Reason's No-Go Theorem against Materialism/Supervenience

Philosopher/independent researcher Catherine M. Reason's "no-go theorem"—subsequently, the “no-supervenience theorem”—is not a positive theory of what consciousness is in the ordinary explanatory sense but a negative claim about what consciousness cannot be. Building on papers by Gilbert Caplain (Caplain, 1995; 2000), Reason's theorem works by imposing a formal constraint on what any successful theory of consciousness can be, and does so via formal proof rather than by metaphysical argument. As she frames the project, the result "relies on a well-defined effective procedure rather than philosophical argument," identifying "a specific operation which is a necessary property of all healthy human conscious individuals — specifically the operation of self-certainty, or the capacity of healthy conscious humans to 'know' with certainty that they are conscious" (Reason, 2019). This operation, she argues, "is shown to be inconsistent with the properties possible in any meaningful definition of a physical system" (Reason, 2019).

Reason’s objective is to show that a whole family of theories—materialist/physicalist, supervenience-based, and perhaps some forms of dual-aspect monism or panpsychist physicalism—cannot account for all the properties of human consciousness. “The no-supervenience theorem limits the capacity of physicalist theories to provide a comprehensive account of human consciousness,” and its “resulting proof shows definitely that any physicalist theory which describes a self-aware, intelligent system must be internally inconsistent” (Reason, 2023). Her conclusion is stark: "for at least one aspect of human consciousness [i.e., materialism/physicalism], the mind–body problem is resolved" (Reason, 2019).

Self-Certainty

The target of Reason’s argument is not neuroscience as such. The key property is self-certainty—the Cartesian cogito ergo sum (I think therefore I am). The crucial capacity is that a conscious human subject can know, or at least assert with certainty, that it is conscious. This is operationalized by asking whether a system can determine that it is not in dreamless sleep, or can assert “I am self-certain” (Reason and Shah, 2021). While founded on the assumption of self-certainty, the No-Go Theorem does not deny that brains correlate with consciousness, that access consciousness is measurable, or that much cognition can be studied in third-person scientific terms. Rather, it challenges the commonly accepted, stronger thesis that conscious states—especially fully self-aware phenomenal states—are entirely determined by objectively observable physical states.

The argument begins from the Cartesian intuition, but Reason attempts to convert it into a formal limitation result. Suppose a physical system—Daniel’s brain, in Reason’s thought experiment — enters a state K if and only if Daniel is not in dreamless sleep. If Daniel knows he is conscious, he must also know that the physical process producing K is reliable. But that reliability judgment, if physicalism is true, must itself be implemented by another physical process. And the reliability of that further process must itself be established by yet another process. Thus, the physicalist explanation generates a recursive requirement: every physical process that underwrites conscious self-knowledge requires a further physical process to certify its reliability (Reason, 2024).

Reason and Shah frame the issue in terms of emergence and supervenience. Neuroscience often assumes that conscious experience is an emergent property of the brain, such that conscious macrostates arise from physical microstates. But they argue that this assumption faces a special difficulty because consciousness is “essentially non-measurable unlike all other proven emergent properties of physical systems” (Reason and Shah, 2021). Their conclusion is correspondingly radical: “conscious macrostates are not emergent from physical systems and they also do not supervene on physical microstates” (Reason and Shah, 2021).

Reason and Shah define a system as conscious if it has two properties: it "must be capable of human-like reasoning" and "must have a capacity for self-certainty, i.e., being able to assert its own conscious experience," where "this self-certainty must be absolute without any possibility of doubt and without depending on any additional assumptions" (Reason and Shah, 2021). They formalize this as the self-certainty function ω, which a system performs if it "is capable of observing that its own state at some time t is not NULL" (Reason and Shah, 2021).

The intuitive flow of the proof is a regress. If every mental process corresponds to some brain process, then the thought "I think therefore I am" corresponds to the claim that the relevant neural correlate "works correctly." But establishing that that correlate works correctly requires a further correlate, NCC*, which "cannot be the same brain state as NCC"; and "each time we deduce the need for a new state NCC*, we can deduce the need for a new state NCC**," so that "one can thereby induce an infinite number of distinguishable brain states, each different from every other" (Reason, 2023). Since "no brain could be large enough to include so many states," it "necessarily follows that any mind or other entity which can perform the Cartesian Cogito, cannot be a physical system" (Reason, 2023).

