I’m a naturalist, an out-and-out naturalist, a philosophical or metaphysical naturalist, a naturalist about concrete reality. I don’t think anything supernatural or otherwise non-natural exists.

You can’t classify anything as supernatural or non-natural until you have a substantive conception of the natural in relation to which something can be classified as non-natural. I do have one: I take it that concrete reality – anything that exists in space-time – is entirely physical. I’m a physicalist naturalist, and I think metaphysical naturalism is the same thing as physicalism as I’ve just defined it (I’m putting ethics aside). There are, however, important questions to be raised about what this amounts to. They’re old questions, but they haven’t received enough attention recently. One result of this is that many – probably most – philosophers who call themselves naturalists are in fact extreme anti-naturalists. They’re false naturalists – noturalists.

Something very strange has happened to the use of the word ‘naturalism’ in the last fifty years or so, when it comes to the question of conscious experience (‘experience’ for short). W.V. Quine, who in his reductive passion and ontological austerity was seen as a standard bearer for naturalism in philosophy, never for a moment doubted or denied the existence of experience, as some of the false naturalists appear to do (even as they deny that they do). Quine never denied the existence of what he called ‘experience in all its richness … the heady luxuriance of experience’. No serious naturalist could deny it. Nevertheless, beginning in the 20th century, self-styled naturalists have seemed to think that questioning or doubting or denying the existence of experience is part of a thoroughgoing naturalism.

How did this happen? It began with the transmogrification of behaviourism early in the 20th century. Behaviourism started out in a highly fruitful way, as a moderate methodological thesis in psychology: roughly, that it’s not worth studying the phenomena of experience, although they obviously exist, because they’re not susceptible of scientifically rigorous quantitative treatment. It turned into a mad metaphysical thesis in philosophy, according to which there is nothing more to experience than behaviour and dispositions to behaviour: i.e., bluntly put, experience doesn’t really exist. Things got worse, especially after the publication of J.J.C. Smart’s paper ‘Sensations and Brain Processes’ in 1959. The denial of the existence of experience came to be thought of as naturalistic independently of any behaviourist (or ‘functionalist’) assumptions. The dubious existence or non-existence of experience was thought to follow simply from the fact – the view – that everything is wholly physical.

Here there is a wonderful irony, for the false naturalists – even as they doubt or deflate or deny the existence of experience, and revile Descartes, their favourite target, for being an outright realist about experience – are themselves in the grip of a fundamentally Cartesian conviction: the conviction that experience can’t possibly be physical, that matter can’t possibly be conscious. The irony is fierce because Descartes was at bottom aware that one can’t rule out the possibility that matter may be conscious. Many of the false naturalists, by contrast, have no such doubts.

Some of them will deny this. They will insist that they do admit the existence of consciousness or experience, and do allow that it can be physical. But they do this by changing the meaning of the word ‘conscious’ into something that involves no consciousness. They ‘looking-glass’ the term, by which I mean use it in such a way that whatever they mean by it, it excludes what the term actually means.

If physicalism is true, experience (consciousness) is wholly physical. For it certainly exists, and everything is wholly physical if physicalism is true. It follows that there are things about the nature of the physical that physics doesn’t characterise, because it doesn’t characterise the nature of experience. Some say that the word ‘physical’ is to be defined by reference to physics, so that if physics doesn’t characterise the nature of something, then it isn’t physical. But this is wrong. The word ‘physical’ is what is known as a ‘natural-kind’ term (like ‘iron’ or ‘water’), albeit of a highly general sort, and it’s generally allowed that we can be very wrong about the nature of anything denoted by a natural-kind term. So we may be very wrong about the physical. We can’t be sure we know the nature and limits of the physical. So we can’t be sure we know the nature and limits of the natural: we can’t be sure we know the nature and limits of the natural even if we’re right (I’m assuming we are) that the natural is the physical.

