I’ve Been Thinking 
by Daniel Dennett.
Allen Lane, 411 pp., £30, October 2023, 978 0 241 51927 1
Show More
Show More

Mydaughter has Pauline Baynes’s map of Narnia hanging above her bed. It’s a lovely object, produced in 1972 as a promotional poster for Puffin Books. One of its pleasures is tracing the way locations from the different stories fit together into a whole. There’s the island they sail to in The Voyage of the Dawn Treader; and those are the marshes they cross in The Silver Chair.

Some stories are harder to integrate than others. On my first visit to India I was introduced to some cousins who were apparently related to my mother on both her mother and her father’s sides of the family. It was confusing: I couldn’t see a way to fit the facts together. It was only once I realised what happens when two brothers marry two sisters that all became clear.

The philosopher Wilfrid Sellars thought that philosophy involved the reconciliation of disparate facts. Consider the disconnect between what he called the ‘scientific’ and the ‘manifest’ images of the world. Science describes a domain of fundamental particles situated in fields of force, spread out across a four-dimensional spacetime. Ordinary human life contains conscious creatures who make decisions, puzzle over problems, and find things good and beautiful. How do we integrate these two stories into a unified whole? One of the tasks of philosophy, Sellars said, was ‘to understand how things in the broadest possible sense of the term hang together in the broadest possible sense of the term’.

Daniel Dennett, who died in April, spent much of his career examining matters that populate ordinary thinking about the mind: beliefs, pain, consciousness, free will, the self. He sought to reconcile these aspects of the mind with the scientific story, to show how they can be integrated with such things as neural pathways carrying information, molecules moving according to mechanistic laws, the fundamental particles of atomic physics. Most important, Dennett’s aim was to show how all this can be achieved without appealing to anything supernatural.

Descartes didn’t believe the mind could be accommodated by the mechanistic stories of science. Instead he posited an immaterial substance, distinct from body, in which thinking was to be found. Contemporary philosophers known as ‘panpsychists’ ply the same trick with consciousness: since it can’t be reconciled with science, consciousness must be smeared across the universe like a magnolia undercoat, adding a basic form of experience to each and every fundamental particle. Dennett had no time for this sort of thing: no magic allowed.

Dennett’s supervisor at Oxford was Gilbert Ryle, the commanding presence of mid-20th-century Oxford philosophy. Ryle had argued in The Concept of Mind (1949) that Descartes’s account of immaterial substance was a bad answer to a silly question. Questions about the way the mind relates to the body assume that the mind is a thing which we need to locate in the material world. But this is a case of language bewitching. We can talk about doing something for another person’s sake without thinking that we need to locate sakes in the material world. We can talk about people who can do sums in their heads without thinking that mental arithmetic is inside the skull in the same way that the occipital lobe is inside the skull. The elements of mind are not denizens of an inner realm but simply a set of abilities, such as being able to answer maths questions without pen and paper.

Dennett wore Ryle’s influence with pride. His doctoral thesis, later published as Content and Consciousness (1969), set out the basic approach to reconciliation he maintained throughout his career. Start with ‘content’. I believe the next train is for Manchester, and I hope there will be a seat available. Philosophers like to say that these states – belief and hope – exhibit ‘intentionality’, which is a fancy way of saying that they are about certain things. But ‘aboutness’ doesn’t look like one of the properties that feature in the scientific story of the world. Fundamental particles have spin, they have charge, but they’re not about anything. So where do we find intentional states like beliefs and hopes in our scientific stories?

One option is to ‘reduce’ beliefs to something within those stories. Perhaps beliefs are simply complicated states of the brain, and when I believe the next train is for Manchester, that belief is represented in a language of thought. On such a view, it remains true that I believe the next train is for Manchester. But that’s not because there is some immaterial substance doing the thinking: it is because there is a state of my brain that is organised in some particular way.

An alternative means of reconciliation is to eliminate the idea of beliefs altogether. Perhaps they are relics of folk psychology, to be dispensed with once we have a good scientific story about neural states. This is the attitude that some of us take to some aspects of alternative medicine. I have a friend who swears by the benefits of healing massage. Her therapist hovers her hands above my friend’s body to facilitate the flow of life energy. Can this life energy be reconciled with our scientific stories? I’m sceptical: we just don’t find ‘life energy’ in our best biological accounts of the way human bodies work. Eliminativists say the same thing about beliefs: they’re a nice fiction but not something that stands up to scientific scrutiny.

These two options – reduction and elimination – serve as fixed barriers between which Dennett tries to navigate. He doesn’t want to say that our familiar mental phenomena are mere fictions – at least, not entirely so. But neither does he think they can be straightforwardly reduced to any specific part of the scientific story. In particular, the reductionist option for belief makes it seem as if the question of what someone believes is like the question of what type of blood they have – a perfectly objective matter, which could be settled by looking inside their brain. But belief isn’t like that. Sometimes it’s fuzzy, sometimes it’s a matter of perspective. Does my daughter believe that the king is the head of state? Well, does that require her to distinguish the state from the government? Or that she understand how the United Kingdom differs from Great Britain? Determining what someone believes is often more like determining whether Hamlet is an investigation of political power. Sort of. It depends.

Instead, Dennett introduced the idea of the ‘intentional stance’. This is the attitude we take towards a system when we view it as having the beliefs and desires that it ought to have, given its place in the world. Think about the ease with which we navigate a crowded street on a rainy day. We assume that people want to avoid getting wet, and that they believe they’ll get splashed if they walk too close to the kerb. That gives us a good basis for predicting the route they will take. Very often, our prediction will be correct. Of course, it’s rare that we actually deliberate over such predictions, but Dennett thought they were ubiquitous in interpersonal relations. We take people to have the beliefs and desires they ought to have and, by doing so, we are able to predict their behaviour with remarkable accuracy.

It is this predictive power which, in Dennett’s view, tells against any attempt to eliminate beliefs. How could mere fiction do such a good job at forecasting the movement of people across space and time? Think of the ease with which I can predict where you will be a week from now, simply by making an arrangement to meet you. I impute to you a desire to see me and a belief about where you should be at what time in order to make this happen, and the prediction is easy. Dennett’s striking claim is that this is all there is to having beliefs. If there is a good predictive model that works by ascribing beliefs and desires to you, then those are the beliefs and desires you really have. Having a belief is a bit like possessing a centre of gravity, something not explained by some inner mechanism but through the role it plays in predictive explanation.

One obvious concern about Dennett’s approach is that it extends the class of believers too widely. If I’m playing chess against a computer, I might think about what it believes or wants, and on that basis make good predictions about its behaviour. Maybe I can even predict what my alarm clock is going to do by thinking of it as believing it is 7.30 a.m. and that it wants to wake me up. So long as there is a good pattern of prediction here, why shouldn’t we think of these systems as believers? The fuzziness of belief allows for generosity of application. This isn’t quite reduction but isn’t quite elimination either.

What about consciousness? Here Dennett veers slightly closer to the eliminative barrier. On his view of content, what you believe is determined not by the goings on inside your head but by the patterns in your behaviour. This is good Rylean dogma, and Ryle’s opponents called it behaviourism. (Not Ryle, though. ‘There is no place for “isms” in philosophy,’ he wrote. ‘To be a “so-and-so-ist” is to be philosophically frail.’) But behaviourism seems doomed as an account of consciousness. Surely there is something it is like to smell coffee, to hear music, to see an impressive sunset. How can the technicolour of experience be reduced to mere patterns of behaviour?

It is the difficulty of reconciling this type of consciousness with scientific accounts that pushes some philosophers to add non-physical stuff. But Dennett regards this as just one more appeal to the supernatural. There are no spooky properties that make mental states conscious; nor is there some special theatre in the brain where the conscious presentation of one’s life is staged. Rather, there are lots of information-processing streams in the brain that construct multiple narratives about what is happening. Different narratives achieve prominence at different times depending on the stimuli involved. But none is canonical. Dennett concludes from this that there are no fixed facts about consciousness. We have explained consciousness when we understand why some narratives come to the fore and others do not. Nothing more is required.

For Dennett’s critics, this is equivalent to denying the reality of consciousness. (One reviewer complained that his book Consciousness Explained, from 1991, should have been called ‘Consciousness Ignored’.) But Dennett didn’t think he was denying the existence of consciousness, only the silly things philosophers have said about it. As to whether he was a behaviourist, he claims in I’ve Been Thinking, the autobiography published six months before he died, that ‘science is a sort of behaviourism; once you’ve got a scientific explanation of all the behaviour, inner and outer, large and microscopic, of any phenomenon, there’s nothing else to explain – except why some people are so uncomfortable with your explanation!’

Philosophers​ often think of themselves as retailers of argument. But Dennett’s primary mode of engagement is the story. One famous paper begins: ‘Several years ago, I was approached by Pentagon officials who asked me to volunteer for a highly dangerous and secret mission.’ The mission involves his brain being removed and placed in a life-support system, from where it continues to relay information to and from the rest of his body. Imagine that. Or imagine people who speak a language just like English, except that when they are tired they talk of coming down with ‘the fatigues’. Or that two coffee tasters have come to dislike the house blend, but one of them says the taste has changed while the other says he no longer likes it. What should we say about the location of the self, the nature of ‘fatigues’, and the modes of consciousness in these respective scenarios?

Dennett calls philosophical stories ‘intuition pumps’ and sometimes seems to suggest that they are signs of philosophical weakness, wheeled in at the point in a piece of reasoning where the argument has run out. (One of Dennett’s most useful insights for anyone reading philosophy: stop whenever you see the word ‘surely’ – it is usually a sign that you are being nudged towards a conclusion without anything by way of support.) But it’s more accurate to say that he is wary of the unselfconscious use of intuition pumps. Just as scientists need to understand their tools in order to understand the results they get by using them, so too a philosopher should be attentive to the use of intuition pumps, to their benefits and limitations.

Dennett’s use of stories is carefully controlled, right up to its seeming naivety. Many philosophers think there is something slippery, perhaps even irresponsible, about his work. One wants a clear answer to a question – do you or do you not think that beliefs exist? – and he offers you a story instead. But it’s helpful to view this tendency as part of a therapeutic tradition in philosophy that Dennett inherited from both Ryle and Wittgenstein. If, as Wittgenstein claimed, the point of philosophy is to show the fly the way out of the fly bottle, then what good is a carefully constructed argument? Better to interrogate the reason for asking the question in the first place. Better to tell a more compelling story.

Dennett’s stories have a dialogical function: they’re a way to get you to see the truth, but not by means of argument. This also explains the variety of intuition pumps one finds in his writings. Some philosophers offer multiple arguments for a view, as if stacking them up somehow made for a more convincing case. But if you have just one good argument, nothing more is necessary; and if your arguments are bad, it doesn’t matter if they are one or many. Stories, by contrast, can strike different people at different times in different ways. You don’t like the story about coffee tasters? Then try imagining a scientist who plays with your memory of colours; or someone who thinks that the dollar is the only real currency; and so on, until you find a story that lands.

One result of this use of stories is that Dennett is resolutely present in his philosophical writing. It’s a long way from standard academic prose. The style is gruff, personal, sharp – the regular at the bar rather than the performer at high table. In the autobiography, he writes in a similar way about his academic life, filling in the personal and intellectual backdrop to the philosophical writing. Here his stories are intended to be exemplary, showing ‘the secrets of my success, my good tricks and policies, my ways of dealing with people and problems’. These involve a healthy disrespect for disciplinary boundaries and an eye for interesting empirical results. As a graduate student in Oxford, Dennett found his own way to experimental psychology, with the tolerance if not quite the blessing of Ryle. His intellectual and institutional lives have ranged freely across neuroscience, psychology and computing.