Reason and Shah render the regress as a pair of functions. To perform ω, a system must also perform a function ξ that certifies "the processes in X are physically real," and verifying ξ requires an observation function f that "is precisely ξ(ξ)" — so "we cannot perform ξ until we already know the result of ξ" (Reason and Shah, 2021). The result is Theorem 1a: "No physical system S(t) which is capable of human-like reasoning (assumed to be sound in principle) can perform the self-certainty function ω (omega) and neither can any such system be recorded by any external device as having asserted 'I am self-certain'" (Reason and Shah, 2021). Theorem 1b extends this to any model that supervenes on a physical system, since a supervening model would require the recording device "to be in two contradictory states at the same time" (Reason and Shah, 2021).

Infinite Regress

The claim is that what distinguishes the proof from ordinary infinite-regress arguments is its diagonal character. Reason and Shah note "a close relationship between the type of 'infinite regress' proof used in this article, and the technique known to mathematicians as a diagonal argument," observing that "an infinite regress is the result when a certain type of diagonal argument is applied to any set," much as "Turing's proof (1937) of the undecidability of the general halting problem is essentially a diagonal argument" (Reason and Shah, 2021). Reason's formalization builds this out using predicate modal logic and Wolpert's (2008) "inference device," a common mathematical structure for "all physical systems capable of observation, prediction and memory" (Reason, 2023). The formal payoff is that "any physicalist theory which describes a self-aware, intelligent system must be internally inconsistent" (Reason, 2023). Crucially, the contradiction "arises from the stipulation that real processes are objective"; because objectivity forces an unending sequence of physical processes (not merely theoretical states), "it is provably possible that whatever M 'knows' about the reliability of its own knowledge is, simply, wrong" (Reason, 2023).

Reason stresses repeatedly that the No-Go Theorem constrains theories, not nature: "the theorem applies strictly not to systems, but to theoretical models of systems… no physicalist model of a system can be [ω] omega-capable" (Reason, 2023). The argument also turns on a modal subtlety: the physicalism postulate "has the form of a material conditional: the argument depends only on the postulate being possible, not on its being necessarily true," and confusion arises "because the English language frequently confuses alethic [metaphysical] modality with epistemic modality" (Reason, 2024).

Consequences

Reason and Shah press the result toward a physical consequence. Because a conscious brain "can have a physical effect" on recording devices yet "cannot be described by any physical theory T," they conclude "there cannot exist any physical theory which fully describes the evolution of a conscious human brain," and therefore that "conscious human brains must violate the conservation of energy in some way" (Reason and Shah, 2021) — possibly via the energy–time uncertainty relation "a loophole of quantum mechanics" is the claim (Reason and Shah, 2021).

Reason’s No-Supervenience Theorem is aimed most directly at reductive physicalism, functionalism, computationalism, and identity theories. If consciousness is nothing over and above physical-functional organization, then self-certainty must be physically implemented; but assuming her proof is sound, Reason claims that such implementation cannot be made consistent for a self-aware rational system. No first-person fact of being conscious can be captured by any third-person objective system without circularity or regress.

The theorem also pressures non-reductive theories, by reframing the conclusion as an inconsistency among three postulates—objectivity, self-awareness, and rationality—showing that property dualism (Chalmers), reflexive monism (Velmans), realistic monism [panpsychism] (Strawson), and illusionism (Frankish) are each "ruled out in their current form," with Chalmers' theory surviving only if it "requires humans to behave irrationally" (Reason, 2024). The general point is that any theory accepting an “objectivity postulate”—that all mental states supervene on objectively observable dynamics—must give up either genuine self-awareness or human-like rationality.

Reason is careful to limit the scope of all this. A strict reading yields only that "there must be at least some states in any human-like conscious system which do not supervene on physical states" (Reason and Shah, 2021). Whether that is deemed a modest conclusion or its bolder extensions as to the nonphysical nature of consciousness depends, in the end, on "whether or not one accepts the Cartesian cogito as a valid basis for self-certainty," which "is a purely empirical question" that "cannot be decided by a purely theoretical argument" (Reason, 2024).