutional objects existing in the world.

The term “mere Cambridge change” was coined by Peter Geach in his book God and the Soul (1969), where he discussed the following Cambridge criterion of change, based on the work of Bertrand Russell and Ellis McTaggart (e.g. Russell 1903: p. 469; McTaggart 1927: p. 14):

The thing called “x” has changed if we have “F(x) at time t” true and “F(x) at time t¹” false, for some interpretation of “F”, “t”, and “t¹”. (Geach 1969: pp. 71–72)

Geach pointed out that this criterion gives the odd result that a person a “changes” when she comes to be shorter than another person b in virtue of the latter’s growth. For a monadic predicate, “is shorter than b”, becomes true of a having not been true of a at an earlier time. This is not a genuine or real change in the person a, according to Geach: it is a mere Cambridge change. The predicate “is shorter than b” comes to apply to a solely as a result of a genuine change that happens to someone else, b, somewhere else (see also Kim 1974; Shoemaker 1980/1997; Mellor 1998: pp. 87–88).³¹

This diagnosis applies quite straightforwardly, I think, to the cases where an institutional property (or “status function”) is imposed on, or ascribed to, a pre-existing object (cf. Searle 2010: p. 99). Consider our example of grading again. An examiner declares, at a certain place, and at a certain time (relative to, say, his own frame of reference), that student a has acquired grade Y in subject S. The examiner’s act of writing or typing is a genuine change in the world, with a location. Given that the examiner is duly authorized,³² and as long as the relevant rules and requirements regarding grading have been followed, an institutional predicate³³ of the form “has acquired grade Y in subject S” becomes true of student a—and it does so irrespective of a’s whereabouts. Since this predicate is now true of the student (in the examiner’s reference frame), it can be truly said that the student has, from this time onwards, a new institutional “status”. She is now entitled to enroll on a new course, etc. It can even be said that the student has acquired a new institutional property, as long as “property” is understood merely in the weak, abundant sense discussed by philosophers such as Lewis (1986a: pp. 59–60), Mellor (1995: pp. 157–159, 185), Bird (2007: pp. 9–12), and Heil (2012: pp. 23–24), according to which, roughly, an object x has a “property” F if predicate “F” applies to x.³⁴ What the student has not acquired, however, is a property in a strong, sparse (or ontic) sense (ibid.): nothing actually happened to the student or in the student’s whereabouts. In this sense, the acquisition of the grade is not a genuine event, but a mere Cambridge event involving mere Cambridge change.

If “acquisition” of a grade is a mere Cambridge event, the difficulties with STR vanish. No causal process propagates from the declaration to the student at a speed faster than light; no sparse, ontic institutional property of the student, located within the absolute elsewhere of the declaration, gets grounded by the declaration.³⁵ Rather, a mere predicate becomes true of the student, as a result of the declaration. The predicate, moreover, can be taken to become true of the student at the time of the declaration—i.e. simultaneously with the declaration—in all reference frames without any need to postulate absolute simultaneity relations. In the reference frame of the student and the examiner, the predicate “has grade Y” becomes true of the student at time t₁ (although the student may not be aware of this at t₁) (see Fig. 1); but in the reference frame of observer c, the predicate can be taken to become true of the student at the simultaneity plane indicated by the drawn x´-axis in Fig. 1. This scenario does not involve the paradoxical situation that there is a single, genuine event e´ (a’s acquisition of the grade Y) that succeeds in being located at two distinct places in spacetime (see Fig. 2)—which, of course, would be impossible.

Fig. 2

One and the same event, e´, is depicted as being located twice over in spacetime

Rather, the claim is that no genuine event e´ occurs at all. This more modest interpretation simply involves the notion that “the acquisition of the grade Y” is to be understood in terms of a predication which is relativized, or indexed, to a reference frame with its own planes of simultaneity or times (see Fig. 3).