In 1965, aged just 23, Dennett was appointed to the new Irvine campus of the University of California, on Ryle’s recommendation, and taught its very first class, on Descartes’s Meditations. He later moved to Tufts, where he remained until his retirement in 2022. Many of his stories involve collaborators and interlocutors, and some antagonists too. He claimed to take pleasure in seeing his ideas ‘rediscovered, reinvented, by philosophers over the past half-century without any acknowledgment’, yet the autobiography is punctuated with the repeated staking of claims and settling of scores. He proudly quotes a philosopher saying of him: ‘Dan believes modesty is a virtue to be reserved for special occasions.’ It shows.

Dennett’s early childhood was spent in Beirut, where his father, a historian of early Islam and an intelligence officer, became the first CIA operative to be killed in action. For forty years he had custodianship of a farm in Maine, where he spent his summers fixing barns, making cider champagne and ploughing the fields – doing ‘tillosophy’, as he puts it. There is something of the handyman too about Dennett’s approach to philosophy proper – a confidence that we can make progress on philosophical questions by getting a grip on the details, and an irritation with those who, it seems to him, prefer easy solutions to the complicated work of getting it right.

This approach goes hand in hand with his naturalism. There can’t be anything mysterious about the mind because its contours must have been formed through a purely naturalistic process. Thus his attraction to natural selection – ‘the single best idea anyone ever had’, as he put it in Darwin’s Dangerous Idea (1995). Through natural selection, heritable traits that enable animals to survive and reproduce build up in a population over time. We can tell evolutionary stories, then, to show why creatures come to display the complex patterns of behaviour that make them predictable within the intentional stance, with no appeal to miracles.

Dennett was active in the debates about religion that surfaced in the wake of the 9/11 attacks. Breaking the Spell (2006) was an attempt to explain organised religion as the natural outcome of evolutionary pressure. It was grouped together with books by Christopher Hitchens, Sam Harris and Richard Dawkins to form the canon of ‘new atheism’. I’ve Been Thinking opens with an account of the open-heart surgery Dennett underwent following a near fatal heart attack. To reduce the risk of mini-strokes, his surgeons reversed the flow of blood to his brain, flushing out bits of debris: quite literally, as he notes with pleasure, brainwashing him. Once he’d recovered, he published a short piece called ‘Thank Goodness!’ in which he contrasted the critical thinking of science with the dogmatism of religion. ‘I have had to forgive my friends who said that they were praying for me,’ he wrote. What they were doing was ‘morally problematic at best’.

Dennett’s​ approach to reconciliation was shaped by his naturalism. But why think there’s any need for reconciliation in the first place? The wish to pull narratives together into a unified whole is often quixotic. It may be necessary where we have some reason to think that our stories are in conflict and that the conflict can only be abated by finding some larger story into which they can be fitted. But Dennett makes clear that this is not true of the manifest and scientific images of the mind. There is nothing in the scientific story that is in itself at odds with mentality. Whether it is or not can only be established by additional claims from outside science – for instance, the claim that beliefs are real only if they are identical to representations inside your head. It’s true that physics doesn’t tell us that there are such things as beliefs. But neither does it tell us that there are no such things. It is simply silent on the question.

‘It seems to be too readily assumed,’ J.L. Austin wrote in 1956,

that if we can only discover the true meanings of each of a cluster of terms … then it must without question transpire that each will fit into place in some single, interlocking, consistent conceptual scheme. Not only is there no reason to assume this, but all historical probability is against it … We may cheerfully use, and with weight, terms which are not so much head-on incompatible as simply disparate, which just do not fit in or even on.

The scientific and manifest stories of the mind are not incompatible, but they are disparate. Why think we have to make them fit in? At least part of the answer for Dennett was his distrust of the miraculous. He seems to have thought that any view which fails to reconcile the mental and the scientific is thereby committed to treating the mental as supernatural. But this is to assume that the natural is exhausted by the scientific, as if there were no more to nature than that which science and science alone allows. If we don’t make that assumption, we don’t have to worry about fitting the mental and scientific together into an integrated whole. The mind’s status as natural is not beholden to successful reconciliation.

