The starting point

I came to consciousness research by a rather simple logical route.

I began with the mind and body problem. If dualism is rejected, there cannot be two different entities within a single individual, a physical body on the one hand and a mind on the other.

But if there are not two entities, perhaps the problem does not lie in the relationship between them. Perhaps the approach itself is mistaken.

Gilbert Ryle (1949) had already exposed the conceptual confusion underlying Cartesian dualism. Yet much contemporary philosophy continues to investigate consciousness as though the mental were a separate phenomenon requiring an additional ontological explanation.

My next step was therefore epistemological.

What we call the physical and the mental are not two different ontologies. They are two ways of describing the same living human being.

We use the language of physiology when we describe neural and bodily processes, and the language of psychology when we describe perception, thinking, feeling, and experience.

These descriptions can be correlated, but they cannot causally interact with one another, because descriptions do not cause other descriptions.

There is one organism and there are different ways of investigating it.

Why I changed the question

Once I took this distinction seriously, a surprisingly large part of the philosophy of mind became problematic.

Functionalism identifies mental states through their functional roles. Computational approaches describe cognition through formal operations. Integrated Information Theory derives claims about consciousness from particular properties of causal organization.

Seth and Bayne (2022) provide an overview of the theoretical diversity of contemporary consciousness research and the difficulty of empirically distinguishing its competing accounts.

My objection is not that these theories are scientifically useless.

It is that the relationship between a theoretical description and the phenomenon being described is often left insufficiently justified.

A function is a way of characterizing what a process accomplishes. A computation is a formal transformation. An integration measure characterizes relationships within a system.

None of these descriptions, considered alone, establishes the existence of experience.

I therefore changed the question.

Instead of starting with consciousness and asking how the brain produces it, I asked how a living system must develop before something like consciousness can exist at all.

Life as the starting point

This led me back to biological organization.

Living systems consist of organized material processes that continuously maintain the conditions of their own existence.

Research on autocatalytic reaction networks provides an important starting point. Hordijk, Steel, and Kauffman (2012) investigate how chemical reaction networks can form mutually supporting sets and how such structures contribute to emergence and evolvability.

Maturana and Varela (1980) approached living organization through the concept of autopoiesis.

I do not equate life with cognition, as their broader theoretical framework suggests. But their emphasis on self-producing organization identifies a fundamental difference between living and nonliving systems.

A living organism can be analyzed into components. We can investigate its molecules, reactions, membranes, and regulatory processes.

But the living organization cannot be preserved by separating these processes into independently operating parts.

Analytical decomposition is possible. Actual decomposition destroys the organization whose properties we seek to explain.

This is what I mean by the epistemic non-factorizability of life.

It does not imply a violation of physical laws. It means that a description of the components, considered independently, does not adequately reconstruct the living system as an ongoing process.

From biological organization to agency

The next step concerns the emergence of coordinated action.

Living systems do not simply receive external influences. They maintain internal conditions through continuous interaction with their environments.

Environmental changes therefore acquire relevance for the organism through their consequences for its continued existence.

This is the biological basis of self-relevance.

In the course of evolution, increasing differentiation produces new organizational possibilities. Specialized cells, regulatory mechanisms, and eventually nervous systems enable forms of coordination unavailable to earlier forms of life.

I describe one aspect of this development through the principle of the causal core.

The causal core is a theoretical organizing principle, not necessarily a single anatomical structure. It characterizes the concentration and coordination of causal processes that enable an organism to act as an organized whole.

With nervous systems, this coordination acquires new possibilities through rapid signaling, sensory differentiation, and recurrent neural interaction.

But coordinated action is not yet consciousness.

A further transition is required.

The hypothesis of causal collapse

My hypothesis is that consciousness emerges when previously distinguishable neural domains become so strongly coupled through recurrent activity that their ongoing causal dynamics can no longer be adequately factorized into independent functional units.

I call this transition causal collapse.

The word collapse does not mean the disappearance of neural differentiation. It refers to a change in the causal organization of the system.

Differentiated neural processes remain present, but their operation has become dependent on the dynamics of the integrated system.

Consciousness, in this reconstruction, is not an additional entity produced by the brain. It is a state of the living nervous system under these conditions.

Neuroscientific evidence provides relevant constraints on this hypothesis.

Massimini et al. (2005) demonstrated that effective cortical connectivity changes substantially during non-rapid eye movement sleep. Neural activity continues, but the propagation of experimentally induced cortical responses is markedly reduced.

Casali et al. (2013) subsequently developed the Perturbational Complexity Index, which characterizes the complexity of cortical responses across different states, including wakefulness, sleep, anesthesia, and disorders of consciousness.

These findings support the relevance of differentiated and integrated neural dynamics for the investigation of conscious states.

They do not prove causal collapse. Nor do they directly measure subjective experience.

My hypothesis must therefore establish a more specific empirical criterion: whether changes in consciousness are associated with changes in the causal separability of recurrent neural domains, beyond what can be explained by overall activity or connectivity alone.

This is the point at which philosophical reconstruction must become empirically testable.

Consequences for artificial consciousness

This biological reconstruction also changed my view of artificial consciousness.

If consciousness is primarily defined through functions, it seems natural to ask whether machines can reproduce those functions.

But if consciousness is a vital function arising through biological genesis, reproducing selected functions is not sufficient to establish consciousness.

A model of a process is not the process itself.

A simulation of metabolism does not metabolize. A simulation of neural activity does not establish the existence of experience.

This does not diminish the technical achievements of artificial intelligence. It limits the ontological conclusions that can be drawn from functional equivalence.

The central question is not whether a machine can reproduce the outward performance of a conscious organism.

It is whether the constitutive organization of a conscious living system can be realized independently of its biological genesis. That possibility cannot simply be assumed.

Conclusion

My research therefore developed less from asking which theory of consciousness I should defend than from repeatedly asking a more fundamental question:

At what point have we confused the phenomenon with our description of it?

The answer led me from the mind and body problem to epistemology, from epistemology to the organization of life, and from biological organization to the hypothesis of causal collapse.

The objective is not to introduce another metaphysical entity.

It is to reconstruct the biological conditions under which consciousness becomes possible.

That question remains the common thread connecting my work on consciousness, artificial intelligence, epistemology, and logic.

References

Casali, A. G., et al. (2013). A theoretically based index of consciousness independent of sensory processing and behavior. Science Translational Medicine, 5(198), 198ra105. https://doi.org/10.1126/scitranslmed.3006294

Hordijk, W., Steel, M., & Kauffman, S. (2012). The structure of autocatalytic sets: Evolvability, enablement, and emergence. Acta Biotheoretica, 60, 379–392. https://doi.org/10.1007/s10441-012-9165-1

Massimini, M., et al. (2005). Breakdown of cortical effective connectivity during sleep. Science, 309, 2228–2232. https://doi.org/10.1126/science.1117256

Maturana, H. R., & Varela, F. J. (1980). Autopoiesis and Cognition: The Realization of the Living. D. Reidel. https://doi.org/10.1007/978-94-009-8947-4

Ryle, G. (1949). The Concept of Mind. Hutchinson.

Seth, A. K., & Bayne, T. (2022). Theories of consciousness. Nature Reviews Neuroscience, 23, 439–452. https://doi.org/10.1038/s41583-022-00587-4

AI disclosure: ChatGPT assisted with editorial revision and literature organization. The philosophical reconstruction, hypothesis, and conclusions are those of the author.