This is putting it mildly, because physics and cosmology are in turmoil. It’s not just that we don’t definitely know the nature and limits of the physical. We definitely don’t know the nature or limits of the physical. It doesn’t help much to say that we have a clear fix on the physical inasmuch as we know that the physical is the spatio-temporal, because we’re very far from clear about the nature of space-time. We may be very wrong about it, even after Einstein. A considerable number of physicists and cosmologists think that the description of reality as spatio-temporal is superficial.

It’s sometimes said that, for all our uncertainty, we have a pretty good fix on the basic nature of the physical. David Lewis once claimed that ‘the physical nature of ordinary matter under mild conditions is very well understood.’ But this isn’t true. It isn’t true even when we put aside the point that the known phenomena of experience are wholly a matter of the physical nature of ordinary matter under mild conditions (the mild if special conditions that obtain in the brain), along with the point that the physics and neurophysiology of the brain don’t enable us to understand how this is so. Even putting these things aside, we’re left with profound theoretical uncertainty about the nature of the ‘fundamental particles’, for example, and this uncertainty obviously extends to all conditions in which the fundamental particles are found, including all mild conditions. The same goes for gravity, ‘dark energy’, ‘dark matter’.

A physicalist conception of the natural has, nevertheless, a great deal of substance. I take it that vast numbers of the claims of current physics are either straightforwardly true or very good approximations to truth. The periodic table is on to something fundamental about the ultimate nature of concrete reality. So are formulae like F=ma, E=mc2, the inverse square law of gravitational attraction, and so on.

Note, though, that all these truths about the physical, outright or approximate, are expressed by statements of number or equations: mathematical equations featuring various constants in addition to various numbers and mathematical functions. They are truths about quantities and relational structures instantiated in concrete reality. This is the fundamental form of our theoretical knowledge of the physical in physics, a point well made by Eddington and then Russell, and one which applies not only to our knowledge of the physical but also to our ignorance of it. Physics may tell us a great deal about the structure of physical reality, but it seems that it can’t tell us anything about the intrinsic nature of reality in so far as its intrinsic nature is more than its structure.

I’m trying to approach a defensible version of physicalist naturalism about the mind by stressing a point that has never entirely disappeared ever since it was made so forcefully by Locke: our ignorance. One way to characterise this ignorance is to start from the Locke-Hume idea that when we observe causation, we can only ever observe regularity, not causal power as such. One can plausibly extend this into the – broadly speaking – Kantian view that when we make epistemic contact with any concrete reality X other than our own current experience, we can only ever have access to an appearance of X, an appearance that is necessarily a function not only of how X is, in itself, but also of how X affects us given how we are in ourselves, given, in particular, our sensory-intellectual constitution. (Kant doesn’t even exempt our access to our own current experience from this limitation.)

It’s plain that the human science of physics can’t fully characterise the nature of concrete reality, even in principle. Those who believe that physics can do this are not realistic or serious physicalists. Serious physicalists are clear on the Eddington-Russell point that physics can’t convey the nature of everything that exists – even though everything is wholly physical. On many matters, such as experience, physics is simply silent. If you’re not clear on this limitation, you have no idea what physics is. This isn’t New Age anti-scientism, it’s hardnosed physicalism.

‘What then do we know about concrete reality?’ When we ask this question we get, first, an old answer. We know, each of us individually, that we exist, as Descartes pointed out. But we also encounter another more general and more interesting certainty, as Descartes also pointed out: the fact of consciousness, the fact of the subjective, qualitative character or ‘phenomenological’ character of experiences: what I’m calling ‘experience’, or ‘the experiential’, or ‘experientiality’. The existence of experience is a certainly known natural fact (it is therefore, for any physicalist naturalist, a certainly known physical fact). This is acknowledged by almost all philosophers; only a tiny group of self-styled naturalists has ever denied it, beginning in the 20th century.

So now we have our starting point: outright realism about experience, conscious experience. This is where real naturalism starts from. What do I mean by ‘realism about experience’? I’m going to try to define the term ‘experience’ in such a way that I can’t be misunderstood. There’s so much terminological wreckage in the philosophy of mind that it’s close to collapse as a shared enterprise. This is why I’m inclined to answer the question of what I mean by ‘experience’ by saying that I mean real experience, and that by ‘realism about experience’ I mean real realism about experience. This isn’t facetious. The trouble is that some who claim to be realists about experience are really no such thing (they have looking-glassed the term).