Marvel Comics used to offer a contest with no prize for readers who found a way to make events across the comics fit together in spite of apparent continuity errors. Over time this morphed into the No-Prize, mailed out to the winner in an empty envelope. Ryle would have loved that shift and the confusions it brought about. (One recipient wrote to Marvel to complain about the empty envelope, imagining that the prize must have gone missing en route.) If there were a No-Prize for reconciling our scientific stories about brains, biology and evolution with the rich topography of mental life, then Dennett would surely have been in the running. But when there are no prizes at stake, why play the game?

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. 46 No. 13 · 4 July 2024

Anil Gomes says that Daniel Dennett was a naturalist, and many would agree with him (LRB, 20 June). They’d be wrong, since Dennett denied the existence of the only wholly natural thing of whose existence we are absolutely certain: consciousness, consciousness as ordinarily understood, sensory experience, emotional experience, pain, experiential qualia of any sort – including the experience of reading this sentence. To be a naturalist, a real naturalist, you have to acknowledge the existence of consciousness. The theoretical physicist Lee Smolin puts it well when he says that qualia must be understood ‘as aspects of nature. That is our commitment to naturalism – the philosophy that asserts that all that exists is part of the natural world science studies.’

Dennett had no such commitment. Like most of those who today call themselves naturalists, he was a false naturalist. He thought – astonishingly – that science, and physics in particular, ruled out the existence of consciousness (as ordinarily understood). He never got the point that Bertrand Russell made so many times: that while physics tells us a great deal about the structure of the world, it is silent on the question of the intrinsic nature of the stuff whose structure its rules and equations describe. Dennett made the great mistake of his philosophical generation. He was sure (just like Descartes, whom he reviled) that we know enough about the nature of the physical to know that consciousness can’t be physical. He made the mistake in the face of all the evidence – the evidence of science – that the stuff of the world is profoundly strange, far stranger, it seems, than we can ever hope to understand. Dennett was, in this respect, a fanatic. Religious fanatics believe in the existence of something for which there is no evidence. Dennett went further: he denied the existence of something whose existence is certain. (If he is right, there has never been any pain or suffering or joy; not really.)

After asserting, with Dennett, that ‘consciousness can’t be reconciled with science,’ Gomes goes on to deride panpsychism. He compares it with Descartes’s positing of an immaterial substance, and calls it a form of ‘magic’. The central tradition of panpsychism is, however, resolutely materialist, from Margaret Cavendish on. It rejects Descartes’s dualism and has nothing to do with magic. Most anglophone philosophers were clear on the point a hundred years ago: ‘Panpsychism must be considered a species of naturalism,’ R.W. Sellars wrote in 1927. Panpsychism may or may not be true, and physicists aren’t very good at metaphysics, but we should perhaps pay attention to what some Nobel Prize-winning physicists think: that ‘the mental and the material are … two sides of the same thing’ (Hendrik Lorentz); that ‘the material universe and consciousness are made out of the same stuff’ (Erwin Schrödinger); that ‘consciousness [is] fundamental’ and ‘matter [is] derivative from consciousness’ (Max Planck); that ‘consciousness and matter [are] different aspects of one thing’ (Louis de Broglie).

Galen Strawson
London NW1

Anil Gomes writes: Galen Strawson claims that he, and not Daniel Dennett, is the real naturalist since only he affirms the one thing whose existence is certain – consciousness. And he takes this alone as sufficient justification for his own controversial views on its nature. But it is not obvious how to isolate what seems clear in introspective experience from our theoretical articulation of it. Strawson’s confidence is misplaced. He tells us also that panpsychism can be reconciled with physics because physics tells us only about the structure of the world and not its intrinsic nature. This, together with his appeal to Nobel Prize-winning authority, is supposed to show that panpsychism is in good scientific standing. Here Strawson and Dennett are closer than Strawson suggests. Both take reconciliation with science to be the standard for a theory of mind – they disagree only over whether panpsychism meets this standard. But the reality of conscious creatures – the reality of creatures like us – is not beholden to our being slotted into the best scientific theories. Strawson’s hurry to defend panpsychism from scientific disrepute prevents him from even recognising this as an option.

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