Contribution
A close critique of the one place Searle gives an account of collective intentions, “Collective Intentions and Actions” (1990, LIT-490). It argues that the paper never analyses a we-intention: the two collective analyses it states are rejected by Searle himself, and the analysis he favours describes an individual’s complex intention. It also argues that Searle’s own theory of intentions, in Intentionality, makes irreducibly collective intentions unlikely. For this record its main value is as a report, with page references, of what the unread 1990 paper argues and what it concedes.
Key insight
“Complex” does not mean “collective” (p. 228). An intention to make the sauce by means of my stirring has the structure of an intention to fire a gun by means of pulling the trigger. Adding that the end is a joint action changes the action described, not the kind of mental state. Searle shows that some collective actions involve complex individual intentions; he does not show that there are collective intentions.
Assumptions
- Searle’s theory of intentions is taken as given. Intentions, unlike beliefs and desires, have only actions as their conditions of satisfaction, are self-referential, and cannot be directed at what is beyond the agent’s control (pp. 211–213).
- Collective intentions and collective intentionality are different claims. Searle’s ontology, as Zaibert reads The Construction of Social Reality, needs we-intentions (CSR 23–26); Searle’s arguments address collective intentionality in general (pp. 222, 229).
- “Intentional” is ambiguous. Zaibert separates intentionality of mental states from intentional action, and the latter into a descriptive, a normative (blaming) and a loose sense (pp. 216–217). He holds that intending X is neither necessary nor sufficient for doing X intentionally (p. 214).
Key results
- Searle’s tone in 1990 (pp. 220–222). As Zaibert quotes it, the 1990 paper is “uncharacteristically tentative”. Searle asks “how far can the theory of intentional action in Intentionality … be extended to become a general theory” (CIA 401). The intuition has two parts: “collective intentional behavior is a primitive phenomenon”, which he calls obvious, and “collective intentions expressed in the form of ‘we intend to do such and such’ … are also primitive phenomena”, which he calls “problematic” (CIA 401–402). On reduction he says: “I have not demonstrated that no such analysis could ever succeed” (CIA 406), and his arguments “suggest that our intuition is right” (CIA 401). His claim of a form of collective intentionality that is neither a group mind nor reducible “has plenty of problems of its own” (CIA 406). Of the means–end step: “the answer to that question is not at all obvious” (CIA 411). And of his conclusion: “I am not sure this is the right analysis, but it does seem better than the three others we considered” (CIA 412).
- The postponed analysis (p. 222). Searle says the individual “I intend”s are “derivative from the we-intend’s” (CIA 403), and that the structure of we-intentions must wait on a prior question (CIA 406). The section that answers it opens with “What exactly is the formal structure of collective intentionality?” (CIA 408), which Zaibert notes is a question about intentionality in general, not intentions.
- The notation (pp. 223–224). “i.a. (this i.a. causes: my arm goes up)” for raising an arm (CIA 408–409). The hollandaise case, where Jones stirs and Smith pours and each could say “We are preparing hollandaise sauce” (CIA 410), is written “i.a. (this i.a. causes: sauce is mixed)”. Zaibert: nothing in this marks it as collective, and with Searle’s dropped capitals it is formally identical to raising one’s arm.
- The by-means-of relation (p. 224). Searle calls the “by and by-means-of relations” “one of the keys for understanding collective intentionality” (CIA 410). Jones’s content is “we are making the sauce by means of me stirring”, Smith’s ”… by means of me pouring” (CIA 410). Zaibert: the two contents differ, so they cannot share a condition of satisfaction, and nothing shows the “we are making the sauce” part is the content of an intention rather than a desire or belief. A wind-mixed sauce would satisfy Smith though not in the way a we-intention requires (p. 225).
- The four analyses (pp. 225–228).
- (I) “collective i.a. (this collective i.a. causes: ingredients are stirred, causes: sauce is mixed)” (CIA 411), rejected by Searle as “collectivist or socialist” because “it leaves out the fact that Jones is making an individual contribution to a collective goal”. Zaibert agrees, and says this is Searle’s own general view of we-intentions.
- (II) “singular i.a. (this singular i.a. causes: stirred, causes: mixed)”, rejected as “capitalist or individualist” because “it is consistent with there being no collective intentionality at all” (CIA 411). Zaibert: this begs the question, since whether there is irreducible collective intentionality is what is at issue.
- (III) a collective i.a. causing a singular i.a. Searle rejects it because one intention would be in the scope of another; his gun case is “only one complex intention, not two intentions where one causes the other” (CIA 411–12). Zaibert agrees.
- (IV) “i.a. collective B by means of singular A (this i.a. causes: A stirred, causes: B mixed)” (CIA 412), the one Searle “half-heartedly favors”. Its individual counterpart is “i.a. B by means of A (this i.a. causes: A trigger pulls, causes: B gun fires)”. Searle says the relation of the collective to the individual intention “is exactly like the relation of the intention to pull the trigger and the intention to fire the gun” (CIA 412). Zaibert: then (IV) differs from the gun case only in that one action described is collective; as to the mental state the two are identical, a single individual by-means-of intention (p. 228).