What is it to be a real realist about experience? One way to answer this is that it’s to continue to take colour experience, say, or taste experience or pain experience to be exactly what one took it to be, quite unreflectively, before one did any philosophy: when you were six years old, for example, and were given food whose taste you didn’t like; or asked a friend what something tasted like, or whether they liked the taste; or turned in the direction of a bright light with closed eyelids and experienced reddish-orange, and then covered your eyes with your hands and watched the reddish-orange fade to black; or wondered, as children often do around that age, whether your brother or sister or friend saw colours exactly as you did yourself. Children of this age experiment constantly with experience. They squeeze their eyeballs to see double, rub their closed eyes to ‘see stars’, block and unblock their ears, spin round until they’re dizzy and then stop and watch the world swirling, fully aware that it’s their experience, not the world, that is swirling.

Children’s realism about experience (all non-philosophers’ realism about experience) is real realism about experience. However many new and astonishing facts real realists about experience learn from psychologists – facts about ‘change blindness’ or ‘inattentional blindness’ – their basic general understanding of what colour experience or pain experience is remains the same as it was before they did any philosophy. It remains, in other words, entirely correct, grounded in the fact that to have experience is already to know what it is, however little one reflects on it. To taste pineapple, in Locke’s old example, is sufficient, as well as necessary, for knowing what it’s like to taste pineapple.

I think this way of defining ‘experience’ is helpful because it guarantees that anyone who claims not to know what is in question here is being disingenuous. I’m talking about the lived character of experience. With this in place, one can say the following. If, as a physicalist naturalist, you think that naturalism or physicalism gives you any good reason to give an account of experience that is in any way deflationary or reductionist relative to the ordinary pre-philosophical understanding of experience, then you have gone wrong. You aren’t being a real realist about experience. It follows that you aren’t a real but a false naturalist – indeed, an anti-naturalist, because you are doubting or denying a known natural fact. So, too, you aren’t a real physicalist, because a real physicalist has to be a real realist about experience.

We can certainly try to question real realism about experience. We can try the hypothesis that experience is in fact illusory or unreal. In this case we admit, as everyone agrees we must, that it seems there is experience, while claiming that really there isn’t any. That, however, leaves the supposed illusion, the seeming, in place as a natural fact – hence a wholly physical fact. And now a familiar and irresistible argument refutes the hypothesis. The problem is that the illusion already possesses, by hypothesis, the very properties whose illusoriness is being hypothesised. The phenomenon of there seeming to be experience – the phenomenon we’re supposing to be an illusion – can’t exist unless there really is experience.

Daniel Dennett tries this move. He proposes that ‘there is no such thing’ as phenomenology: ‘There seems to be phenomenology … but it does not follow from this undeniable, universally attested fact that there really is phenomenology.’ In fact it does follow, for the reason I’ve just given: for there to seem to be phenomenology is for there to be phenomenology. When it comes to experience, you can’t open up the is/seems gap. Descartes makes the point. To suggest that the apparently sensory aspects of phenomenology (say) are in some sense illusory – in that they aren’t the product of sensory mechanisms in the way we suppose, but are somehow generated by processes of judgment or belief – is to put forward a surprising hypothesis about part of the mechanism of this rich seeming. It is in no way to put in question its existence or reality. Whatever the process by which the seeming arises, the end result of the process is, as Dennett agrees, at least this: that it seems as if one is having a phenomenally rich experience of (in his examples) green-gold sunlight, Vivaldi’s violin music and so on. And in this case, what seems, is. Consider pain.

