Ned Block (1987), “Functional Role and Truth Conditions”, symposium with John Campbell, Proceedings of the Aristotelian Society, Supplementary Volumes 61, pp. 157–181 (Block’s paper) — DOI-10.1093/aristoteliansupp/61.1.157.
This is a distinct work from “Advertisement for a Semantics for Psychology” (LIT-648), shorter and narrower, and it cites that paper for the claim that functional role determines the nature of the referential factor (p. 170, pointing to pp. 643–644 of LIT-648).
Key takeaways
- Five facts (pp. 162–166). Grasping some concepts involves grasping others; having a thought involves being able to have related thoughts (Evans’s generality principle); phlogiston cannot be translated, only learned with its role (Kuhn); translation cannot attribute irrationality without limit; grasping a thought involves grasping its structure. Functional role semantics explains each, which “shift[s] the burden to the competitors”.
- Non-reductionist roles cannot fix Twin Earth truth conditions (pp. 168–169). Defining ‘aluminum’ by its links to ‘light’, ‘grey’, ‘conducts’ does not help: those words have the same extensions on both planets, so nothing in the inferences separates aluminum from twin aluminum.
- Public use converges with private use (pp. 172–178). Public use “in the sense in which it could be recorded by a movie camera” is identical on the two planets; appealing to what the words are used to refer to begs the question. The only workable move is long-arm roles that include aluminum in their inputs and outputs (Harman). Then the space traveller who radios “Good place to get aluminum” from Twin Earth needs a “normal context” or “dominant causal source” twist, ‘Moses’ needs roles reaching thousands of years back, and ‘elm’ needs experts. The external part is a theory of reference; the internal part is “the same for earthians and twin-earthians” and “can be identified with the narrow meaning of ‘aluminum’ — if narrow meaning exists” (p. 177).
- Which to prefer is a matter of purpose (n. 31): “The short-arm roles are better for psychological explanation, long-arm roles are better for thinking about translation and communication.”
- Burge’s arthritis (Appendix, pp. 179–181). Causal chains cannot explain the case, since the language change in the counterfactual community need never reach Fred. What is needed is “functional role in a language”, cashed out through “the structure of deference” and perhaps an idealised individual role. But individual roles matter too: if A thinks a fortnight is ten days, we may report him as believing B arrives in ten days.
Standing in the record
Filed on 2026-10-03 at the owner’s request. It is the work the record’s “Block 1987” citations refer to. NOTE-117, reading Drayson (LIT-112), cites “Block 1987” four times (in its key results, claim C3, its limitations and its corrections) as a functional role semantics on which mental representations take content from “their role in the communicative and social environment”. Drayson’s reference list, checked against her preprint, gives “Block, Ned (1987). Functional Role and Truth Conditions. Aristotelian Society Supplementary Volume 61 (1):157-184”. So NOTE-117’s “Block 1987” is this paper, and the coordinator can point those citations here.
What the paper actually supports in Drayson’s use: its public-use section (pp. 172–178) is exactly a functional role theory in which communicative use and the community’s experts enter the role, and its appendix adds a community-level role for Burge cases. So Drayson’s reading is fair. But Block’s own conclusion is that such a theory is equivalent to a two-factor theory with a narrow internal role. On Block’s account, then, the social content Drayson needs lives in the referential or community factor, not in the internal role of the mental representation itself. NOTE-117’s limitation, that Block 1987 “is named, not developed”, is answered here: it is developed, and it is more equivocal than Drayson’s one sentence. That reading of Drayson against Block is mine.
philosophy-of-language primary, cognition for its claims about thought
and psychological explanation. No anthology topic holds it, and it carries
no instruction for machine-learning practice.
Relation declared. extends: LIT-648. The paper develops two
points of the “Advertisement”: the argument against Harman’s one-factor
long-arm roles becomes a general argument that public-use theories converge
on two factors, and the social dimension the “Advertisement” left to a
possible third factor is given a first treatment in the appendix.
Elsewhere in the record. “Inverted Earth” (LIT-634, n. 26) cites this paper for the claim that Harman’s long-arm roles are equivalent to short-arm roles plus a referential component, and builds its argument on long-arm roles.