Fig. 3

The predicate “has grade Y” becomes true of a, because of b’s declaration e, at distinct simultaneity planes of spacetime, according to a and b, and c, respectively

Similar considerations apply, I submit, to other impositions of putative status functions to pre-existing objects through declarations: to someone’s appointment as president, promotion to a professorship, or sentencing to a term in prison.³⁶

This mere-Cambridge-change account also enables us to make sense of the special cases where declarations are directed towards the past from the point of view of the one who is performing the declaration.³⁷ Consider grading yet again. In the Swedish university system (and in the Dutch, I am told), grades are often assigned retrospectively: a student completes an assignment on a certain date t, but a busy professor grades the assignment at a later date t´ and in so doing gives the student an official grade which is valid from the earlier time t onwards. If something was really created at t by the declaration at t´, we would have a case of backward causation or grounding—but this time, within the past light cone of the declaration (i.e. within the so-called absolute past of the declaration, entailing that the acquisition of the grade occurs before the declaration in all reference frames). By contrast, on the mere-Cambridge-change view a mere predicate of the form “has grade Y” becomes true of the student, in virtue of the declaration at t´. The student does not genuinely change as a result of the declaration. Interestingly, though, on the assumption that the B-theory of time is correct—according to which all events (i.e. events deemed “future”, “present” or “past” from the point of view of someone existing in spacetime) are equally real³⁸—the present-tensed predicate “has grade Y” is true of the student already at time t, owing to the existence of the later, past-directed declaration at t´. However, at t the declaration is epistemologically inaccessible to everyone, so at that time no one acts on its “consequences”; but the institutional predicate is nevertheless true of the student at t, on this worldview.³⁹ Thus, the genuine change or event (i.e. the making of the relevant past-directed declaration) occurs at time t´, and the “result” of that event is a mere Cambridge change concerning a predicate which, on the B-theory of time, is true of the student at t and onwards, but not at times before t. After t´, of course, the declaration gradually becomes epistemologically accessible to the relevant society at large, and people, including the student, start behaving in accordance with it (cf. Sect. 5 below).

So far, I have been discussing the imposition of an institutional property or “status function” to a pre-existing physical object. I have not yet addressed the alleged creation of institutional objects such as corporations (i.e. instances of so-called free-standing Y-terms) through Status Function Declarations.⁴⁰ About this second variety of institutional creation, Searle writes:

In this case we seem to have created a remarkably potent object, a limited liability corporation, so to speak out of thin air. No pre-existing object was operated on to turn it into a corporation. Rather, we simply made it the case by fiat, by Declaration, that the corporation exists. (Searle 2010: p. 98)⁴¹

I think the idea that an object is literally created in this way needs to be avoided. It is better to adapt the mere-Cambridge-change account of institutional properties and apply it to objects. On this account, a declaration to the effect that a corporation exists is made, and because of this declaration it becomes true to say, in a tensed language and within a suitable reference frame, “A corporation, founded in such and such a way, exists”.⁴² The truth-maker for such an existential assertion, however, should not be taken to be a new, ontic, institutional object that is somehow brought into existence in the world (perhaps, at its “institutional location”, cf. Hindriks 2013: p. 418) simultaneously with the declaration. That would lead to difficulties of the kind discussed above. Rather, the truth-maker (or truth-makers) should be assumed to consist simply of the declaration itself, together with representations of the relevant legal regulation.⁴³

In effect, I am suggesting that the institutional object case is in an important respect similar to many of the so-called deflationary cases discussed in the truth-maker literature (e.g. Armstrong 2004: p. 33). These involve existential statements which are made true by entities ostensibly quite different from the entity claimed to exist. For example, Hugh Mellor (2012: pp. 98–99) argues that true existential propositions about rainbows, such as <there is a rainbow east of car c driving north>, are not made true by peculiar rainbow-entities with remarkable properties (in the example in question, a rainbow that keeps pace with c as it moves north); they are made true by sunlight-reflecting raindrops. Similarly, David Armstrong (2004: pp. 137–142) argues that true existential propositions about dispositions are made true by categorical properties governed by contingent laws of nature, not by ontic dispositions or so-called powers. And John Heil (2012: pp. 19, 153–155; forthcoming) argues that propositions (or “judgments”) about macroscopic objects are made true by micro particles appropriately arranged, not by macro-objects.⁴⁴ Similarly, my suggestion is that existential propositions or statements concerning corporations, and other institutional objects, are not made true by ontic institutional objects, but by declarations and other human activities and representations.⁴⁵