It’s worth noting that all materialists until the 20th century were real materialists, where this means, crucially, that they were real realists about experience, and took experience so conceived to be wholly physical. Joseph Priestley, a strict materialist, holds in the 1770s that ‘the faculty of thinking is the result of a certain arrangement of the parts of matter’; that ‘sensation and thought do necessarily result from the organisation of the brain’; that ‘mind … is not a substance distinct from the body, but the result of corporeal organisation.’ This is also Hobbes’s view in 1641. It’s Regius’s view in 1647, Locke’s suspicion in 1689, Toland’s view in 1704, Collins’s view in 1707-8. It’s extremely widespread in 18th-century France, it’s old news in the powerful 19th-century movement in Germany that followed German idealism and was known as German materialism. There’s no reason to doubt that Democritus and other ancient materialists and atomists held essentially the same view.

The point about ignorance of the intrinsic nature of the physical doesn’t need to be expressed in the Eddington-Russell way. Suppose someone rejects Eddington-Russell, and claims that physics does give us knowledge that is not merely numerical-structural (they may claim that it gives, for example, some positively descriptive, non-numerical knowledge of the nature of properties like spin and charge, or indeed motion). We can grant the claim, for then a further point comes to the fore: our knowledge of the nature of these properties doesn’t give us any good reason to think that the physical can’t be experiential. Locke saw this, and Priestley, again, was very clear:

I define … matter … to be a substance possessed of the property of extension, and of powers of attraction or repulsion. And since it has never yet been asserted, that the powers of sensation and thought are incompatible with these … I therefore maintain, that we have no reason to suppose that there are in man two substances so distinct from each other, as have been represented.

So we know the experiential exists, but do we know that the non-experiential exists? No. As a physicalist naturalist, I take it that experiences are spatio-temporally located events, neural electrochemical goings-on, and I’m also going to assume for the moment that experiences have, as such (in having mass, charge, shape, size and so on), certain irreducibly non-experiential concrete properties. It is, however, important to stress that this is an assumption. It’s a highly substantive and unverifiable assumption. One has to grant that it’s an assumption as soon as one accepts some version of Eddington-Russell: that 1) there is a fundamental sense in which physics offers only abstract structural descriptions; 2) there must be something more to concrete reality than just abstract structure, something concrete that has or exemplifies the structure. As soon as one accepts 1) and 2), one must grant 3), that nothing in physics requires or entails that the structure-transcendent nature of concrete reality is or must be fundamentally or irreducibly non-experiential in character. And one mustn’t think that anything in one’s everyday conception of the physical can count against 3. That would be like thinking that our everyday experience of solidity in sitting on chairs, walking on the ground, bumping into doors and so on, can give us insight into the actual physical nature of solidity, i.e. electric charge. It can’t. ‘Kick at the rock, Sam Johnson, break your bones:/But cloudy, cloudy is the stuff of stones,’ Richard Wilbur wrote in ‘Epistemology’.

It may be claimed that the meaning of the term ‘physical’ is such that to say ‘X is physical’ entails ‘X has some non-experiential property’. But there is nothing in physics to warrant this claim. To claim to know with certainty that spatio-temporal extension entails non-experientiality is to claim to know more about space-time than is warranted by anything in science. One can re-express this by saying that physicalism is compatible with panpsychism. This makes it vivid, because many think it obvious that physicalism isn’t compatible with panpsychism. They think physicalism is not only incompatible with pure panpsychism (according to which all concrete phenomena have only experiential intrinsic properties), but also with the more moderate version of panpsychism (all concrete phenomena have experiential as well as non-experiential intrinsic properties).

They’re wrong. Serious discussion of the so-called ‘mind-body problem’ begins only on the far side of this mistake. Getting the point about our ignorance of the physical is a revolutionary experience in a standard Western philosophical career. It involves a massive intellectual jolt. (It spaced me out for about two weeks, back in the 1990s.)

I’ve argued that a physicalist naturalist can’t rule out the idea that the physical is wholly mental or wholly experiential. I have nevertheless, and for the moment, assumed that physical reality does have non-experiential concrete properties. One doesn’t also have to assume that it has experiential character, for this is a certainly known fact. Given that concrete reality is physical, we know that physical reality has experiential character. But we needn’t now rule out the possibility that panpsychism is false in all versions, and that physical reality has experiential character only when organised in certain specific ways – e.g. in the way in which it is organised in brains. Either way, all realistic physicalist naturalists respect the basic data of experience: most notably, experience itself. Real naturalism is accordingly in direct conflict with the wildly anti-naturalist doctrine now commonly known as ‘naturalism’, which has for the last fifty years or so treated its first and fundamental datum – experience – as if it were its greatest problem, and has tried to deal with it by questioning its existence, more or less covertly, or at least questioning its claim to be, in a fundamental respect, exactly as it seems, and indeed is.