- Tuomela and Miller (pp. 224, 226, n. 16). Searle calls their account “the best he has seen” (CIA 404) and rejects it because “it attempts to reduce collective intentions to individual intentions plus beliefs” and “no such reduction will work” (CIA 404). His argument, as n. 16 reports it, is one case: ethical egoists who each act selfishly, believing the others do too, for humanity’s good, meet Tuomela and Miller’s conditions yet have no we-intention, which needs “a pact or agreement of some sort”. Zaibert replies that the case works only if every collective benefit is the aim of a collective goal, and cites Searle’s own “It is not of course to be supposed that a single counter-example can refute a philosophical thesis” (IOQ 120) against him (p. 226).
- Animals (pp. 219–220). Searle attributes intentional action to a dog chasing a ball (I 101) and collective behaviour to hunting animals. Zaibert holds animals have beliefs and desires but not intentions, and that groups likewise cannot have irreducible we-intentions, though they may have “we-desires or we-beliefs” (p. 220).
- Conclusion (pp. 229–230). Searle “has neither shown that the attempt to explain we-intentions in terms of individual intentions is doomed nor presented a positive account of we-intentions”. The gap could be filled by a real analysis of collective intentions, reducible or not, or by letting collective desires carry the weight: “A definition of a collective desire … seems, after all, a more plausible suggestion than a collective intention.”
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Searle’s 1990 paper concedes it has not demonstrated that collective intentions are irreducible, and offers its arguments as suggesting an intuition | strong as a report | quotations of CIA 401, 402, 406, 411, 412 (pp. 221–222, 226) |
| C2 | Searle’s favoured analysis (IV) states a complex individual intention, not a collective one | moderate | formal comparison of (IV) with the gun case, CIA 412 (p. 228) |
| C3 | Searle’s rejection of the individualist analysis (II) begs the question | moderate | the stated reason, CIA 411, is that it allows no collective intentionality (p. 226) |
| C4 | Searle’s rejection of Tuomela and Miller rests on one doubtful counterexample | moderate | n. 16; the egoist case could be met by restating the egoists’ goal |
| C5 | Searle’s ontology needs collective intentions but argues only for collective intentionality | moderate | CSR 23–26 against CIA 408; the paper’s reading of CSR, unread here |
| C6 | Intentions, unlike desires and beliefs, cannot be irreducibly collective | weak | analogy with animals and the uniqueness of intentions; no argument from the structure of groups |
| C7 | Intending X is neither necessary nor sufficient for doing X intentionally, because “intentional” has a normative sense | moderate | Butler’s dice and revolver puzzle, Bentham’s oblique intention, criminal law (pp. 214–218) |
Concepts
- We-intention — a collective intention expressed “we intend to do such and such”, which Searle holds primitive and irreducible (CIA 401).
- Intention-in-action (i.a.) — “the Intentional content of the action” (I 84), as against a prior intention; every intentional action has one (I 85).
- By-means-of intention — one complex intention to do B by doing A, not two intentions one causing the other (CIA 411–12).
- Intentional, descriptive, normative and loose — Zaibert’s three senses of “intentional” as said of actions: how an action was done; a ground of blame or praise; merely accompanied by some intentional state (p. 216).
Connections
Searle’s 2006 paper (LIT-482, read in NOTE-388) refers its notion of collective intentionality to the 1990 paper, and THEORY-625 records that it defers the irreducibility argument there. This reading supplies that paper’s content at second hand. Zaibert’s distinction between collective intentions and collective intentionality bears on what THEORY-625 calls collective acceptance: if acceptance is not an intention, Zaibert’s argument against irreducible we-intentions does not reach it, and he himself leaves collective desires and beliefs open (p. 220). Tuomela and Miller, Bratman’s Faces of Intention and Gilbert are cited or named and not held here.
Bearing on the record
- THEORY-625, against, at second hand. Its promote_when needs “an argument that collective intentionality is irreducible to interlocking individual attitudes”, which the 2006 paper defers to LIT-490. By Zaibert’s quotations, LIT-490 does not give one: Searle says “I have not demonstrated that no such analysis could ever succeed” (CIA 406), offers the irreducibility as an intuition his arguments “suggest” (CIA 401), and rejects the best reductive account by a single case (CIA 404). So the reading does not meet what THEORY-625 waits on, and it indicates that a first-hand reading of LIT-490 would not meet it either. It does not refute THEORY-625’s account of institutions, which needs collective acceptance rather than collective intentions in Zaibert’s narrow sense. This is a critic’s report and does not replace reading LIT-490.
- THEORY-414, the rival that treats institutions as rules in equilibrium among individuals, is not discussed, but the report removes one support THEORY-625 might have had against it: the 1990 paper’s irreducibility claim is stated as an intuition.
- No instruction for machine-learning practice; nothing for the anthology.
Limitations
- Second hand. Every claim about LIT-490 rests on Zaibert’s quotations and his notation. He drops Searle’s upper-case expressions after Searle does, and some of his formal comparison depends on restoring them.
- The positive thesis is thin. That groups cannot have irreducible we-intentions is argued by analogy with animals and from the uniqueness of intentions, not from anything about groups (pp. 219–220). The suggestion of collective desires is one paragraph.
- Section II–III is a separate argument about the ambiguity of “intentional”, which connects to collective intentions only through the charge that Searle uses “intentional behavior” ambiguously (p. 219).
- Later Searle is not considered. The paper reads the 1983, 1990 and 1995 texts only.
Open questions
- Does LIT-490 itself, read first-hand, contain more than Zaibert reports, for instance an argument against reduction beyond the egoist case?
- Is collective acceptance, as Searle’s later work uses it, an intention, a desire or a belief? Zaibert’s critique reaches only the first.