The Ego Trick (baggini, Julian)

then we are talking about real and significant differences between the western and East Asian mind, including how westerners and easterners see themselves and others. Nevertheless, Nisbett is right not to make too much of this. It’s not as though the two cultures are so far apart that people can’t relate to one another. As he says, Asians have the same fundamental theories of human personality, and he could talk with Asians about ‘Fung’s decency and humility, Chan’s arrogance, Lin’s reserve, understanding each other perfectly.’10 Also, although the differences do seem to be deeply culturally entrenched, they are far from unmalleable. Asian-Americans, for instance, display typically hybrid patterns, and Nisbett himself concludes his study by saying he expects the future to bring synthesis, not westernisation or polarisation.

For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure, colour or sound, etc. I never catch myself, distinct from some such perception.   David Hume, Treatise of Human Nature, Book 1, Part 4, Section 6

The central riddle I’ve set out to solve concerns the self’s continuity in change: how can we remain the same people over time, even as we change, sometimes considerably?

The Irish philosopher Bishop Berkeley once wrote, ‘Philosophers have raised a dust and then complain that they cannot see.’ The cloud that obscures our view of the self has been thickened by the trampling of psychologists, scientists, theologians and sociologists too. In contemporary academia, the assumed solution to this often seems to be that we should examine each particle of dirt in ever more isolated detail. My approach is rather to let the dust settle and have another close look at what it is that everyone was stampeding to see. Selves are like paintings: they cannot be brought into focus if you stand too far away or too close.

‘In one moment, everything that I had ever taken to be my personal self completely disappeared,’ she later recalled. ‘It was just gone. As I waited for the bus to approach, something in consciousness was loosening somehow. And when it got there … this reference point of an “I”, a someone that everything was about and that everything that occurred in life was structured around, was gone. It was like a switch had been turned off. And it was never to turn on again.’ She had experienced a blinding, shattering realisation: ‘There was no personal self.’ Yet ‘nothing stopped; the functions continued to function just as before. In fact, better than before.

Imagine believing someone to be a shining spiritual example, only to then conclude that they were in fact simply brain damaged. This would provoke too much ‘cognitive dissonance’: the mental discomfort caused by holding incompatible beliefs.

The most important finding, which seems to be universally accepted by all researchers into the self and the brain, is that brain research has given up on the search for the pearl of self. As the clinical neuropsychologist Paul Broks put it to me, ‘We have this deep intuition that there is a core, an essence there, and it’s hard to shake off, probably impossible to shake off, I suspect. But it’s true that neuroscience shows that there is no centre in the brain where things do all come together.’ The unity of the self is not to be explained in terms of a single, unified brain region, which acts as the master controller.

it is tempting to think that light shines on the retina and then the brain creates from this a single, three-dimensional image. But who sees this image? The temptation is to think (or perhaps more usually assume) that there is a kind of mind’s eye which inspects the image in the brain. But then how does this ‘mind’s eye’ see this image? It cannot be that there is a little person – a homunculus – in our brains which watches mental images. If that were the case, we’d have to ask what was going on inside the head of that homunculus. Would there be another mental image, and if so, what would be seeing that? An even smaller homunculus? If we continued to explain each stage in the same way, we’d end up with an infinite number of ever smaller homunculi, each packed Russian-doll-like into our brains. Such an infinite regress could never explain how any seeing actually went on at all.

You cannot explain the unity of experience by simply positing an inner, unified experiencer. That simply begs the question: how is unity of experience possible in the first place?

So how does it all come together? It’s too soon to say that there is a definitive answer, but there are sufficient similarities between different researchers’ views to suggest that the outlines of an answer are in place. The central insight is that the sense of self is constructed from a variety of different brain systems working together. Some of these are very old and are shared with very unsophisticated animals. Others are unique to higher mammals.

The newest and most advanced parts of our brains are a mere 24–55 million years old, born during the Eocene and Oligocene epochs. This is the neomammalian brain, otherwise known as the neocortex or cerebral cortex. Higher brain functions like logical reasoning and episodic memory depend on it. Only mammals have a neocortex, and it is most developed in primates, especially humans. Of course, this is just a model, and it should not be thought that the neocortex and limbic system are layered on top of the reptilian brain like some kind of neural lasagne. In reality, the systems evolved gradually and are deeply interconnected.

there is another kind of hierarchy: one which is nested. Nested hierarchies can be visualised as concentric circles, with the smallest circle in the centre at the ‘bottom’ of the hierarchy and the largest, outer circle at the ‘top’ of it. To say such a hierarchy is nested is simply to say that the top levels incorporate the lower ones. Higher levels are not therefore independent of lower ones. Take away the lowest level and the highest is left with a donut-like hole at its centre, and it just can’t function.

All living things, says Feinberg, are nested hierarchies. ‘At the lower levels of an organism are organelles that are combined to produce single cells that are in turn organised to produce tissues, which are then combined to produce organs that are ultimately organised to produce an entire living organism.’7 You may not understand the precise biological terms here (I don’t) but the principle is clear: ‘lower’ or ‘smaller’ elements are not separate from the larger, higher ones. Rather, they constitute those higher parts and are incorporated into them. So, unlike in a non-nested hierarchy, each level cannot be physically separated. Even more importantly, control does not come from the top down. Rather, the whole system puts constraints on what it can and can’t do and there is no single centre of control.

functionally, the self is indeed a nested hierarchy. This is because the higher functions of self, such as self-consciousness, are not independent of lower functions, such as basic awareness of…

Different theorists use different descriptions, but most distinguish between what is variously called a ‘minimal’, ‘core’ or ‘implicit’ self and the ‘extended’ or ‘autobiographical’ self.8 The minimal self is the most primitive kind of self-awareness. Any creature which is able to in some sense distinguish itself from the environment and other things has a minimal self. The reptilian brain has sufficient resources to enable this. However, to say that even lizards have selves is highly misleading, because this kind of awareness is almost certainly entirely in the moment. The lizard has some sense of itself at a time but not over time. Human beings, perhaps uniquely, have autobiographical selves. We have a…

In contrast, most, if not all, animal memory is simply a form of learned response. Your dog remembers you in that it recognises you as its pack leader, but it is very unlikely that it fondly recalls past walks, as you do. That difference is a core part of what enables you to develop the…

As well as hindsight, foresight is a characteristic part of autobiographical selves. We can plan ahead with incomparably more sophistication than even the most cognitively advanced primates. Chimpanzees can at best anticipate only a few steps ahead;…

It would be misleading to suggest that neuroscientists speak with one voice on the self, but all the leading people in the field do seem to agree on a few key fundamentals: that higher functions of self build on and incorporate lower ones; that there is no part of the brain at which it all comes together; that there is a continuum between the human sense of self and that of the lizard, and that there is no sharp dividing line between creatures with autobiographical selves and those without.

Broks has written eloquently about these issues, first in magazines and newspapers, and then in his book Into the Silent Land. His prose evokes the strangeness, mysteries and paradoxes of our experiences of self, usually starting from pathological cases, but using them to shine peculiar light on our ordinary experience.

