Performative Self-Reference and the Metalogical Impossibility of Computational Consciousness: A Formal Axiomatic Development
PhilPapers (PhilPapers Foundation) December 6, 2025 DOI: 10.5281/zenodo.17839032 (opens in new tab) via OpenAlex
Summary
AI-generated from the abstractConsciousness cannot be identified with any purely computational structure, because any computational description treats it as an object, yet the subjective standpoint of an experiencing subject is logically prior to and presupposed by any such description. A minimal formal axiomatic system is introduced that strictly separates a subject-type S (experiencing consciousness) from an object-type O (all computable, objective states). Within this framework, three theorems are demonstrated: it is logically impossible to reduce consciousness to computation; any attempt to deny or explain away the subject already presupposes it; and the subject is ontologically prior to all objective manifestations. Machines are not excluded from consciousness because they are too simple, but because any wholly computational description can only model consciousness as an object, never capturing the subjective standpoint required for such modeling.
Study at a glance
| Characteristics | Theoretical or philosophical paper Peer reviewed |
|---|---|
| Keywords | Impossibility Performative utterance Subject documents Development topology Axiomatic system |
| Key finding | Argues that consciousness cannot be identified with a purely computational structure because any computational description models consciousness only as an object and cannot capture the subjective standpoint required for such modeling. |
Abstract
This work supersedes and radically extends the author’s earlier publication [DOI Link-1]. This paper introduces a minimal formal axiomatic system that strictly distinguishes a subject‑type S (the experiencing consciousness) from an object‑type O (all computable, objective states). Within this framework, it shows: (T1) It is logically impossible to identify consciousness with a purely computational structure. (T2) The subject‑type S is unavoidable: every attempt to deny or explain it away already presupposes S in the very act of critique. (T3) S is ontologically prior to all objective manifestations O: whatever is described as an “object” already assumes a subject that does the describing. This reframes the question of “artificial” or “computational” consciousness: machines are not excluded because they are too simple, but because any wholly computational description can only model consciousness as an object, and thereby can never logically capture the subjective standpoint that is required for such modeling in the first place.