On the first two occasions I made these points in public, I was told, with some heat, that nobody now disagrees, and that I was attacking a straw man. If this is true, I’m delighted. But I’m not actually sure that everyone does now agree. Here’s a test. Everyone who does agree must be fully open to panpsychism. This doesn’t yet seem to be so.

When self-styled naturalists argue from physicalism or naturalism to the non-existence of experience, or question its existence, the argument isn’t just wrong, it’s perfectly wrong. It gets things exactly the wrong way round. If we call experience ‘E’, and anything non-experiential ‘non-E’, then the correct argument is as follows.

1) If concrete reality contains something other than E that we as naturalists take to be a natural phenomenon, e.g. physical stuff conceived as something that is in its intrinsic nature non-E and that is (for this reason) such that we find it hard to understand how E exists as it does if non-E exists, given the highly intimate relation we suppose to exist between E and non-E, then non-E must be a problem for naturalism; but not E.

We are in this case in no position to say, as naturalists, that 2) non-E certainly exists, as a matter of natural fact, and it’s most unclear, given non-E and the relation we suppose to exist between non-E and E, how E is possible (and perhaps E is not possible). We may be in a position to say that 3) if non-E exists, as a matter of natural fact, then, given the relation we suppose to exist between non-E and E, it is most unclear how E is possible. But the next step must then be this: 4) E certainly exists, as a matter of certain natural fact; so if it is most unclear how E is possible, given the relation we suppose to exist between non-E and E, then it is presumably most unclear how non-E is possible, given that we know that E exists; and we have in fact no good reason to believe non-E is actual. We have in other words no reason to think that 5) physical stuff is wholly non-experiential in its fundamental nature. In fact we have no reason not to think that 6) physical stuff is wholly experiential in its fundamental nature.

6) may seem very strange, even after matter is acknowledged to be a form of energy, almost inconceivably insubstantial by our everyday lights, ‘as ghostly as anything in a spiritualist séance’ (Russell in 1927); but reasons of theoretical parsimony, simplicity and elegance tell in its favour. This is because, having committed ourselves to physicalism, we know that 7) some physical stuff is experiential (we know that common-or-garden physical stuff is experiential in nature under the mild if extraordinarily intricate conditions found in the brain). And given that some physical stuff is certainly experiential, 6) is the simplest hypothesis: all physical stuff is in its fundamental nature wholly experiential in all conditions and in all respects and all the way down.

6) is pure panpsychism. It makes a claim to be the most plausible version of hard-nosed physicalist naturalism. I call it ‘pure’ because it goes beyond the version of panpsychism according to which all physical stuff has experiential being in addition to non-experiential being. It’s important to see that it is wholly compatible with physics. It leaves everything that is true in physics untouched.

Im not particularly disposed to make the case for panpsychism here. I’ve simply argued that there’s no reason to favour the view that the energy patterns of which physics gives structural descriptions are fundamentally non-experiential over the view that they’re fundamentally experiential, especially given that we know that some of them are experiential in nature. Nor does the situation change if we allow that physics gives us more than purely structural information in furnishing us with concepts of spin, charge, motion and so on.

Some may still think that we have positive reason not to think that physical stuff is fundamentally experiential, because we have no positive evidence for its experientiality, except in the case of our own brains. But this is no positive reason to think that physical stuff other than that which constitutes our brains is not fundamentally experiential in nature, for we have no evidence for its non-experientiality either. If we ask what evidence there is for the existence of non-experiential concrete reality, the answer is easy and mathematically precise. There is zero evidence. There is zero observational evidence for the existence of any non-experiential concrete reality. Nor will there ever be any. All there is is one great big, wholly ungrounded, wholly question-begging theoretical intuition or conviction: we don’t see most of the matter around us doing things that we think of as showing signs of experientiality, so we conclude we know it’s not experiential, and that it’s ridiculous to think otherwise. This appears to be the great foundation of the wildly anti-naturalistic naturalism de nos jours.

