NOTE-388: Social ontology: Some basic principles

Contribution

A compact statement of the theory of The Construction of Social Reality (LIT-485), written for social scientists, with three additions to the book:

• status indicators can take on “a kind of life of their own” (after De Soto), §III;
• some status functions have no X: “free-standing Y terms” (after Barry Smith), such as corporations and money without currency, §III;
• “X counts as Y in C” is placed under the more general “We accept (S has power (S does A))”, §III. Searle says the book already gave that formula in a later chapter (p. 65); what is new is ranking it above the first.

There is also a first sketch of a taxonomy of status functions (§IV), a defence of the method as conceptual analysis rather than empirical generalisation (§V), and a demarcation of which “institutions” the theory covers (§VI). Nothing here argues for the account of collective intentionality. That is referred to Searle (1990), LIT-490 (fn. 3).

Key insight

Status functions are the glue. A thing performs a function not because of its physical structure but because a group collectively accepts that it has a status. That status brings rights, duties and permissions, which Searle calls deontic powers. To recognise them is to have a reason to act that does not depend on what one wants. Institutions are systems of such powers. They exist only as represented, which is why they need language.

Assumptions

• Physicalist background. “The world consists entirely of physical particles in fields of force” (p. 52). The problem is how social reality fits into that world, not whether it does.
• Collective intentionality as a primitive. Collective intentionality is “a genuine biological phenomenon”, “not mysterious or inexplicable”, and the “psychological presupposition of all social reality” (p. 57). Its analysis is deferred to Searle (1990). In this paper it is illustrated by one formula, “I am doing something only as part of our doing something” (p. 56), and no argument is given that it is irreducible.
• The examples are the author’s own society: the US presidency, California corporation law, a twenty dollar bill, American football. §V answers the charge that this is a narrow base. The claim is about logical structure, which any society with a deontology must have.
• Animal capacities. The human/animal line is drawn on deontology. It is held provisionally: if some primates turn out to have deontic relations, “so much the better for them” (p. 62).

Key results

The paper has no theorems. Its results are distinctions and classifications, each argued informally or by example.

• Observer-relative versus observer-independent (p. 53). A rough test: could the fact exist with no conscious agents? Observer-relative facts are created by observer-independent attitudes.
• Epistemic versus ontological objectivity (pp. 54–55). “Observer relativity implies ontological subjectivity but ontological subjectivity does not preclude epistemic objectivity.” That, he says, is why social science is possible.
• Three primitives (p. 56): collective intentionality, the assignment of function, and constitutive rules.
• Definitions (p. 57). A social fact is “any fact involving collective intentionality of two or more human or animal agents”, so a wolf pack hunting is one. Institutional facts are the subclass that involve status functions.
• Functions (p. 57). “All functions are observer relative”, argued from the normativity of function. He adds that the main argument does not depend on it.
• “X counts as Y in C” (p. 58) iterates upward (noises → sentence → promise → contract) and interlocks laterally.
• Deontic powers (p. 59). Accepting a status function is accepting obligations, rights, duties and so on. These are “desire-independent reasons for action”, and “animals lack a deontology”.
• Language’s four roles (pp. 61–62). The fact exists only as represented. Its content is deontic. The deontology persists after its creation (“no language, no deontology”). The institution itself must be recognised, not only its instances. The dog can see a man carry a ball across a line but cannot see a touchdown (p. 60).
• Status indicators (p. 63). De Soto’s squatters have recognised status but no deeds, so their deontic powers “stop at the point where the larger society requires some official proof”.
• Free-standing Y terms (pp. 63–66). A corporation is created by declaration with no X; money needs no currency, only recorded numerical values. In such cases “the real bearer of the deontology is the participant”. Chess pieces and bills are “markers for the amount of deontic power that the players have”.
• The general form (p. 65). The book’s early chapters gave “X counts as Y in C” as the basic form, but “a later formulation I gave in the book gives us a much more general account”: We accept (S has power (S does A)). The two are related thus: we accept that S has the power because S = X and X counts as Y, and the Y status carries the powers.
• Desire-based reasons presuppose desire-independent ones (p. 66). The desire to be rich or be president presupposes recognised deontic relations.
• Taxonomy, tentative (pp. 66–68). Status functions where physical capacities are essential (a licensed driver) versus where they are not (money). Certification versus authorisation. Positive versus negative powers. Taxes versus punishments, and salaries versus prizes, as distinct status functions with the same monetary upshot.
• Scope (pp. 69–71). A date, a colour word or a unit of weight is not an institutional fact. Friendship is, “because friendship carries obligations”. “Nerd” can become one where a group recognises obligations attached to it. Science, religion and education are practices containing institutions, not institutions in this sense.

Claims

id	claim	strength	support
C1	Social facts are ontologically subjective yet epistemically objective, and are created by observer-independent attitudes	strong as a distinction	conceptual argument, pp. 52–55
C2	Three primitives (collective intentionality, assignment of function, constitutive rules) suffice for institutional reality	assertion	”I am going to be very brief because I am repeating things I have said elsewhere” (p. 56); illustrated, not argued
C3	All functions are observer-relative	weak	one argument from normativity (p. 57); declared inessential to the paper
C4	Status functions carry deontic powers, which give desire-independent reasons for action, and these are what animals lack	moderate for the human half (examples, pp. 59, 66); weak for animals	the animal claim is hedged by the author (p. 62)
C5	Institutional facts require language in a broad sense, because they exist only as represented	moderate	the touchdown argument and the four roles (pp. 60–62)
C6	”X counts as Y in C” is not general; free-standing Y terms exist, and “We accept (S has power (S does A))” is more general	moderate	corporations and money (pp. 63–66); the relation between the two forms is stated, not derived
C7	The analysis is conceptual, so a society lacking promising or driver’s licences is no counterexample	moderate	the analogy with speech-act taxonomy (pp. 68–69); the author admits it could be refuted empirically by an institution of another structure
C8	Status functions divide into certification/authorisation and positive/negative powers, among others	assertion	offered as unsatisfactory by the author (“I still do not have a taxonomy … that I find satisfactory”, p. 66)