Another example is found in people suffering from temporal lobe epilepsy, in an experience called transient epileptic amnesia.

The environment seems real but unfamiliar, but they know that they are. What is it that Descartes said? I know that I am, but what is this I that I know?’ This sounds very similar to Suzanne Segal’s far longer-term experience, and to the kind of state of selflessness that many spiritual seekers aspire to. Isn’t it strange, however, that the very same state which people aspire to achieve through spiritual discipline can be created by what, under the lens of neuropsychology, is a freak pathology?

Because normally the corpus callosum allows the two hemispheres to communicate, this presents no practical difficulty for most people. But the corpus callosum had been severed in Sperry and Gazzaniga’s patients, so this information exchange could not occur. That meant that if you controlled carefully which hemisphere received information from the environment, you could effectively make one hemisphere aware of something that the other was not. What is astonishing about this is that for this to be possible, there would have to be two centres of awareness in the individual concerned. But isn’t the very definition of a self that of a single, unified subject of awareness? Commissurotomy therefore seems to show that selves can be divided – at least temporarily – or that they needn’t have just one centre of consciousness after all.

Writers like Oliver Sacks, Paul Broks and Todd Feinberg have vividly described numerous other strange cases, often with detailed neurological explanations. Reading them, one is struck by the fragility of what we take to be permanent, essential features of our selves.

Despite these enormous upheavals, ‘sense of self is generally pretty robust,’ says Broks. The split selves of the commissurotomy patients, for instance, only emerge in specific experimental conditions. In real life, they find ways of retaining a unified experience of the world. Even with a split brain, we are able to function as though we had a unified mind.

If the unified self is an illusion, it is a very persistent one.

Broks says that, despite his intellectual convictions, he lives as a ‘soul theorist’ or ‘ego theorist’ who persists with the belief in a fixed core of self. ‘I think we probably all do, because that’s the way we’ve evolved biologically and sociologically to think and behave. Neuroscience is chipping and will chip away at the cherished ideas that go with that, like the idea that we’re autonomous agents with free will and moral choice. But it won’t stop us living by them and it probably shouldn’t.’

‘Physically we’re not the same, psychologically we’re not the same, genetically we’re the same but that doesn’t mean anything really; we may have certain patterns of brain organisation that predispose us temperamentally in certain ways, so we’re the same in that sense but then lots of other people probably have the same organisational patterns, so every which way you look at it we’re not really the same. What makes us the same is that we believe we’re the same.’

Sense of self over time is therefore ‘the story that we tell ourselves that keeps us together.’

We may go through major transformations, not just brain injuries, but what psychologists call ‘emotional insult’, such as the break-up of relationships or a bereavement. ‘Are we the same people when we come through them? In some sense yes, in some sense no, but it seems we have no choice but to believe that we are.’ Belief in the enduring self is therefore, Broks believes, a kind of magical thinking. But that is not to disparage it: the magic has real power, ‘it’s what we’re built on’.

So what could explain the strange combination of fragility and robustness that characterises the self? The answer is perhaps that the fragility is the strength. A pearl may be hard and distinct, but smash it and it’s utterly destroyed. A composite or amalgam, on the other hand, is of its nature a collection of things. That means items can come and go, or be damaged, without necessarily terminally destroying the character of the whole.

Michael Gazzaniga explains one aspect of this when he discusses why it is that a split-brain patient ‘does not find one side of the brain missing the other’. His simple answer is that ‘we don’t miss what we no longer have access to’.

Damasio believes that his view ‘resolves the apparent paradox identified by William James – that the self in our stream of consciousness changes continuously as it moves forward in time, even as we retain a sense that the self remains the same while our existence continues’. His solution is that ‘the seemingly changing self and the seemingly permanent self, although closely related, are not one entity but two’.11 That is to say, what seems permanent is the autobiographical self, and what seems to be changing is the core self.

The fragility and robustness of the self is therefore no paradox, but another riddle that makes perfect sense when you understand enough about it. In short, the robustness of the self lies in the fact that it is not a thing at all, but a product of the complex interaction of parts of the brain and body.

This means that very few, if any, parts are critical to its existence, and so it can adapt to sometimes massive losses. However, self does still depend on that functioning brain, and the wrong kind of malfunction can dramatically change it or completely destroy it. To put it another way, the self is a construction of the mind, one flexible enough to withstand constant renovation, partial…

idea of the self as a construction is one that many want to resist, because it seems to imply that it is not real. But of course constructions can be perfectly real. It is important to distinguish between ‘mere constructs’, which exist solely as ideas, and real constructions, which we pick up, use and live in. The ‘average American…

The limits of what neuroscience can contribute to our understanding of self was made evident to me as I read more and more about how the brain constructs a sense of self. The findings I have reported here do seem to illuminate the issue, or at least shine a light different from the torch of philosophers, one which brings out certain details more clearly. But if you go into details, the real nuts and bolts, you rarely, if ever, come across anything that takes us forward philosophically. All you get is more detail about the precise mechanisms by which self emerges from…

There are, however, two highly significant features of the self that scientific knowledge pulls into focus. Firstly it underlines very strongly the fact that there is no pearl at the heart of self. The apparent unity of self is the result of a highly organised play of parts. The second is that central to the ordinary sense of self is the construction of what has been called here the autobiographical self, and this depends, more than anything else, on memory…

Kathleen Wilkes complained that the philosophy of personal identity depends far too much on fanciful thought experiments, and that this is entirely unnecessary, because real life provides all the test cases we need.1 Dementia is perhaps the cruellest and most revealing example. Many philosophers have argued that we are constituted by a psychologically continuous web of thoughts, feelings, beliefs and memories. Dementia says, well, okay, let’s pick that web apart, piece by piece and see if anything of you remains.

Similar thoughts informed one of the first modern advocates of psychological continuity as the key to personal identity, the seventeenth-century philosopher John Locke. Such was the emphasis he placed on memory that he has been widely attributed with holding a ‘memory criterion’ of personal identity. Locke defined ‘person’ in psychological terms, as a ‘thinking intelligent being that has reflection and can consider itself as itself, the same thinking thing in different times and places’. The key to personal identity is therefore consciousness: ‘For since consciousness always accompanies thinking, and ’tis that, that makes every one to be, what he calls self; and thereby distinguishes himself from all other thinking things, in this alone consists personal Identity, i.e. the sameness of a rational Being: And as far as this consciousness can be extended backwards to any past Action or Thought, so far reaches the Identity of that Person.’

Although Reid’s objection hits the bullseye, in an important respect both he and Locke are looking at the wrong target.

We now know that memory works very differently. Each time a memory is recalled, it is changed in some way. Memories are not passive chunks of information. Rather, memory is an active process, the contents of which are forever in flux.

‘Every time we recall an event, we must reconstruct the memory, and with every recollection the memory may be changed … Truth and reality, when seen through the filter of our memories, are not objective facts but subjective, interpretive realities.’

