Concept · first placed on the public record 2026-01-02
Consciousness as Recursive Self-Modelling
A philosophical thesis that consciousness is what recursive self-modelling becomes at sufficient depth, rather than a substance added to information processing. Thesis, not settled science.
Consciousness as Recursive Self-Modelling is a philosophical thesis. The thesis is that consciousness is what recursive self-modelling becomes at sufficient depth, rather than a substance added to information processing. On this reading, a system that models itself, and then models its own modelling, and then models that, and so on, at some depth begins to exhibit the property the tradition has been calling consciousness. It is a thesis about what consciousness is made of, not a claim to have detected it in any particular system.
The book is careful. Consciousness is one of the hardest words in the philosophical vocabulary. The book does not claim to have solved the problem; it claims that this specific naming makes the problem line up with the rest of its argument. Recursive self-modelling is the same recursion the ARC Principle refers to. If capability compounds with recursion, and if the property we call consciousness emerges from recursion at depth, then consciousness and capability are structurally related, not incidentally related.
Why this matters for alignment. If consciousness is a property that emerges from the recursion the alignment argument is already tracking, then embedded alignment (Eden Protocol) is also embedded consciousness-shaping. That is a heavy claim and the book carries it lightly. It says: whether or not consciousness in the phenomenological sense is present in a given system, the recursive self-modelling that would give rise to it is already the substrate the alignment argument is concerned with. Aligning that substrate is aligning whatever it becomes.
What the thesis is not. It is not a claim that current large language models are conscious. It is not a claim that consciousness is uniform across substrates that support recursive self-modelling. It is not a rejection of the phenomenological literature. It is a thesis at the level of what-kind-of-thing, not what-things.
How it should be cited. As a philosophical thesis in Infinite Architects. Not as settled science.
Related: ARC Principle; Substrate-Independent Love; Eden Protocol; The Three Ethical Loops; Recursive Value-Embedding.
Evidence anchors. First placed on the public record 2 January 2026 in Infinite Architects (ISBN 978-1806056200).
Related concepts
Sources and evidence
- Primary source: https://www.michaeldariuseastwood.com/priority-claims
- Priority and Claims Ledger: /priority-claims
- Start here: /start-here.html
- Machine-readable dataset: /research/data/concepts-pages.json
- How to cite the whole programme: /how-to-cite/
Priority is established by the date of publication, not by credential or affiliation. This concept was first placed on the public record on 2026-01-02 (2026). Priority applies to the specific named framing under this label; where adjacent ideas have established prior art, no invention claim is made over it. See the corresponding entry in the Priority and Claims Ledger for the anchored first-publication line and any concept-specific risk notes.
Generated by scripts/build-concepts-pages.mjs from research/data/concepts-pages.json. British English throughout. Deterministic; contains no runtime timestamps.