# Section 3 Machery role reconsideration — checkpoint 25 Mar **State**: The March 24 checkpoint recorded "Machery developed across two paragraphs, NOT compressed" as a decided item. This session reopened that decision from first principles — Nick explicitly asked not to treat prior adamance as evidence for current preference. The session involved: (1) reading Machery's Chapters 1 and 3 in full from the markdown extraction; (2) researching the exceptionalist counter-positions (Bealer 1998, Chudnoff 2013, Bengson 2015) via the SEP article on intuition, NDPR review of Chudnoff, web sources, and extraction of Bengson 2015; (3) working through what Machery's argument actually is, what the exceptionalists actually claim, and how the debate bears on Section 3. ## What Machery actually argues (from the source) Chapter 1 argues for "minimalism" about the method of cases: the judgements philosophers make when responding to thought experiments (Gettier cases, trolley problems, Twin Earth) are ordinary concept applications, not deliverances of a special faculty. His specific arguments: - No "faculty of intuition" — psychology and neuroscience have no place for one (lines 2002-2006) - Seemings that philosophers cite as evidence for irreducible intuitions can be explained as mere inclinations to judge, analogous to Müller-Lyer illusions (lines 2027-2070) - No common phenomenology across case-responses — "it is outlandish to assert that the judgments elicited by philosophical cases share a common phenomenology" (line 2212). But this passage is specifically about Bealer's claim of a phenomenology of NECESSITY, not about phenomenological character in general. - Actual philosophers don't describe their case-responses using exceptionalist vocabulary — Kripke says "We simply are not," not "I have a rational intuition that we are not" (lines 2154-2156) Chapter 3 argues that philosophical cases produce "cognitive artifacts" — unreliable responses caused by features of the case design (unusual scenarios, pulled-apart properties, entangled framing) rather than by the philosophical facts being tested. The cases have "disturbing characteristics" that make judgements about them systematically unreliable, and these characteristics are non-accidental — cases NEED to be unusual and property-pulling to do their dialectical work. ## What the exceptionalists actually claim Bealer (1998, "Intuition and the Autonomy of Philosophy"): Intuitions are "a sui generis, irreducible, natural propositional attitude that occurs episodically" — intellectual seemings analogous to perceptual seemings. They carry a phenomenology of necessity and provide non-inferential justification for philosophical beliefs. He defends "the Autonomy of Philosophy" — philosophy has its own independent epistemic source (rational intuition) distinct from empirical science. Chudnoff (2013, *Intuition*, Oxford): Intuitions have "presentational phenomenology" — when you intuit that p, it seems as if you're directly aware of the chunk of reality that makes p true. Modelled on perception: seeing red isn't just being caused to believe "that's red" — the redness is presented to you. Chudnoff thinks the same structure applies to intellectual intuitions. More Platonic than Bealer — direct intellectual apprehension of abstract facts. Bengson (2015, "The Intellectual Given," *Mind* 124): Intuitions and perceptions are at a certain level of abstraction the SAME type of mental state — "presentations." Both provide justification without requiring justification — they're "given." The strongest version of the perceptual analogy: not just that intuitions are LIKE perceptions, but that they share the same deep structure. This debate is genuinely live. These are not fringe positions — Bealer was a significant epistemologist, Chudnoff published with OUP, Bengson published in *Mind*. The minimalist side (Machery, Williamson, Cappelen) is equally serious but hasn't won. The SEP notes recent empirical work suggesting intuitions are "surprisingly stable across demographic groups" (Knobe 2021), which actually complicates Machery's reliability critique. ## How this bears on Section 3 The current draft uses Machery to DENY that intuitions exist as a distinctive mental state (¶9) and then redirects via cognitive artifacts (¶10). Problems identified: 1. Saying "Machery shows" treats a contested position as settled. He argues; he doesn't show. 2. The cognitive artifacts material (Ch. 3) is risky — Machery's own conclusion is that philosophical cases are FUNDAMENTALLY unreliable because their disturbing characteristics are non-accidental. The paper doesn't want this scepticism. Using cognitive artifacts to redirect ("the problem is textual quality, not phenomenological access") cherry-picks from an argument that goes further than the paper wants. 3. Two full paragraphs of Machery in a section Enrico wants "more distilled" is a lot of space for a secondary source. 4. The input/response distinction I initially drew (Machery is about responses, the section is about inputs) doesn't hold up — engaging with a philosophical case is one integrated act. The exceptionalist claim about intuitions IS a phenomenology-in-the-mind claim with the same shape as Zahavy's. ## The late-emerging alternative Instead of deploying Machery's deflationary argument, frame the intuitions objection as having the SAME SHAPE as the Zahavy embodied-simulation objection: - Zahavy: physics requires embodied simulation → LLMs lack it → LLMs can't do physics - Exceptionalists (Bealer, Chudnoff, Bengson): philosophy requires rational intuition / intellectual seemings / presentational phenomenology → LLMs lack it → LLMs can't do philosophy Same shape, same reply: whatever cognitive process generates philosophical insight, its contribution enters the discipline as propositional content — judgements about cases, stated reasons, arguments — and that content is in the corpus. Advantages over the current approach: - Doesn't take a side in the exceptionalist vs. minimalist debate — works even if Bealer and Chudnoff are right - Connects to the Merleau-Ponty point already in the section (an LLM can't originate the insight, but once articulated, it's available for further work) - Compact — Nick's instinct was "a straightforward small thing" - Machery could appear in a footnote for the deflationary alternative rather than occupying two paragraphs of main text ## New source acquired Bengson (2015) "The Intellectual Given," *Mind* 124, pp. 707-760. Downloaded to `Learning/generating-philosophy/The Intellectual Given by John Bengson 2015.pdf`. Saved to Readwise Reader. Extracted to `/tmp/bengson_2015.txt`. This is distinct from the Bengson, Cuneo, Shafer-Landau (2022) methodology book already in the project sources — the 2022 book deliberately brackets the intuitions question that the 2015 paper takes head-on. **Decided**: - Nothing decided — Nick explicitly said "I haven't decided what to do" - The previous decision ("Machery gets two full paragraphs, not compressed") is reopened for reconsideration **Rejected**: - "Machery shows there's no faculty" — the verb "shows" treats a contested philosophical position as settled fact - Framing Machery as about "responses" while the section is about "inputs" — this distinction doesn't hold up; engaging with a philosophical case is one integrated act, not separable input/output stages - Claiming "most philosophers accept the deflationary view" — unsupported assertion about disciplinary sociology - The specific claim that the exceptionalist view is a "minority position" — I have no evidence for this and can name roughly equal numbers of philosophers on both sides **Open**: - Whether to use the Zahavy-parallel framing for the intuitions objection (compact, doesn't take sides) or keep Machery's deflationary argument (takes a side, needs more space) - If using the parallel framing: how many sentences/paragraphs does this need? Nick's instinct was "a straightforward small thing" - Whether Machery appears in the main text at all, or only in a footnote - Which exceptionalists to cite — Bealer (1998) is the most cited, Bengson (2015) is the strongest version, Chudnoff (2013) is the most developed phenomenologically - How this decision affects the CEV's question (B) — does the three-question structure survive, or does it simplify?