All in all, I think it is deeply problematic (and certainly highly misleading, if Searle does not want his claims to be taken literally) to maintain that we have an “amazing” or “quasi-magical” ability to “create” institutional facts—non-identical with mental states and true propositions—“corresponding” to the content of extralinguistic declarations. Rather, we should simply maintain that in virtue of the declarations, predicates that formerly did not apply to a certain object or set of objects come to apply to the object or sets of objects; and that in virtue of the declarations it becomes possible to quantify over institutional objects, at least in certain types of discourse.⁴⁶ Thus, saying so can indeed make it so, as Austin (1962: p. 7) originally taught us; but not because declarations create worldly, extralinguistic, institutional facts corresponding to their propositional contents, but rather because declarations “create” institutional truths and are themselves one of the truth-makers for those truths. With some provisos (see below), it would appear that Searle is in the grip of a misunderstanding: he seems to think that institutional truths must be made true by extra-linguistic institutional facts.

I want to end this section by acknowledging that, in spite of numerous formulations suggesting the contrary, Searle might in the end agree with my core ontological point that nothing is really created in the world when we perform Status Function Declarations. In The Construction of Social Reality (1995) and Making the Social World (2010), Searle says that institutional features are “ontologically subjective” (1995: p. 10, 2010: p. 18) because they are “intentionality-relative” (2010: p. 17) or “observer-relative features of the world” (1995: 9 ff.). In Speech Acts (1969: pp. 103–106), he defends a version of predicate nominalism with regard to the ontology of properties (cf. Armstrong 978: pp. 12–14). Clearly, a predicate nominalist account of putative institutional properties would combine very well with the mere-Cambridge-change view defended above.⁴⁷ Also, in Speech Acts (1969: p. 52) and in The Construction of Social Reality (1995: p. 28), Searle puts a lot of emphasis on the role of constitutive rules of the form “X counts as Y” in the “construction” of institutional “reality”. Saying that something X counts as a Y sounds a lot more innocent—especially within predicate nominalism—than saying “in the nonlinguistic Status Function Declaration, we do more than represent: we create” (Searle 2010: p. 114), or “We (or I) make it the case by Declaration that the Y status function exists” (ibid: 101). I also want to point out that in Making the Social World, Searle remarks, in passing, about corporations, that:

…we need to specify not just that the [status] function exists but that there is an entity Y, the corporation, that has the function, even though the entity is, as they say, a “fictitious” entity. […] On the standard accounts of the semantics of quantificational logic, the quantifiers range over a domain of existing objects. But in the case of the creation of entities that have status functions, there is no such domain. (Searle 2010: pp. 100, 120)⁴⁸

Moreover, in The Construction of Social Reality, Searle denies that the “correspondence” in the correspondence theory of truth—a theory which Searle defends in Ch. 9 of the book—needs to involve “resemblance” or “isomorphism” between statements and facts (1995: pp. 213–214).

It is unclear what we should make of these claims and doctrines, however. To say that something is intentionality-relative or mind-dependent, or even ontologically subjective, is not necessarily to say that it does not belong to the furniture of the universe and does not really exist. Jonathan Schaffer (2009), for example, argues that ontologically dependent or grounded entities—mind-dependent entities included (ibid: 362)—are as real or existent as the entities they depend upon. Searle himself says, about pains and intentions, that they are “ontologically subjective” in that “they exist only as experienced by human or animal subjects” (Searle 2010: p. 18), but he nevertheless maintains that mental states are “features of reality” (Searle 1995: p. 12). And, crucially, if saying that something is ontologically subjective is to say that it does not really exist, how could institutional facts, being held to be ontologically subjective, be said to be truth-makers for institutional truths (e.g. Searle 1995: p. 221)? Relatedly, how could Status Function Declarations with their putative double direction of fit, ↑↓, ever be successful in Searle’s view, if the allegedly fitting relata at the other end, the institutional facts created, were strictly speaking non-existent?

Moreover, it is unclear whether Searle is still a predicate nominalist: as far as I know, he has not defended predicate nominalism since Speech Acts (1969) and has never mentioned predicate nominalism in connection with status functions. In Making the Social World (2010), he has also made it clear that he now thinks the theory of Status Function Declarations is the fundamental theory of the “construction” of institutional reality: constitutive rules are to be interpreted as various forms of “standing declarations” with double direction of fit (e.g. Searle 2010: pp. 13, 97). Constitutive rules of the paradigmatic form “X counts as Y in C” are played down: Searle maintains that we can create status functions without there being a brute object X which is counted as a Y (i.e. in cases of free-standing Y terms; e.g. Searle 2010: pp. 20–21, 97–102).

