Contribution
It splits the notion of group identification that the shared-agency literature borrowed from social psychology into two processes:
- adopting the we-perspective: reasoning about what is best for the group, as in team reasoning;
- self-transformation: coming to understand oneself as a group member.
The two dissociate, and self-transformation is developmentally prior. It then gives self-transformation a cognitive kind. It is a Pushmi-Pullyu representation (Millikan): a state that represents a fact and directs action at once, so identifying with a group is already being moved to act as a member, with no further desire needed.
Key insight
Thinking about what is good for one’s group is not the same as being one of the group. A committee member caught in a discursive dilemma can work out what “we” should conclude while personally disagreeing (§3.2). A two-year-old can act as one of “us” without being able to take the group’s perspective at all (§3.1). Only the second is identification in the self-involving sense. It is motivating by its nature: it tells you both what you are and what to do.
Assumptions
- Joint action is explained by shared intention (Bratman, Gilbert) for full-fledged agents. The paper’s quarrel is with “lite” versions meant to cover infants (Pacherie 2013).
- Developmental evidence that adopting a group perspective, shown by the black sheep effect, appears around age three and matures by eight, later than joint action at 18–24 months (§3.1).
- Millikan’s category of PPRs is real. The paper does not commit to her theory of content (§4).
- Internalism about group identification (note 6). Externalists are not addressed.
Key results
- The two-condition view (§2). Pacherie’s theory of shared intention requires (i) a self-conception as a team member, (i′) the belief that others have it, (ii) team reasoning from the group viewpoint, and (iii) the intention to do one’s part. Petersson extends group identification to all joint action in the “we-mode”, via Bacharach’s “agency transformation”.
- Team reasoner versus team benefactor (§2, after Tuomela; Hakli et al.). The team reasoner holds “We intend to J”, with the “we” in subject position. The team benefactor reasons in the first person singular about the group’s well-being and holds “I intend that we J”. Only the first is a group member’s attitude.
- Self-transformation without the we-perspective (§3.1). Precursors of the black sheep effect appear by three, mature forms by eight, while joint action appears at 18–24 months. So infants’ joint action cannot rest on adopting a group perspective. The paper calls this “conditional and defeasible”.
- The we-perspective without self-transformation (§3.2). In the discursive dilemma (List & Pettit), a member can form “We intend to J” while holding “I do not intend to J”. This is Goldie’s “in-his-shoes” perspective-taking applied to a group. What closes the gap, turning it into “empathetic perspective-shifting”, is self-transformation, not more perspective-taking.
- The social self as a PPR (§4). A belief “I am a member of us” does not motivate, and people can hold it without identifying (Ruble et al.; Bennett & Sani). The social self is a PPR: “you are a member of us; act as such”. Like the bee’s waggle dance, it represents and directs at once. It is triggered by group cues: Bacharach’s list, plus joint attention, which the authors propose for children.
- Three implications (§5).
- Minimal joint action needs no complex mentalizing.
- PPRs spread, which may explain why cooperation is “contagious”.
- Self-transformation is not under direct rational control, though one can arrange the situations that elicit it; adopting a perspective is.
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Adopting the group’s perspective and understanding oneself as a member are distinct processes that can occur without each other | moderate: two arguments, one from developmental timing (indirect) and one from the discursive dilemma (conceptual) | §3 |
| C2 | Infants’ joint actions rest on self-transformation alone, not on adopting a group perspective | weak: inferred from the later appearance of the black sheep effect; the authors call it conditional and defeasible | §3.1 |
| C3 | The social self is a Pushmi-Pullyu representation: descriptive and motivating at once, not a belief plus a desire | weak to moderate: argued by elimination of the doxastic view and by analogy; no direct evidence | §4 |
| C4 | Group identification in the sense of self-transformation is not under direct rational control | weak: follows from C3 | §5 |
Concepts
- self-transformation: transformation in self-understanding, coming to understand oneself as a group member (a “social self”).
- adoption of the we-perspective: framing a situation from the group’s point of view, as needed for team reasoning.
- team reasoner / team benefactor: one who intends as part of “we” (“We intend to J”), versus one who intends, as an individual, to promote the group’s good (“I intend that we J”).
- Pushmi-Pullyu representation (PPR): a representation undifferentiated between describing a fact and directing action (Millikan).
- in-his-shoes perspective-taking / empathetic perspective-shifting: imagining oneself in the other’s (here the group’s) situation, versus imagining being the other (Goldie).
Connections
- The owner’s thesis (CLAIM-491). The paper bears on it at three
points. These are inferences; the paper does not discuss selflessness.
- The team benefactor reasons from the first person singular about the group’s good, which is the thesis’s “self-interest on a group” read as one would naturally read it. The team reasoner is something else: their “we” is in subject position, so the good pursued is the subject’s own good only if the subject is the we.
- The PPR account says identification motivates directly (“act as such”). That fits the thesis’s picture that collectivist action flows from a self-understanding, not from a separate desire for others’ welfare.
- The paper supplies a criterion of self-extension independent of caring: the activated social self, triggered by cues. That is what CLAIM-492 says the thesis needs.
- Self-categorization theory (THEORY-1114). The working definition is Turner’s (1982): group identification internalizes a categorization into the self-concept. The PPR account is a cognitive gloss on depersonalization, and Bacharach’s group cues (common fate, competing group, we-language) are close to fit.
- Skorich (THEORY-1112). Both make joint attention and shared mental states central to group self-perception, by different routes.
- Sugden (LIT-1782) and List & Pettit (LIT-391) supply the team reasoning and the discursive dilemma.
Bearing on the record
- The owner’s thesis. The team-reasoner/team-benefactor distinction gives CLAIM-491 a sharper form. The thesis is strongest for team reasoners whose social self is activated, and weakest for team benefactors, who care about a group they do not identify with as self-transformed members. A dated note is added to the claim.
- No THEORY filed. The PPR account is a conjecture argued by elimination; it is recorded here.
- No ML instruction.
Limitations
- Conceptual. The developmental timeline is borrowed and the authors flag its use as defeasible.
- The doxastic view is dismissed quickly. That one can believe one is a member without identifying shows only that belief is not sufficient, not that identification is not a belief plus something else.
- Externalist accounts are set aside (note 6).
Open questions
- Can self-transformation be measured apart from perspective-taking in adults? A task that separates “we-reasoning while disagreeing” from identification would test C1 directly.