Concepts

• Observer-relative: a feature that exists only relative to the attitudes of agents (money, citizenship). Observer-independent: one that does not (mass, photosynthesis, and the attitudes themselves).
• Collective intentionality: intentionality “shared by different people”, in the forms “we believe”, “we desire”, “we intend”. In action, “I am doing something only as part of our doing something” (p. 56).
• Status function: a function a person or object performs not by its physical structure but by collective acceptance of a status (pp. 57–58).
• Constitutive rule: a rule of the form X counts as Y in C. It creates the activity it governs.
• Deontic powers: rights, duties, obligations, permissions, authorisations, requirements, and the like, positive and negative.
• Desire-independent reason for action: a reason that comes from recognising a deontic power, whatever one wants.
• Status indicator: a uniform, deed, licence or passport marking a status that nothing physical marks.
• Free-standing Y term: a status function with no physical object on which it is imposed (Barry Smith’s term).

Connections

The paper restates The Construction of Social Reality (LIT-485) and sends the reader to “Collective Intentions and Actions” (LIT-490) for collective intentionality. Both are filed Deferred. It does not engage with any rival account of collective intentionality: Gilbert (LIT-477, LIT-492), Bratman (LIT-471, LIT-488) and Tuomela & Miller (LIT-476) are not mentioned.

Its rivals in this record are later. Hindriks & Guala (LIT-481, read in NOTE-395) target the book’s constitutive rules. They argue that the Y term is a theoretical term eliminable in favour of regulative rules, and that “desire-independent reasons” is either a platitude or an unargued anti-Humean thesis. This paper’s free-standing Y terms bear on that dispute. When Searle says the deontic power of a corporation or of money goes “directly to the individuals”, and that bills are “markers” for powers, he treats the Y term much as they do. He does not draw their conclusion, and he keeps “desire-independent” as a primitive distinction.

Bearing on the record

• THEORY-043, indirect. The account says nothing about group minds. It bears on the anti-nesting question only through its ontology. A society such as “the United States of America” (p. 52) exists as that society only because its members represent it as existing. A brain’s parts do not represent the brain. That is the shape of Kammerer’s exception (LIT-409) and could be a principled asymmetry of the kind THEORY-043’s promote_when asks for. Searle does not argue this, and it is the reader’s inference. It would also need an argument that a society’s being observer-relative bears on whether it can be a subject, which is a separate question.
• Tomasello (LIT-125, NOTE-185). Searle’s human/animal line on deontology is the conceptual counterpart of Tomasello’s developmental claim that felt obligation comes with joint commitment from about age 3. They differ in vocabulary: Searle’s collective intentionality includes hunting wolves, while Tomasello’s shared intentionality is species-unique.
• List & Pettit (LIT-391). The corporation here is a deontic construction, and its bearers of power are individuals. This is compatible with, but does not supply, an account of the corporation as an agent.
• Chinese Room (LIT-400). A different subject. The common premise is that intentionality is a biological feature of individual brains, which here makes the creating attitudes observer-independent.
• No instruction for machine-learning practice. Nothing here points to the Anthology of the SOTA.

Limitations

• It is a summary and update, and it says so. The theory of collective intentionality is not argued here, and neither is the claim that the three primitives suffice.
• Searle does not reconcile his own revision with the theory’s original form. A status function is defined as imposed “on a person or an object” (p. 57), and then corporations have none. The new general form is offered (p. 65), but the paper does not say whether “X counts as Y in C” is now derivative or merely the common case.
• The test for observer relativity is admitted to be “rough-and-ready” (p. 53), and the paper offers no sharper one.
• The taxonomy (§IV) is offered as unsatisfactory by its author.
• The scope criterion in §VI, a collectively recognised deontology, is wider than the X-counts-as-Y form. It makes friendship an institutional fact without saying what X counts as what Y.

Open questions

• Is “X counts as Y in C” eliminable in favour of “We accept (S has power (S does A))”? If so, how does the account differ from the rules-in-equilibrium view (LIT-481)?
• What is the content of the “acceptance” in “We accept”? Is it Searle’s irreducible we-intention, a joint commitment (Gilbert), or interlocking individual intentions (Bratman)? The paper uses it without saying, and nothing here would decide it.
• Where exactly is the animal line? Evidence of deontic relations in primates would, by Searle’s own statement, move it, and LIT-125 is the record’s nearest evidence on the human side.

Corrections

• none to a seeded skim (there was no seed)
• Two printings, not one. The article appears as Anthropological Theory 6(1): 12–29 (March 2006) and as Papers: Revista de Sociologia 80: 51–71 (April 2006). The Crossref records differ in the abstract and the references. The AT record has ten references including Barry Smith, “John Searle: From Speech Acts to Social Reality” (2003). The Papers text has nine and credits Smith only in prose (p. 63). Anything here that depends on the exact wording is from the Papers text.
• The text predates its publication. It speaks of “the current presidential election campaign” (p. 52) and says “today is the 15th of July, 2004” (p. 69). It was written in mid-2004.
• Reference slips in the Papers printing. Rosaldo (1982) is listed as “Llongot” in “Language and Society”. Crossref’s record of the article is “Ilongot” in Language in Society 11: 203–237.