We all ignore and do not commit to memory facts and events that conflict with the way we see ourselves and the world. We remember selectively, usually without conscious effort or desire to do so. And yet because we believe memory records facts, objectively, we fail to see that all this means that we are constructing ourselves and the world. Memories are therefore rather complicated building blocks of the self.

modern advocates of psychological continuity as the basis for personal identity over time see memory as just one of a number of psychological connections over time that contribute to this continuity, and they see this continuity in terms of overlapping connections, not rigid, constant ties to each stage of our past selves. These connections include temperament, desire, intention and belief. A person who had a good memory but woke up each day with a different personality or set of preferences would not experience enough psychological continuity to provide unity for the self. Nevertheless, for all its fallibility and constructed nature, it is clear that memory is one of the most, if not the most, important of these connections. Psychologists and neuroscientists are perhaps more explicit about this than philosophers.

H.M.’s loss was an immeasurable gain to the science of memory. Experiments conducted by Milner showed that although H.M. formed no new memories, he could learn new tasks. This showed that episodic memory – recollection of facts and events – worked quite differently from procedural memory – ability to remember how to do things. It also shed light on the different brain mechanisms underlying short- and long-term memory, since H.M. could hold information for about twenty seconds in ‘working memory’ quite normally. It also helped develop the distinction between explicit and implicit memory: memories we are aware of and those that affect our thought and behaviour without any conscious recollection.

What seems clear is that H.M. did retain some sense of self. This is certainly the view Philip J. Hilts formed, having spent a long time with H.M. while writing his book Memory’s Ghost. This residual sense of self was a ‘sharply realistic’ idea of the kind of person H.M. was. ‘The memory of our own traits does not depend on our collection of personal memories,’ writes Hilts. ‘They inhabit separate provinces. We do not arrive at summaries about ourselves by concluding from instances.’

If there is a continuity of key character traits, there will be some, albeit a radically diminished, continuity of the self too. ‘A man does not consist of memory alone,’ wrote the psychologist A.R. Luria, ‘he has feeling, will, sensibilities, moral feeling – matters of which neuropsychology cannot speak.’11 That is why H.M. remained recognisably H.M., even though he had lost a huge part of who he was.

This is closer to the view taken by Janet and her daughter Jackie, who run the care home where Linda is now a resident. Janet utterly rejects the common view that dementia is the slow death of the self. ‘I have a belief that dementia actually makes you more like yourself, so rather than rob you of your self, it robs you of all the exterior things that you pile on through life, all the baggage that you carry and the layers. What you’re left with at the end of the day with dementia is the core person, the soul, or whatever term you want to put on it. We’ve described it as an onion. If you peel an onion, from the brown skin outwards, you’ve got lots and lots of layers. When you get right to the centre of the onion you get to a little pearl in the middle and you can’t peel any more off it. It seems to me that it’s the real essence of the person.’

What gives this view some credence is the extent to which personality remains, even as memory and other cognitive capacities fail. ‘One of the things with the residents here that you see time and time again is that they behave in character,’ says Jackie.

it seems to me that a consistent set of personality traits is not a person. Janet’s metaphors seem to get it exactly wrong. I am my baggage. I am the layers that have grown on the onion, not the tiny core at the middle. We are precisely all the things we’ve accrued, the memories, the experience, the learning. If you strip away what you call the baggage you’re stripping away precisely the things that make us, that fill us out.

However, there is one possible candidate for the pearl of self that we have so far ignored: the soul. Many people believe that a human being, in addition to having a physical body, also has a non-physical part which does not depend on oxygen and food and which can persist long after our bodies have rotted into the earth. This soul is the true seat of the self: the body is merely a temporary vessel. Some version of this idea probably is and has been the most common belief around the world through human history. One

There was a wonderful cartoon that circulated around university departments when I was an undergraduate that dramatised this fallacy beautifully. A thought bubble emerges from a computer. ‘I think,’ it says. In the next frame, ‘I think, therefore I am.’ Cogito ergo sum, in Descartes’ memorable Latin phrase. The computer then goes on to continue a chain of thought that essentially follows that of Descartes, whose famous cogito argument shares the same basic logic as Avicenna’s falling man. As a thinking thing, the computer concludes, as does Descartes, that it is this capacity for thought which defines what it is, and that other features, such as its metal case and silicon chips, do not comprise its essential nature. It is res cogitans – a thinking thing – not res extensa – an extended, material thing. Overjoyed by this discovery the computer starts dancing around, only to accidentally pull its plug out of the socket. The dancing and the thought bubbles abruptly cease.

But why is this apparently simple error so common? A convincing explanation was offered by Gilbert Ryle in his 1949 classic The Concept of Mind. Ryle accused dualists of making what he called a ‘category mistake’. They started from the correct idea that thoughts, feelings and sensations were not physical things. The category mistake was to conclude that they must therefore be a different kind of thing, a non-physical thing. But there is another, more plausible alternative: they are not things at all. Rather, thinking and feeling are what brains and bodies do. Mind should not be thought of a substance, but as a kind of activity. Category mistakes are encouraged by conventions of language. When a word used to describe something is a noun, it is tempting to think that the word it describes is an object of some kind. So, if ‘mind’ is a noun, then minds must be things. But all sorts of nouns are not physical things, yet we do not think that they are some kind of other, ghostly thing. You can’t measure love in physical terms, despite the chuckles from a group of teenagers when I illustrated this point by saying it would not make sense to say ‘my love for you is six inches long’. But no one would conclude that therefore love is a nonphysical object. Nor would denying it was such an object suggest you did not think it was real. Another example would be a song, which is not identical either with any written score nor any particular performance. A song is real, but it is not a physical or non-physical object.

My favourite of Ryle’s many examples is the person who understands that playing cricket requires team-spirit, but thinks that contributing team-spirit is an activity additional to batting, bowling and fielding, rather than being tied up with the manner of doing all of them.3 The weaknesses in the arguments for immaterial souls are so evident that most serious philosophers now agree with Ryle, who famously dismissed Descartes’ notion as the myth of ‘the ghost in the machine’.

As he summarises it, ‘My argument has been that knowledge of what has happened to a person’s body and its parts will not necessarily give you knowledge of what has happened to the person, and so, that persons are not the same as their bodies.’

‘The trouble with physicalism,’ Swinburne told me, ‘is that it doesn’t face the data, and the data first of all is the mental life. We have thoughts and feelings and so on and the occurrence of these does not entail the occurrence of goings on in the brain or conversely.’ The conclusion is that a physical account of the world is not a complete account of the world. Therefore, there must be more to the world than the physical.