I’m not arguing (like an old empiricist) that there’s a sense in which our knowledge can’t go beyond the experiential because experience is all we know for certain to exist. Our certainty that there is experiential reality, and our relative uncertainty that there is anything else, doesn’t in itself provide any grounds for an argument that there’s nothing but experiential reality. Nor does the fact that experientiality is the only concrete thing whose nature we have some direct acquaintance with provide any grounds for the view that everything concrete is experiential. Nor am I arguing that we have reason to suppose that the nature of concrete reality is wholly experiential via the claim that there’s a fundamental sense in which we have no positively descriptively contentful conception of non-experiential concrete reality, and so can’t even genuinely formulate the hypothesis that there is any non-experiential concrete reality.

Many will still think that it’s somehow ontologically or theoretically cheaper to postulate non-experiential stuff rather than experiential stuff as the basic stuff of reality. This is a mistake. It’s not cheaper. It’s either more expensive, because we know there is experiential reality and don’t know there is any non-experiential reality, and we now have to posit some kind of ‘radical emergence’ as a way of getting experiential reality out of non-experiential reality; or the cost is at best equal.

I end with a prediction and a challenge. The prediction is that no philosopher who disagrees will take any notice of this argument. I’m sorrowfully confident about this for a reason Hobbes gave in 1645, which has a vast amount of empirical support: ‘Arguments do seldom work on men of wit and learning, when they have once engaged themselves in a contrary opinion.’ The challenge is this. If you think anything I’ve said is anti-naturalist in any way at all, then you’re not a real naturalist. You’re not a serious, realistic naturalist. You haven’t got to first base.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.


Vol. 35 No. 20 · 24 October 2013

Galen Strawson can turn as many rhetorical backflips as he likes, but when he comes back to earth, there’s not a hair’s breadth of difference between ‘pure panpsychism’ and what the rest of us would call ‘supernaturalism’ (LRB, 26 September). Having excluded ethics from the realm of the ‘really real’, he can stipulate that experience is ‘wholly physical’, while folding ignorance and uncertainty about the nature and limits of the natural into that experience, precisely where the supernatural once was. This leaves him holding both ends of the definitional rope: it is not others who have ‘looking-glassed’ the terms, but his own verbal dexterity. He is welcome to say that physicalism and psychism come to the same thing, but since most of us use those terms to mark a distinction rather than an identity, it will be difficult to get others to play the game.

Rick Livingston
Ohio State University

‘If you think anything I’ve said is anti-communist in any way at all, then you’re not a real communist. You’re not a serious, realistic communist. You haven’t got to first base.’ If you swap ‘naturalist’ for ‘communist’ in this passage, you have the last three sentences of Galen Strawson’s essay ‘Real Naturalism’, which illustrates how naturalism functions now as an ideology. Strawson asserts his membership of the Naturalist Party and then seeks to determine the party line.

Strawson says, very properly, ‘You can’t classify anything as supernatural or non-natural until you have a substantive conception of the natural in relation to which something can be classified as non-natural,’ but then claims that he does have such a substantive conception: ‘I take it that concrete reality – anything that exists in space-time – is entirely physical.’ He denies, however, that what is physical can be defined by the science of physics. ‘Physics can’t convey the nature of everything that exists,’ he writes, leaving us no way of knowing what he means when he says that ‘everything is wholly physical.’

Clearly, what matters is the terminology: you’re a naturalist just so long as you say that ‘everything is wholly physical,’ even if you don’t know what ‘physical’ means. This is not an obviously ‘substantive conception of the natural’, so it remains unclear how we might distinguish between the natural and the supernatural or non-natural. Words like ‘natural’ and ‘physical’ function here not as descriptions of reality, but as symbols of group membership.

Christopher Eddy

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN


Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences