The Hard Problem Isn't Hard Enough

Arguing that Chalmers undersells how strange consciousness is by framing it as a property, not a process

David Chalmers gave philosophy one of its most useful formulations when he distinguished the easy problems of consciousness from the hard one. The easy problems concern function: explaining how the brain integrates information, how it discriminates stimuli, how it produces reports about internal states. These are hard in the ordinary scientific sense, meaning they will take time and ingenuity to solve, but there is no conceptual mystery about why a physical system could, in principle, accomplish them. The hard problem is different. It asks why any of this processing is accompanied by subjective experience at all. Why is there something it is like to see red, to feel pain, to hear music? Why is there anything it is like to be a conscious creature, rather than simply a very sophisticated information processor that behaves as if it had experiences without actually having them?

This is a genuine and important distinction. It correctly identifies the explanatory gap that haunts all purely functional accounts of mind, and it does so with a clarity that earlier philosophy of mind, in its rush toward functionalism, often failed to achieve. Chalmers deserves the influence the formulation has had. But I think the framing contains a subtle distortion that makes consciousness seem less strange than it actually is, and that the distortion matters because it shapes which solutions look promising and which look beside the point.

The distortion is this: by asking why physical processes are accompanied by experience, Chalmers implicitly treats experience as a property that attaches to an otherwise characterisable physical process. Consciousness, on this picture, is something that certain neural events have, the way a surface has a colour or a wire has a voltage. The hard problem is then the problem of explaining why they have it. But I want to suggest that treating consciousness as a property is already to have made a move that prejudges the answer, and that the stranger and more defensible position is to treat consciousness as a process in its own right, one that does not attach to physical events but constitutes a different kind of occurrence altogether.

An apple that has the property of the colour red

What a Property Is

A property, in the philosophical sense, is something that an object or event has or instantiates. Properties are predicated of things that exist independently of them: the redness of the apple is something the apple has, and the apple would exist even if it were a different colour. When we say that a neural process has phenomenal character, has the quality of seeming like something from the inside, we are implicitly committing to a similar structure. The physical process is one thing; the experience is another thing that the physical process has or gives rise to.

This structure is what makes the hard problem feel so intractable. If experience is a property that certain physical events have, then we need an account of why those events have it and others do not. We need a bridging principle, a law or identity or supervenience relation that connects the physical description to the phenomenal one. And no such principle seems to be available without either trivialising the phenomenal (by reducing it to the functional) or mystifying the physical (by positing some new fundamental feature of reality that attaches to certain configurations of matter).

Chalmers himself eventually opts for something like the latter. His property dualism holds that phenomenal properties are genuine features of the world, not reducible to physical or functional properties, and that they are correlated with physical processes by fundamental laws of nature analogous to the laws that govern mass or charge. This is a coherent position, and it takes the explanatory gap seriously in a way that most physicalist accounts do not. But it inherits all the difficulties of treating consciousness as a property, chief among them the question of why those particular laws hold, and what it could even mean to say that a physical process has phenomenal properties as a brute matter of fact.

Consciousness may not be a feature that certain physical events possess. It may be a different kind of event altogether, one that the physical vocabulary was not designed to capture and cannot capture without distortion.

The Process Alternative

Consider an alternative framing. Rather than asking why physical processes have phenomenal properties, ask whether phenomenal experience might be better understood as a process, something that happens rather than something that obtains, and whether it might be a process that is not simply identical to, or correlated with, a physical process, but that intersects with physical processes in a way that neither reduces to the other.

This might sound like a merely verbal shift, but I think it has real philosophical content. A process, unlike a property, is not predicated of a substance; it is an occurrence, something that unfolds in time and has its own internal structure. Processes can be more or less fundamental; there is no requirement that they supervene on properties of the physical constituents that participate in them. And crucially, processes can be relational in a way that properties are not. The process of digestion is not a property that the stomach has; it is a dynamic relation between the stomach and its contents, and between the stomach and the organism of which it is a part. You cannot fully specify what is happening by looking at any one component in isolation.

If consciousness is a process of this kind, then the question changes. We are no longer asking why a physical event has phenomenal properties. We are asking what kind of process phenomenal experience is, what participates in it, what makes it the specific experience it is, and how it relates to the neural processes that are clearly implicated in it without being identical to it. These questions are no easier in one sense. But they are easier in another: they do not require a bridging principle between two fundamentally different ontological categories, because they do not treat the physical and the phenomenal as separate categories requiring bridging.

Where Chalmers Goes Wrong

The heart of my objection is this. Chalmers correctly observes that there is an explanatory gap between physical descriptions and phenomenal descriptions. He incorrectly concludes from this that consciousness must be a non-physical property that attaches to physical processes. But explanatory gaps do not always signal ontological novelty of the property kind. Sometimes they signal that we are carving the phenomenon at the wrong joint, that the conceptual framework we are using is not the right one for the thing we are trying to understand.

The history of physics offers relevant examples. Before the development of thermodynamics, heat was not well understood. It was sometimes treated as a substance, sometimes as a property of bodies, sometimes as a form of motion. The explanatory gap between macroscopic thermal phenomena and the mechanical behaviour of microscopic particles was not resolved by postulating that thermal properties were fundamental features of reality correlated with mechanical ones by bridging laws. It was resolved by reconceiving the phenomenon: temperature turned out to be mean kinetic energy, not because kinetic energy has temperature as a property, but because what we were tracking when we measured temperature was all along a feature of a process, the collective motion of particles, that the earlier vocabulary had misdescribed.

I am not suggesting that consciousness will turn out to reduce to a physical process in the way that temperature reduces to mean kinetic energy. The cases are importantly different, and I take the explanatory gap seriously. What I am suggesting is that the gap might persist not because consciousness is a non-physical property requiring bridging laws, but because the process that consciousness is has not yet been correctly characterised, and because our current physical vocabulary was developed to track a different class of phenomena and may be systematically unsuited to tracking this one.

The Strangeness Chalmers Misses

When consciousness is treated as a property, the hard problem becomes: why does this physical process have this property? This is a strange question, but it has a recognisable shape. Properties attach to things; we want to know why they attach in this case. The question has a form that invites certain kinds of answers, even if none of them turns out to be satisfactory.

When consciousness is treated as a process, the question becomes stranger and, I think, more honest. We are forced to ask what kind of occurrence it is, what the boundaries of the process are, where it begins and ends, what counts as the same conscious process continuing and what counts as a different one beginning. These questions reveal that consciousness does not have the stable, bounded character that a property typically has. It is not a feature that a brain state either has or lacks. It is something that is happening, and what is happening has a structure that is profoundly difficult to specify without already deploying phenomenal concepts.

The temporal structure of conscious experience is particularly revealing here. Experience does not present itself as a series of discrete states, each having phenomenal properties, one after another. It presents itself as a flow, something that William James was tracking when he coined the phrase stream of consciousness, but which he did not fully theorise. The present moment of experience contains what Edmund Husserl called retentions and protentions: a trailing edge of the just-past and a leading edge of the about-to-come. Experience is not a property of an instant; it is a process that spans time and that constitutes its own temporal structure rather than merely occurring within objective time.

This temporal character is something that property dualism has no good account of. If phenomenal properties are fundamental features of reality that attach to neural events, what accounts for the fact that those features are experienced as flowing rather than as a series of distinct states? The answer cannot simply be that the neural events form a temporal sequence, because that would be to explain the phenomenal character of temporal flow in terms of objective temporal succession, which is precisely the kind of move that the hard problem shows to be inadequate.

The question is not why this process has phenomenal properties. The question is what kind of process the phenomenal is, and whether our current vocabulary is even equipped to ask that question well.

Relational Consciousness and the Dissolution of the Gap

The move I am recommending connects to a broader tradition in philosophy of mind that has received less attention than it deserves, one that treats consciousness not as an intrinsic property of individual neural systems but as a relational phenomenon that arises in the interaction between a system and its environment, or between different aspects of a system, or perhaps between the physical and whatever it is that the physical stands in relation to when experience occurs.

Relational accounts of consciousness are not committed to any particular metaphysics. They do not require panpsychism, though some versions of them are compatible with it. They do not require substance dualism, though they resist the reductionism that treats consciousness as nothing over and above the physical processes that realise it. What they require is a willingness to take seriously the possibility that consciousness is a kind of occurrence that cannot be fully specified by reference to any one of its relata in isolation, that it is essentially a between-thing rather than a within-thing.

On this view, the explanatory gap that Chalmers identifies does not indicate that consciousness is a non-physical property requiring bridging laws. It indicates that our usual way of carving up reality into objects with properties misses something about the structure of conscious experience, that experience is the kind of thing that falls between the categories rather than fitting neatly into one of them. The gap is real; it is just not located where the property framing suggests.

This is a harder position to defend in detail than property dualism, because it requires developing a positive account of what relational occurrence means, and that account does not currently exist in satisfactory form. But I think the difficulty of developing the account is less troubling than the difficulty of making sense of why a physical process would have irreducible phenomenal properties. The former difficulty is a research problem. The latter is a conceptual one, and conceptual problems of this kind tend not to dissolve with more research; they require reconceiving the question.

A Harder Problem

The hard problem, properly understood, is not: why does this physical process have phenomenal properties? It is: what kind of occurrence is consciousness, such that it is neither fully captured by the physical vocabulary nor straightforwardly independent of it? This is a harder problem in the sense that it does not admit of the same kind of solution that the property framing invites. You cannot solve it by finding the right bridging laws, because the point is that bridging laws presuppose a relation between two already-understood categories, and one of the categories here is not yet understood.

What it requires, I think, is a new conceptual vocabulary, one that can describe processes that are neither purely physical nor purely phenomenal, that can track the structure of conscious experience without reducing it to neural correlates or elevating it to a separate realm of fundamental properties. Developing that vocabulary is the real philosophical work that remains to be done on consciousness, and it is work that will require contributions from phenomenology, from the philosophy of physics, from cognitive science, and from metaphysics, none of which can do it alone.

Chalmers gave us the hard problem, and that was a genuine contribution. What I am suggesting is that the hard problem, as he framed it, is a pointer toward a deeper difficulty that his framing does not quite reach. The strangeness of consciousness is not just the strangeness of a physical process having non-physical properties. It is the strangeness of an occurrence that does not fit comfortably into any of the ontological categories we have developed for thinking about what there is. That is a harder problem. It is also, I think, the right one.

The author is Chair of The Philosophical Society: Oxford and a researcher in philosophy of mind. This essay reflects work-in-progress on consciousness as a relational phenomenon. Published at magicinharmony.com.