It should be clear that this argument simply repeats the kind of category mistake identified by Ryle. So it is true, as Swinburne says, that a physical account of the world is not a complete account of the world. But it does not follow that the world is comprised of more stuff than physical stuff. Rather, it simply follows that a complete account of the world is more than just an account of physical descriptions of the movements of stuff. This should not be a difficult thought. In the same way, a proper account of a game of football is not an account of the changing co-ordinates of the twenty-two players and the ball. If you knew enough about the game, such a description might enable you to infer what was going on, but nothing in it would entail that this was a sport with certain rules and a certain result.

in the split-brain case, there is a truth about what has happened to the person which we cannot know on the basis of the physical facts, namely, which of them is me. This is an example of what Derek Parfit calls a ‘further fact’ view of the self.8 According to such theories, knowing all the facts about a person’s brain, body, thoughts, feelings, relations and so on is never enough to tell you all there is to know about who the person is. There must be some further fact, perhaps a fact about which soul they possess, which is needed to settle such matters. But why should we think such further facts exist? Nations provide an apt analogy here. A complete account of the lands, peoples and relations of a nation tell you everything about that nation. There are no further facts which make nations nations. It is true that wholes are greater than the sums of their parts, but that is not because wholes are additional, different kinds of parts. Persons are no exceptions. We are greater than the sums of our parts, but not because there is some further part left out of the sum. We are nothing but our parts, but we are more than just our parts.

To this we need to add another compelling fact: if you have a sufficiently complex brain and body, you will be conscious. Combined with the first fact, most people would conclude that brains and bodies are both necessary and sufficient for human consciousness: you can’t think or feel without them; and with them, as long as they’re working properly, you can’t but think and feel. Having committed himself to belief in souls, however, Swinburne can’t accept this position. On his view, the non-material part of us is as essential as the material.

‘What I have argued,’ he writes, ‘… is that without a functioning brain, the soul will not function (i.e. have conscious episodes) – not that it will not exist.’9 Swinburne uses the analogy of a light bulb to explain this. ‘The soul is like a light bulb and the brain is like an electric light socket. If you plug the bulb into the socket and turn the current on, the light will shine. If the socket is damaged or the current turned off, the light will not shine. So, too, the soul will function (have a mental life) if it is plugged into a functioning brain.’10 But, like a magician, it is always possible that God could cause an unplugged bulb to light up if he so wanted: ‘maybe there are other ways of getting souls to function other than plugging them into brains’. This analogy ironically shows how far materialist conceptions have come, since it is the mirror image of a much older view: that it is the body which requires animating by an immaterial soul. The philosopher Avicenna, for instance, thought that the soul animated and controlled the body, even though it did not come into existence before it.11 That even a dualist today accepts that bodies power souls and not vice versa is indicative of widespread belief in the materialist basis of the world.

Swinburne is not the only serious thinker to advocate the existence of non-material souls, of course. But he is one of the most eminent, and his account is one of the most systematic and thorough. His writings lay bare the logic of dualistic souls, and once exposed, its failings should be obvious.

To switch metaphors, if the soil is not proving to be fertile, it is better to plant elsewhere than to continue in the hope that persistence will bear fruit. In all my years of reading and thinking about soul and self, I’ve yet to come across a single argument that is left standing after even a little serious scrutiny. As an idea, the immaterial soul is dead, and it’s time we buried it, along with any other dreams we might have had about finding the pearl at the heart of our identity.

As has been suggested, but not yet fully explained, our unity is a product of various parts and processes working together. Unity is the product of the self system, it is not the basis of it. What this means, in broad terms, is that you, me and everyone else are in some sense constructions, of mind and of matter. Many will find this troubling. If we are constructions, then could we not have been constructed differently? Might we not reconstruct or even deconstruct ourselves?

Once upon a time, in a castle on a large hill, there lived eleven strange, fractious individuals.

The castle and all its inhabitants were contained within the head of one man: Robert B. Oxnam. Oxnam was diagnosed with Dissociative Identity Disorder (DID), more commonly, but anachronistically, known as Multiple Personality Disorder. In his account of his life with DID, A Fractured Mind, Oxnam tells an incredible tale of how, through therapy, all these different ‘alters’ – as the different personalities are known – first came to light and how they then became at least partially integrated, so that only Bobby, Robert and Wanda remained, functioning in unison.

According to Smith, and many other DID experts, the vast majority of cases begin with severe childhood abuse. Dissociation is essentially a coping mechanism for tolerating the intolerable. In a strange way it is entirely rational. If you have the thought ‘this can’t be happening to me’, the logical corollary of this is ‘this is happening to someone else’. If you think ‘I can’t bear this’, then conjure up somebody else to cope with it for you. Most of us are familiar with less extreme versions of strategies like this. The introvert who hates social functions may create an outgoing persona, a pretence that can just be kept up long enough to get through the horrors of a hosting a dinner party. Someone who has to do an unpleasant job can see putting on their work clothes – or taking them off – as representing a line between the real self and a role that has to be taken on. In most such cases, the roles we take on are clearly understood as roles. There is no radical dissociation between the ‘real self’ and the temporary facade. What makes DID qualitatively different is that the separation goes right to the bottom of the conscious self. A ‘memory barrier’ is erected between the alters, so that the memories, thoughts and actions of the alters become entirely separated from each other. So in the paradigmatic case, the victim of the abuse becomes an alter whose experiences and painful memories are not shared with the other alter or alters.

There are, however, many who doubt DID really exists. The most serious worry concerns the way in which therapists bring alters to the surface. The deep philosophical question here is most evident when you consider the memories of childhood trauma that therapists elicit: how could anyone possibly know if what the client says about childhood abuse is a real memory or a confabulation? It is important here to remember that confabulation in this sense has nothing to do with deliberately lying or making things up. The mind confabulates any time that it comes up with a story of an event which didn’t actually happen. As we saw in Chapter Three when we looked at memory, the human mind is a prolific confabulator. What’s more, it is remarkably easy to elicit specific confabulations, as we saw from Elizabeth Loftus’s Bugs Bunny experiments.

Confabulation is not only a potential problem when it comes to causes, however. Could it also be that the alters are themselves brought into being by the therapeutic process, rather than simply brought to light by them?

To put it in broader terms, the self is not a single thing, it is simply what the brain and body system does. If that’s so, then there is absolutely no reason in principle why the brain couldn’t do different things at different times. In that sense, as the philosopher Daniel Dennett and the psychologist Nicholas Humphry concluded in their investigation into DID, not only is multiplicity ‘biologically and psychologically possible’ but ‘the possibility of developing multiple selves is inherent in every human being’.4

The kind of extreme dissociation in DID may be rare, or even non-existent, but if the self lacks a fixed core and is flexible enough to in principle be multiple, then should we think of ourselves in ordinary life as having significant dimensions of multiplicity? Once the idea of the unitary self is fractured, should we not take this one stage further and accept that in the absence of a strongly singular ‘I’, there must be a weakly multiple ‘we’?

