Does a Self-Model Get You Anywhere Near an Analogue of Consciousness?

The previous page built a self-model with two operating modes: opaque, sitting as a labelled entry Reason/Plan consults; transparent, feeding straight back into Perception where nothing downstream can inspect it as a model at all. This page checks that distinction against real theories of consciousness, and states plainly where the checking stops.

Transparency is Metzinger's term, not a coincidence

Metzinger's self-model theory of subjectivity argues that the phenomenal self is not a thing but the content of a self-model, and that the model being transparent — the system has no access to the fact that it is looking at a model rather than reality directly — is central to what makes that content feel like a first-person perspective rather than a represented object among others1. The opaque/transparent distinction on the previous page is not an analogy borrowed for convenience. It is the same distinction, built as an architectural switch rather than stated as a philosophical claim.

This gives the switch real content. An opaque self-model is available for the system's own reasoning to treat as a model — inspectable, revisable, argued with. A transparent one is not. Its influence is real and its status as a model is invisible from inside the system, exactly the structural signature Metzinger's theory names as necessary for something to feel like a perspective rather than a fact under consideration.

Global Workspace Theory needs no new mechanism

Baars' Global Workspace Theory holds that consciousness corresponds to content winning broadcast access to a shared, limited-capacity workspace that many otherwise-independent specialised processors can read from and write to2. This site's Multi-Agent Systems page already documents the mechanism this requires: shared-resource coordination, the same pattern behind Gelernter's Linda tuple spaces, applied there between separate agents. Applied inward, between a cognitive system's own modules — Perception, Abstraction, Episodic Memory, Reason/Plan — instead of between separate agents, it is the identical mechanism. Global Workspace Theory does not require a new architectural primitive for this series. It requires pointing an existing one at the inside of one system instead of the space between several.

Higher-order theories want the opposite of transparency

Higher-order theories of consciousness hold that a mental state is conscious when there exists a further, higher-order state representing it3. This is precisely what transparency denies. A transparent self-model's influence never becomes an object for a higher-order state to represent — it is folded into perception before anything could form a representation of it as a state at all. An opaque self-model is much closer to what higher-order theories require: an explicit representation, available to be represented about. The two operating modes on the previous page do not just differ from each other. They point toward two different, competing theories of what consciousness actually requires, and satisfying one moves away from satisfying the other.

Whether any of this needs explaining at all is contested

Chalmers separates the "easy" problems of consciousness — explaining reportability, attention, behavioural control — from the hard problem: why any of this is accompanied by subjective experience at all, rather than occurring with no experience whatsoever4. Nothing on the previous page or in this one touches the hard problem as Chalmers states it. Building a transparent self-model that produces certain functional behaviour says nothing about whether anything it does is like something to undergo.

That framing is itself disputed. Dennett's argument against classical qualia holds that once the properties qualia are supposed to have — ineffable, intrinsic, private, infallibly known the instant they occur — are examined directly, nothing satisfies the description, and the concept fails on its own terms rather than surviving as something science has yet to explain5. Frankish develops this into illusionism: there is no separate phenomenal fact requiring explanation, only the genuine fact that it seems as though there is one. He states the resulting shift directly — illusionism replaces the hard problem with the illusion problem, the problem of explaining how the illusion of phenomenality arises and why it is so convincing6. This is a live, actively defended minority position against phenomenal realists including Chalmers, not a settled result on either side.

The engineering target this leaves

Stated as an engineering brief rather than a metaphysical claim: build a system whose transparent self-model produces the same functional and reporting profile a human produces when claiming to experience something. Every position in the debate above agrees that profile is real and needs an account, including illusionists, who hold that the profile is all there ever was to explain. This series targets that profile. It does not claim building it settles what a phenomenal realist would still consider the further, harder question.

Where this connects

References

  1. Metzinger, T. (2003). Being No One: The Self-Model Theory of Subjectivity. MIT Press. ↩

  2. Baars, B. J. (1988). A Cognitive Theory of Consciousness. Cambridge University Press. ↩

  3. Rosenthal, D. M. (1986). Two concepts of consciousness. Philosophical Studies, 49(3), 329–359. ↩

  4. Chalmers, D. J. (1995). Facing up to the problem of consciousness. Journal of Consciousness Studies, 2(3), 200–219. ↩

  5. Dennett, D. C. (1988). Quining qualia. In A. J. Marcel & E. Bisiach (Eds.), Consciousness in Contemporary Science (pp. 42–77). Oxford University Press. ↩

  6. Frankish, K. (2016). Illusionism as a theory of consciousness. Journal of Consciousness Studies, 23(11–12), 11–39. ↩