Ned Block (1995), Behavioral and Brain Sciences 18(2), 227–247, a target article with open peer commentary and the author’s response (the issue’s pages run to 287). A corrected version is reprinted in Block, Flanagan and Güzeldere (eds), The Nature of Consciousness (MIT Press, 1997), which was not read.
Key takeaways
- Two concepts (§§3–4). P-consciousness is experience, which Block says he cannot define non-circularly and can only point to. He takes P-conscious properties to be distinct from any cognitive, intentional or functional property, while allowing that differences in intentional content often make a phenomenal difference. A state is A-conscious if, in virtue of having it, a representation of its content is poised (1) as a premise in reasoning, (2) for rational control of action and (3) for rational control of speech. A is a functional, system-relative and cluster notion; reportability is its least weighty element, so that chimpanzees can have A-conscious states.
- The fallacy (§6). Marcel, Flanagan, van Gulick, Schacter and Baars argue that because the thirsty blindsight patient does not reach for the water, phenomenal consciousness must function to let information guide action. Searle argues the same from Penfield’s petit mal patients. In blindsight both P and A are missing, so the failure to reach is explained by the missing access. In the epileptics nothing shows that P is missing at all. “The fallacy is: an obvious function of the machinery of access-consciousness is illicitly transferred to phenomenal consciousness.”
- What is right in the target reasoning (§6). A and P are almost always present or absent together, and superblindsight, A without P, seems not to exist. So P may be the gateway to access (Schacter’s model), may be epiphenomenal, or may be empirically the same as A. Making the distinction is what makes these alternatives visible.
- Possible P without A (§§4.2, 7). The drill one has heard all along but notices only at noon; Sperling’s arrays, where Block reports seeing all the letters but can report only about half; hypnotic analgesia; pain under general anaesthesia (aerodontalgia). Block says these “don’t show anything on their own”.
Standing in the record
Filed on 2026-10-03 at the owner’s request, with six later papers by Block on consciousness. No anthology topic holds a distinction between kinds of consciousness, and the paper carries no instruction for machine-learning practice.
It is the work the record’s readings mean when they write “Block 1995” or “Block’s access/phenomenal distinction”. Frankish, read in NOTE-143, places it in his “Cartesian” paradigm. Bruckner, read in NOTE-093, cites it for the definition of phenomenal consciousness. Block’s own later papers build on it. Block (2005), LIT-629, extends it from concepts to neural correlates, a phenomenal NCC and an access NCC.
Bearing on the new theories. THEORY-100 states as a limit that a report paradigm cannot separate access from phenomenal consciousness. That limit is this paper’s distinction, applied to report. The paper also shows why the limit runs one way: a report proves A, and it is silent on whether P can occur without A. THEORY-023 concerns attributing consciousness to AI. The robot “computationally identical to a person” whose silicon does not support P is this paper’s example of a full A-without-P zombie (§4.1). Block calls it conceptually possible and “very controversial”.
Against Dennett. The paper argues (§5) that Dennett’s claim that consciousness is a cultural construction is “trivially false” of P-consciousness, false of A-consciousness and “banal” of a sophisticated self-consciousness. It reads Dennett’s Multiple Drafts theory as a theory of A-consciousness. The record holds the article form of that theory, Dennett and Kinsbourne’s “Time and the Observer” (LIT-431). Block cites it in note 2 for the Cartesian materialism after which he names his “Cartesian modularism”. The paper does not engage that article’s argument, so no relation is declared here. The direct clash comes in Block (2007), LIT-646.
No relation is declared. The papers it argues against (Marcel, Searle, Flanagan, van Gulick, Schacter) are not held.