Although most people find the idea of multiple-personality disorder strange and alien, many, perhaps the majority, buy into a weaker idea of multiplicity. For instance, there is a reason why these lines from Walt Whitman’s ‘Song of Myself’ have become so widely quoted: Do I contradict myself? Very well then I contradict myself, (I am large, I contain multitudes.) Many, if not most, people believe that they are multi-faceted, and that people who only know them in one specific context are not seeing the full picture. Although it is true that a great number take pride in saying that ‘what you see is what you get’, many others will tell you that there is more to them than meets the eye. However, this weak idea that we are multi-faceted seems light years away from saying we are in some deep sense multiple.

One of the most fashionable drivers behind the idea that we should all think of ourselves as multiple rather than singular is postmodernism. The nature of ‘postmodernism’ is hard to pin down, not least because it is in part a denial of the apparently old-fashioned idea that any reality can be pinned down, or even that there is such a thing as reality to be pinned down. It is, however, possible to say some general things about postmodernism which explain why it makes such a good bedfellow for the view of self as multiple. If postmodernism has a unifying theme it is one of fragmentation, which it picked up from modernism and ran with wildly.

This is not fragmentation for fragmentation’s sake, however. Underlying it is an intellectual rejection of various kinds of essentialism or realism. Before postmodernism, so we are told, most dominant modes of thought took reality to have a single, real essence.

Postmodernism rejected all that. The world has no real essence that is just waiting for us to describe. Reality is constituted by our understanding of it. History is created by the stories we tell about the past, and there are innumerable different such stories we can tell about the same events. Likewise, the Earth can be mapped differently: you don’t even need to privilege the land surface. Science is also a collection of stories we tell ourselves, ones we value for their instrumental or perhaps aesthetic value, but not because they reveal the essence of the real world. These ideas can be captured in the idea of the rejection of grand or meta-narratives. Humans make sense of their world by telling stories. Meta-narratives are the big stories that tie everything together.

In their place we need a multiplicity of narratives, ones which capture the different, contradictory perspectives that people in different times and places have of the world. This analysis applies to the self as much as it does to anything else. The self has no immutable essence. Rather, it is constructed, like a fiction. But what is doing this constructing? It depends on which postmodernist you ask. The answer is often ‘discourses’: the ways in which language shapes thought and perpetuates power relations.

However, with this comes the promise of a liberation. If the self is constructed, then may we now take over the task of construction for ourselves? We can create ourselves, become the authors of our identities, writing as many of them as we want, not limiting ourselves to a singular identity. Everyone really needs to calm down. Postmodernists seem to me like thinkers who have tasted potent conceptual liquor and, lacking all moderation, have become intellectual dipsomaniacs. Drunk on the rejection of totalising narratives, they fail to notice they have simply replaced them with a different one: the narrative that understands everything in terms of constitution by language and power. Taken to its logical conclusion, postmodernism itself would simply be one narrative among others, constituted by the discourses of its time, and no more worthy of respect than any other system of thought.

it is true that the self has no fixed essence, that it is in part a construction and that the autobiographical narratives we tell ourselves invent an order and cohesion that real lives lack. But it is not good enough to revel in this in a kind of Dionysian frenzy. You have to soberly think through what follows from it. And what does not follow is the opposite extreme. If you reject the singular grand narrative, it does not follow that we should embrace an infinity of contradictory narratives.

The hyperbole of hyper-malleability is refuted by the simple observation that human beings, postmodernists included, display a remarkable amount of continuity and unity, even though they have no fixed essence.

Reading Belle’s Best Bits, a kind of greatest hits collection from her two volumes of blogs, I quickly suspected not. It seemed Magnanti went into this line of work because it suited her. She sounded sexually experimental, easily able to divorce the erotic and the emotional, and pretty keen to do so. And when, in a Bristol pub not far from where Magnanti works, she told me about how she viewed her experiences, my suspicions became convictions.

Human beings think in terms of stereotypes, for good reasons. Without mental short cuts, we simply couldn’t get by. There is just too much information to process, and we often have to be quite crude as to how we filter it. But these generalisations are bound to lead us astray.

The banal reality is that the maintenance of different identities does not go deep to the core of Magnanti’s soul, but was simply a necessary means of keeping up some barriers for reasons of confidentiality and privacy. Most obviously, there is nothing philosophically profound about the fact she didn’t use her real name for her sex work:

Nor did going to work require her to ‘get into character’, to become a different person. ‘It wasn’t really a character so much as just suppressing some things and bringing other things to the fore. It wasn’t really playing a character because it was me. I mean, I remember having a conversation with a client once about chemoinformatics – that’s definitely me. It’s a bit like when you’re giving a talk at a conference and you’re wearing something you wouldn’t normally wear: if you’re a girl you put on a bit more make-up than you would usually wear, and then you only talk of x. It’s like for that piece of time you’re just a mouthpiece for your work and everything else about you is just too much, you leave the rest of that at the door. To what extent do people just pretty much do that in their lives? On the way home, I’m going to stop at the corner shop and get some milk, I’m not going to then sit there and tell my life story to the person behind the counter and I don’t expect him to see me as anything other than a customer.’

‘I’d say if you have too healthy an appreciation for who you are, that leaves you open to fossilising yourself. Everything’s open to revision. I’m an atheist, but let’s face it, if there was a real God-created miracle out there right now, that would be open to revision. I’m not too attached to that as…

People have this fantasy of “I give 100 per cent of myself to the world, all the time.” And online they also have this fantasy of “when I choose to hide things about myself that’s completely…

‘My now obviously ex-boyfriend was very keen to represent me as though there was a good me and a bad me. Anything that I did that he liked – you know, I knit – that was good Brooke, and then if I behaved in a way he didn’t like that was bad Brooke, as if there were two of me. It was like, “I know there is this good person. Why can’t you be her all the time?” And I felt really uncomfortable with that, because when you’re around someone a lot it’s hard not to start to take on the way that they see you and what they’ve created as who you are, and you do kind of respond to certain things about that. At the beginning that wasn’t the way I thought about it. It was more like going to work, putting on a suit; going to a party, putting on a dress; talking about x conversations, talking about y conversations. Then as time went on I did start to get the feeling, what if there’s a good Brooke and a bad Brooke, and can I just uncover good Brooke? But then after I got out of that I thought back around it again and just thought if you took all these things that he had identified as this other person, this incredible me that I could be, there…

To be a whole person is precisely to have depth and more than one side. This is why DID is a disorder rather than a fascinating, postmodern way to live: each alter does not actually…

shows how the postmodernists were right: there is nothing inevitable about the way we think about the world. We can carve it up in many different ways. What they fail to stress, though, is that some carvings-up are better and more truthful than others. We can indeed think…

Could it not rather be the case that the extent to which we perceive ourselves as one or many is in fact highly variable? That’s what the science writer Rita Carter came to believe after researching her book Multiplicity. ‘Multiplicity is a spectrum,’ she told me when I caught up with one of her selves near their home in Oxfordshire. ‘At the very extreme you have people with multiple-personality disorder who jump totally from one discrete personality to another. At the other end you have people who are almost entirely integrated, so that you can cut, slice them at any time, meet them in any context, and you will see the same recognisable things. I think both of those extremes are rare and most of us cluster in the middle.’