Complicating matters further, although Searle occasionally describes institutional objects such as corporations as fictitious entities, he also explicitly maintains, as we have seen, that they are “created” and that we make this the case by declarations that they “exist” (Searle 2010: pp. 120, 122).⁴⁹ And although he says that he is not committed to correspondence as isomorphism, he does at various places claim that facts are what make true statements true (e.g. 1995: pp. 211, 221), that facts have a “structure” (1995: 221), indeed “a propositional structure” (2014: p. 20), and that facts “have to be propositionally specified” (1995: p. 212), or alternatively, “can only be stated and not named” (1995: p. 211). All of which strongly suggests that facts have an object-property structure somehow mirroring the subject-predicate structure of the statements or propositions that specify them and are made true by them.

Concerning institutional facts, he explicitly says:

Once you see the power of the Declaration to create an institutional reality, a reality of governments, universities, marriages, private property, money, and all the rest of it, you can see that social reality has a formal structure as simple and elegant as the structure of the language used to create it. (Searle 2010: p. 16, my emphasis)

Such remarks strongly suggest that a Status Function Declaration, whose propositional content is of the form “a is F”, brings into existence a corresponding and structurally similar institutional fact, a fact of the form a’s being F, where F is an institutional property.

Be this as it may. If, appearances to the contrary, Searle is defending nothing more than a mere-Cambridge-change account of the creation or imposition of “status functions”, if he is, more generally, wedded to no more than the notion that our declarations create institutional truths, not corresponding worldly institutional facts or states of affairs, then this paper can be seen as a friendly plea to him (or to his followers) to explicitly formulate the theory in such more modest terms.

5 What follows?

If nothing is created by Status Function Declarations except institutional truths, what follows? One important implication concerns causation. If the institutional properties, objects and facts allegedly created by declarations do not exist in a worldly sense, they cannot be relata of causal relations (or any other relations for that matter). Yet, we often speak as if such properties/objects/facts have a causal impact or partake in causal processes. We do so in everyday speech (“His poor grades caused him to lose his scholarship”, “Money causes more harm than good”, “…this was caused by the fact that the company had…” etc.); and the academic literature (in philosophy, sociology, political science, etc.) is replete with such institutional causal claims.⁵⁰ And Searle, as we have seen, sometimes seems to suggest that institutional facts and status functions are causally potent: “facts [in general, institutional facts included] function causally in a way that true statements do not” (1995: 206), and “a function [status functions included] is a cause with a purpose” (2006: p. 17, 2010: p. 59). But again, if status functions and institutional facts do not really exist as worldly entities, they cannot be causes.⁵¹ At least, they cannot be what we might call “sparse causes” (in analogy with the distinction between sparse/ontic and abundant/predicate-nominalistic properties, see Sect. 4 above). A sparse cause, let us say, is an ontic entity that is a relatum of an ontic causal transaction or process of some kind. A “mere abundant cause”, let us stipulate, figures as a (putative) cause in a true causal statement, without being a sparse cause. To illustrate: it is sometimes argued that causal statements involving “absences” and “omissions” as causes can be true, although absences and omissions do not exist as worldly or ontic entities (e.g. Mellor 1995, Ch. 13; Dowe 2000, Ch. 6; Lewis 2000/2004: pp. 99–102; Woodward 2003: pp. 86–91); on such a view, absences and omissions fall out as mere abundant causes, according to the terminology suggested here.⁵² Similarly, one could maintain—in an attempt at rescuing some watered-down form of institutional causation—that institutional properties/objects/facts can figure as mere abundant causes in true, institutional, diachronic causal statements (such as “His poor grades caused him to be depressed”, “…caused him to lead a miserable life”, etc.). But importantly, if what I have been arguing is correct, the truth-makers for such causal statements (assuming they are true) will not involve institutional properties/objects/facts.⁵³ Rather, the key truth-makers will in many cases, in all likelihood, involve (although certainly not exhaustively) people’s mental representations of institutional properties/objects/facts (see also my forthcoming). Nothing about what I have said rules out that representations of institutional entities can be sparsely causally efficacious and influence the way that people feel, reason and behave.⁵⁴ A detailed positive account of these causal issues will have to wait for another occasion, however.⁵⁵