Source
Drayson (2025), What motivates mental fictionalism?, LIT-112, read in full in NOTE-117 from the author’s preprint; page numbers are the preprint’s.
What was actually shown
The target. Toon’s case for mental fictionalism has the two steps any fictionalism needs (§1): a negative claim, that the discourse lacks truthmakers, and a positive claim, that it is worth keeping anyway.
Against the negative claim (§2). Drayson reconstructs it (p. 4): P1 on the Cartesian reading, folk psychology is committed to attitudes as internal, intentional, causal states; P2 that commitment is a commitment to the representational theory of mind (RTM); P3 RTM is false, because internal mental representation is incoherent; so folk psychology lacks truthmakers. The argument is valid only if P2 is an entailment, and Toon’s text shows only that Cartesians “tend to subscribe” to RTM.
- P2 is false (pp. 6–7). RTM’s core is that mental representations are individuated apart from attitudes, so believing-that-p and desiring-that-p can token the same one. Non-representational functionalism (Lewis 1966, Putnam 1967, Loar 1982) individuates attitudes as functional wholes with no further content-bearing constituents. Such states are internal in Toon’s Sellarsian sense (theoretical, not spatially inner), intentional at the personal level “even if none of its subpersonal realizers” are, and causal as brain states. Making P2 true by definition would class these theories as anti-Cartesian and weaken Toon’s positive argument (p. 7).
- P3 is unsupported (pp. 7–8). Toon’s case is metasemantic: intentionality belongs first to public, socially normed representations. RTM is not committed to a reductive, mind-first metasemantics; that is one motive for it (Fodor’s), and functional role semantics could let inner representations take content from social use (Block 1987, LIT-625).
- Upshot (p. 9): “Even if there were no mental representations, we could still coherently claim that there were ‘Cartesian’ truthmakers.”
Against the positive claim (§3, pp. 9–12). Toon holds that pretence about an inner notebook lets us make genuine assertions that a person is in “a certain state”, which is what gives fictionalism its causal advantage over behaviourism and instrumentalism. Drayson’s dilemma: if that overall state is a set of behavioural facts, fictionalism has no causal-explanatory advantage over anti-Cartesian realism; if it is the state S that causes the behaviour (Toon p. 52), the assertions commit Toon to S, and “why not identify state S as Mark’s belief?” Realist views on which attitudes are functional or dispositional states of the whole person with non-representational causal bases are available (Egan 2025; Langland-Hassan 2020). So the positive argument “provides Toon himself with a reason to be a realist”.
What this does not say
- That mental fictionalism is false or incoherent. Coherence is assumed for the sake of argument (n. 2), and §4 says fictionalism need not wait for all realisms to fail. The paper defeats two motivations.
- That RTM is true. Drayson shares Toon’s skepticism about “an architecturally-innate and linguistically-structured” RTM (§4). Her point is that realism about beliefs does not stand or fall with it.
- That whole-person functional states are shown to be causes. That they are causally efficacious “as a state of the brain” is asserted, not defended against exclusion arguments (NOTE-117, Limitations).
- Which realism is right. Footnote 20 grants interpretivist views (Curry 2025, Williams 2019) the same “overall state” without fictionalist semantics. The argument needs only that some realism is available.
- That the dilemma refutes a fixed position. Toon “sometimes suggests” the behavioural reading (p. 11), so the objection is a dilemma.
- Anything about artificial systems. The paper does not discuss them; the application to fictionalism about chatbot minds below is the record’s.
Connections
- The metasemantic reply in developed form. Drayson names Block’s functional role semantics and does not develop it. THEORY-089 states Block’s two-factor account and says how it answers Toon’s “locked away” premise, with a qualification of Drayson’s wording.
- Fictionalism about chatbots. Roberts’s account (THEORY-461) is a fictionalism about AI self-ascriptions motivated by embodiment, not by skepticism about representations. Drayson’s dilemma carries over to any chatbot fictionalist who also wants chatbot talk to track the real causes of output; Roberts’s own anti-realism leaves belief open, which is where a functionalist realism would sit.
- Interpretationist realism. Dennett’s intentional strategy (THEORY-538) is an interpretivist realism of the kind footnote 20 grants, though Drayson does not cite it; Goldstein & Lederman (LIT-206) apply such a view to language models.
- The notebook. Toon’s pretence is a “memory-as-a-notebook game” (p. 10). Clark and Chalmers use the same image realistically (THEORY-579).