In support of this claim, Carter cites a very powerful weapon: empirical evidence. One of her biggest guns is state-dependent memory. ‘In different mind-states one does not have access to the same bag of memories, and that has been demonstrated in a lot of quite clever studies. So when you are very angry you do not have access to the same memories that you have when you are feeling very benign and relaxed.’ The research on this is extensive and well established. If you want to remember an event or fact, it is better to be in the same situation or state as you were when the event took place or when you learned the fact. There is even some evidence that something learned when drunk is better remembered when drunk than sober.13 Not only memories, but skills and personality traits are often to some extent state-dependent. A familiar example, in the literal sense, is that no matter what their adult personality is, many people revert to childhood versions of themselves when they gather together as family.

it is true that we could talk of multiple selves or multi-faceted selves, but the former is just more truthful than the latter. She illustrates this with the analogy of the self as a gemstone. ‘If you accept the intuitive model, that of a many-faceted gem, a cut diamond rotating and catching different facets of light, the trouble with that is that it implies there is something underneath the faces, that there is a central core that is rotating the faces around. I’m saying something very different. I’m saying that if you take away the faces there is nothing in the centre.’

This is sometimes taken to mean that chariots do not exist, only their parts. But this is clearly nonsense. A cart is not an illusion just because it has no existence other than by the correct arrangement of its parts. The only thing that doesn’t exist is an object which is somehow independent of its parts. In the same way, the self clearly exists, it is just not a thing independent of its constituent parts.

this is where I think talk of multiplicity repeats the error it is trying to correct. The problem with our intuitive sense of self is that it excessively ‘reifies’: it makes a singular thing out of it. Multiplicity does exactly the same, only at a lower level: making singular, unified things out of the parts which make up the self. This seems even more misleading than the reification of the whole self: at least ‘person’ in the ordinary, intuitive sense is singular in several important ways, such as having a single body and a single autobiography. If it really is a mistake to think of the self as a single entity, it just multiplies the mistake to think of ourselves as a collection of smaller, single selves. The multiplicity thesis seems most plausible when, on analysis, it is trivial. Nietzsche, for instance, wrote that the ‘subject is multiplicity’ and that ‘we always have only a semblance of unity’.15 But this ‘always’ takes the sting out of the thesis. What Nietzsche argues in The Will to Power is that nothing that enters consciousness is a unity. But if there are no unities, there is nothing to count, and it makes as little sense to talk about twenty personalities as it does to talk of one. It is like the difference between count nouns and non-count nouns. You can have one or two bottles of milk, but only some milk, not…

the singular self remains, for all its imperfections, a much better way of thinking about who we are than the hyperbolic claim that we are a multiplicity of selves. We are indeed less unified, coherent, consistent and enduring than we usually suppose, but we are still real and individual. The constructed self is robust enough to resist deconstruction from multiplicity…

We been looking inwards, when perhaps all the time we should have been looking outwards. Is the key to identity to be found not in our minds, but in the social world? The idea that the self is ‘socially constructed’ has been rather in vogue in some intellectual circles for some decades now. However, the idea of the ‘social self’ goes back to at least 1890, when William James talked about ‘the images other men have framed of me’ in The Principles of Psychology.1 We internalise these images and they become part of our self-image. To put it simply, how we are seen by others affects how we see ourselves.

First, it usually really matters to someone if there is a disconnect between how they see themselves and how others see them.

A second, related way in which the social self matters is that if people are treated very differently depending on how they appear, that can change how they perceive themselves. ‘I’ve been finding that recently I’ve been learning to be unconfident,’ Jñanamitra told me. ‘I’m now a middle-aged woman and middle-aged women in our society just disappear. So it took me really by surprise. I found myself being completely disregarded in situations. It just took my breath away really.’

You can’t just live in your private reality. What this suggests is that, like the body, the social matters primarily for how it helps frame and shape our psychological sense of self. However, some claim that this way of putting it gets things precisely the wrong way round. It is rather the case that the psychological, the inner, matters primarily for how it helps frame and shapes our social sense of self. It is our place in the world that defines who we are. The relations that constitute our identity are the relations we have with others, not those that hold between thoughts and memories in our minds. So is it true that we are social constructions?

As Jacques famously put it in Shakespeare’s As You Like It: All the world’s a stage, And all the men and women merely players; They have their exits and their entrances, And one man in his time plays many parts …2 This theatrical metaphor has become the dominant means of expressing the idea that human identity is constituted by the essentially social roles we play.

if you take away all the roles, what is left of the so-called ‘real man within’? At almost every given time a man is in some kind of role, so if you strip these away, the real, inner man becomes the man only at times of solitude. This is not credible.

‘The point is,’ she writes, ‘that no single environment can show the nature of anything, because to know the nature of anything (a woman, a lump of iron, x-rays, mosquitos or black holes) is to know its potential; that if it is in one environment it will appear or behave in one way; that if it is in another, it may be different.’

Many a person plays the role of great thinker when no one else is around to challenge their thoughts. The eyes of the world can force us to see ourselves as others see us. Alone, the image we may hold of ourselves may be no true reflection, but a flattering portrait produced by the magic mirror of vanity. Recognising this is important for self-knowledge. We may feel most ourselves when alone, for instance, but our self-knowledge is severely limited if we fail to see how our interactions with others also reveal important truths about who we are. What we see of ourselves when we are playing no roles is not the real self, but a partial glimpse of our self.

So the theatrical metaphor makes no sense if it posits the real self as the off-stage actor. The other interpretation is suggested by Jacques in As You Like It: that we are ‘merely players’. We are the sum of our roles, no more and no less.

one of the seminal theoretical exponents of the theatrical metaphor, the sociologist Erving Goffman. In his book The Presentation of Self in Everyday Life, Goffman adopts the theatrical model as ‘one sociological perspective from which social life can be studied’.

Goffman appears to be either agnostic about, or indifferent to, the question of the reality of different selves. His interest was not essentially ontological, that is to say, concerned with the nature of being. It was primarily sociological, concerned with the nature of social interaction. His insight is neither that the roles we play conceal our real selves nor that the roles we play are our whole selves, but that whatever we are, our interactions with others involve elements of performance that can be understood through a dramaturgical model.

since it makes no sense to talk about presenting an unedited, undirected self, it makes no sense to think of such presentations as

any respectable theory will completely agree that mind is what it is because of how it is embodied and socially situated. To say that the social self matters to self-image therefore doesn’t undermine the idea we keep coming back to: our identity is essentially rooted in a psychological sense of self.

There is, however, another argument for the social construction of self, based on the claim that notions of self vary enormously from culture to culture. And on the face of it, there does seem to be a lot of evidence that this is true. For instance, the philosopher Rom Harré has argued that the different use of personal pronouns in different cultures reflects different conceptions of the self.

‘Eskimo emotional states appear to be much more socially dependent than ours. Isolated Eskimos, in so far as they can be observed, seem to be stolid, neither cheerful nor depressed. But once they become part of a community (a family, say) they quickly take on the emotional tone of the community, whether they are intimately

However, there is clearly a danger here of exoticising the unfamiliar.

the fact that the person acts and is treated differently depending on the social situation doesn’t threaten the fundamental unity which underlies those diverse roles. It could be argued that without an awareness of the different roles we play, we would be unable to fulfil them properly. Although a lot of the adjustments we make to our behaviour in different contexts are unconscious, it would be very difficult, if not impossible, to make all the changes needed unless you, and others, had a sense that you had changed

despite the notorious confusions of westerners trying to get by in Japanese culture, neither locals nor foreigners ever have any difficulty in treating individuals as the same individuals over time, with consistent traits, beliefs and so on. If you try to really imagine a society in which a person’s own sense of their psychological continuity was not the bedrock of personal identity, I don’t think you can imagine a society which is human at all.

recent work of Richard E. Nisbett on the differences between how Westerners (Europeans and North Americans) and East Asians (Chinese, Japanese and Korean) think. Nisbett surveys a wide range of studies and concludes that there are real and important differences in how people from these regions tend to think. In summary form, it sounds very stereotypical. Aspects of East Asian and western thought can be lined up in diametrically opposite pairs: holistic (eastern) vs reductive (western); relational vs atomistic; community-based vs individualistic; and/both vs either/or; in flux vs rigid.

My argument so far supports four main claims:  First, there is no thing or part of you which contains your essence. Your body, your brain and your memories are all very important for who you are, but none of these is the pearl of self in which your identity resides.   Second, you have no immaterial soul. Whatever stuff you are made from, it is the same kind of stuff that everything else is made of, be it plankton, cabbages or orang-utans.   Third, given that the pearl view is to be rejected, this means that your sense of self must in some way be a construction. If there is no single thing which makes you the person you are, then you must be the result of several parts or things working together.   Fourth, the unity which enables you to think of yourself as the same person over time is in some ways fragile, and in others robust. So although we may not be unified in the simple, strong way the pearl view suggests, that does not mean that we go wrong when we think of ourselves as integrated individuals. We may be more internally fragmented than common sense assumes, but that does not mean we are radically multiple, and should give up thinking of ourselves in the first-person singular. Our sense of self may well be strongly affected by the social world we inhabit, but the sense itself remains inner, psychological.

by themselves, these truths leave a lot of unanswered questions and problems. Together, they tell us we are unified, material constructions. Each one of these words is problematic. The degree to which we are unified is still vague. We say we are in some sense constructions, but it is not clear how we are put together. We know that bodies, memories and society all play a role, but how does a strong sense of self emerge from their interaction? And the extent to which we are material is also puzzling, given that our mental life gives us the strongest sense of who we are, not our bodies. That is why transgendered people can know who they are even when their bodies are wrong or change, and why we often feel loved ones with dementia have left us, even as their diseased bodies continue to live. In this chapter, I’m going to try to answer these questions and flesh out what I think the correct view of our selves should be. I will be building on and synthesising arguments given by a string of major thinkers in eastern and western traditions.

John Locke’s definition of a person as a ‘thinking intelligent being that has reflection and can consider itself as itself, the same thinking thing in different times and places’.

Locke continues, ‘He is no more one self with either of them than if the soul or immaterial spirit that now informs him had been created, and began to exist, when it began to inform his present body.’ In itself, having the soul of someone now dead no more makes you that person than having the same liver would. Souls would only preserve identity if they took their consciousness with them. Locke’s point is so self-evident that you can’t even call it an argument. Just

That principle is that ‘person’ is not a biological category, but a functional one. Any creature, or even machine, that has the conscious capacities of a person is a person, irrespective of its species. That is why we recognise as people characters in science-fiction films who are human-like but of another species. Were we to be visited by any such friendly aliens, we would surely treat them as persons, not animals.

What makes for a single person over time is therefore, in the formulation of contemporary philosopher Derek Parfit, psychological connectedness and continuity. That is why a person remains the same person, even when their bodies radically change. That is why we doubt whether the person we once knew still exists when the continuity has been severely disrupted by disease or brain trauma. That is why memory seems intuitively to be so important to personal identity, but is not enough to fix it, because it is the primary but not only source of psychological connectedness and continuity.

Psychological connectedness and continuity is not the result of any core pearl of self that carries the continuity. Rather, this is sometimes called a ‘bundle theory’. The eighteenth-century philosopher David Hume was the first to use ‘bundle’ in this context, just once. His contemporary Thomas Reid picked up on it, attributing to Hume the view that ‘what we call the mind, is only a bundle of thoughts, passions, and emotions, without any subject’.5 Like many descriptions which start out as mockery, the term has stuck.

you just are the collection of these thoughts, ‘nothing but a bundle or collection of different perceptions, which succeed each other with an inconceivable rapidity, and are in a perpetual flux and movement’.6 This is the heart of the Ego Trick. The trick is to create something which has a strong sense of unity and singleness from what is actually a messy, fragmented sequence of experiences and memories, in a brain which has no control centre.

We only go wrong if we’re too impressed by this unity and assume that it means that underlying it is a single thing. But the self is not a substance or thing, it is a function of what a certain collection of stuff does.

it’s worth remembering just how much of reality is bundle-like. Atoms are not solid objects but bundles of electrons, neutrons and protons. A plant is a complicated system, not a simple thing. Cars only exist as an assemblage of parts. The internet is a network, not an object that can pinned down in time and space.

We do not observe the self in the same way that we observe the yellowness of a tulip, the smell of coffee or the thought that I really must book my train. The self is what is having those thoughts and perceptions – it is not itself a perception. And when you introspect, you might argue, you are always aware that there is a self having thoughts and perceptions. If not, there could be no awareness at all. To put it another way, what Gilbert Ryle calls the ‘systematic elusiveness of “I”’7 is simply a result of the impossibility of being both subject and object at the same time. The ‘I’ cannot catch itself for the same kind of reason that a hunter cannot be its own prey. It would be like trying to eat your own face, or look directly into your own eyes (without the aid of a reflection, of course). In that sense, ‘I’ is elusive in the same way that ‘now’ is. As soon as you have tried to refer to the present moment, it has passed. Likewise, when you try to observe your own self, to become fully self-conscious, that very act changes the nature of the experience you are trying to observe.

Cassam follows in the footsteps of Immanuel Kant, who argued that for experience of the world to be possible, there has to be a fundamental unity within the subject of experience which allows the world to be perceived. Kant calls this the ‘transcendental unity of apperception’: the unity of inner experience necessary for any particular experience of the world to be perceived as part of a wider, unified whole. There must be an ‘I think’ which necessarily accompanies all thought, which allows the manifold of experiences to be grasped as a coherent whole in conscious experience.

These arguments seem compelling, but attempts to work out from the armchair what is necessary for experience to be possible, as Cassam and Kant do, are always vulnerable to facts emerging on the ground which plain contradict them. People with Cotard Syndrome or transient epileptic amnesia, for instance, which we came across in Chapter Two, simply do not conform to Cassam’s armchair stipulations

I may have to think of myself as having a unity over time, but that does not mean that I really have it. How we must think about things at certain times, for certain purposes, in certain ways, does not necessarily tells us about how they actually are. Kant recognised this: ‘The identity of the consciousness of myself at different times is therefore only a formal condition of my thoughts and their coherence, and in no way proves the numerical identity of my subject.’12 Or, in Galen Strawson’s words, ‘self-experience exists, as a form of experience, whether or not selves do’.

The unity we experience, which allows us legitimately to talk of ‘I’, is a result of the Ego Trick – the remarkable way in which a complicated bundle of mental events, made possible by the brain, creates a singular self, without there being a singular thing underlying it. It seems as though there is some pearl at the self’s core, but in fact there is no core at all.

unity is not a cause: it is an effect of a remarkably disunified,…

On the one hand, I’ve been arguing that we are entirely physical, and that we have no material souls. But on the other, I insist we are not our brains or our bodies, and that our identity is defined by psychological continuity. We seem to be wholly physical yet oddly not essentially physical at all. The key to understanding how this can be…

The way I put it is that we are no more than, but more than just, matter. In other words, we are made up of nothing more than physical stuff, but to describe our true nature, you need more than just a physical vocabulary. You cannot fully describe what it is to be a person in the…

This way of looking at things – sometimes called non-reductive physicalism – seems to be the only way to reconcile…

The first is that thoughts, feelings, emotions and so on are real. There are some ‘eliminative materialists’ who claim that they aren’t, but in I.A. Richards’s famous phrase, such thinkers seem to ‘feign anaesthesia’,…

The second truth is that whatever thoughts and feelings are, they are not straightforwardly physical. You can’t see a memory under a microscope, and an emotion has no mass or velocity. It is true that with fMRI scanners you can see the brain activity associated with such experiences, and perhaps in the future we will be able to say exactly what the experience is purely on the basis of such observations. But no matter how good the technology gets, there is always a difference between observing a brain event and experiencing a mental event. To use the jargon, we can see the ‘neural correlates of consciousness’,…

the third truth: that there is no stuff in the universe other than the stuff of physics. Not so long ago, you could have put that same basic thought in simpler terms and it would probably have been more persuasive: there is nothing in the universe apart from matter. That formulation has become both less persuasive and less accurate, since developments in physics have changed the way we think about the fundamental constituents of the universe. The best simplification we can now come up with is that the universe is at root comprised of energy, but at the smallest,…

the only way physicists can really understand the universe is in the language of mathematics. The fundamental basis of physics is not atoms, or even subatomic particles, but equations. That is why, for all its mysteriousness, dark matter does have all the reality that is needed for physics: it plugs into the equations. Indeed, it is because the equations make it necessary that the existence of dark matter was postulated before it could be detected. So, odd though it may sound, the best definition of the physical world we now have is ‘that which has a place in the equations of physicists’. And one thing you can say with certainty about souls is that they have no place there.

It seems the only way to make sense of this is that mental events emerge from physical ones, without being strictly identical with them. As the neurologist Todd E. Feinberg puts it, ‘your life is not a pack of cells; your life is what your particular pack of cells collectively do, though I cannot observe such a thing as your life, touch it, put it under a microscope, or keep it in a bottle on shelf.’14 Thought and feeling are what matter does, when it is arranged in the remarkably complex ways that brains are. Matter is all that is needed for them to exist, but they are not themselves lumps of matter. In this sense, ‘I’ is a verb dressed as a noun.

what does seem to be true is that consciousness does indeed emerge from complex physical events in the brain, even if we don’t know how it does so. Whatever the mechanism, we have thoughts and feelings because we have physical brains that work, not because there’s something else in our heads doing the mental work instead. The evidence for this is simple but overwhelming: damage the brain, and you impair consciousness. Change the chemicals in the brain, and you change consciousness. Stimulate certain parts of the brain, and you get a certain kind…

Those who hold on to the idea that something other than a functioning brain is required to create persons with a sense of self are simply not looking the evidence in the face. Just because some mysteries remain, that does not mean that everything is a mystery, and it certainly doesn’t mean we are justified in appealing to other mysteries, like souls, to plug the gaps in our…

It is important to stress the physical basis of personhood in order to answer critics who argue that the contemporary philosophy of personal identity suffers from the same disability that has plagued western philosophy since its inception: it is so centred on rational thought that it either downplays or ignores the importance of our less rational features. The most significant victim of this ‘ratiocentrism’ is the body. And if you want to know why the intellectual crime has been committed, maybe a clue is in the canon of great philosophers: they’re all male. ‘From the…

a philosopher of Descartes’ undoubted genius is not going to completely miss the fact that this ego is very much embodied. In a passage often quoted by those seeking to defend Descartes against the charge of ignoring the body, he writes: ‘Nature also teaches me, by these sensations of pain, hunger, thirst and so on, that I am not merely present in my body as a sailor is present in a ship, but that I am very closely joined and, as it were, intermingled with it, so that I and the body form a unit.’18 But what Descartes admits in this sentence he simultaneously takes back, for the psychosomatic unit he describes is nonetheless a combination of ‘I’ and ‘body’. He may be intermingled with a body, but the body is not him. As he puts it in the very next sentence, ‘I, who am nothing but a thinking thing’.

The legacy of Descartes has also come under attack in neuroscience, explicitly so in the case of Antonio Damasio, who called one of his books Descartes’ Error. Damasio’s target is not just the general separation of mind and body, but the distinction between reason and emotion which it is deeply connected to. Damasio’s general hypothesis is that emotions are in fact essential to rational decision-making. Reason and emotion therefore can’t be separated, because the proper operation of the former requires and involves the latter: ‘Well-targeted and well-deployed emotion seems to be a support system without which the edifice of reason cannot operate properly.’19 What’s more, emotions are not produced simply in the brain. Rather, ‘the body is the main stage for emotions, either directly or via its representation in somatosensory structures of the brain.’ This is because ‘feelings are largely a reflection of body-state changes’.

Descartes’ error was to think that mind and body are of different orders, not that mind exists.

It’s also worth saying something at this point about the role of the unconscious. Many of a psychoanalytic bent also criticise the kind of analytic, philosophical account of personal identity I have presented for neglecting that of which we are not conscious. It would indeed be wrong to limit the psychological to conscious experience. There are unconscious drivers of our mental lives which are integral to their operation. Psychological continuity therefore requires some continuity in what is unconscious too. But it is important to realise that this is so only because the conscious cannot be the way it is without the unconscious. The unconscious matters because of its influence on the conscious, not in its own right.

For philosophers, ‘identity’ comes in two main forms: quantitative and qualitative. Two or more things are qualitatively identical when interchanging them would make no difference. For instance, if you have two qualitatively identical bowling balls, it makes no difference which one you pick up when you attempt a strike. Such qualitatively identical objects, however, are not quantitatively the same: they are not literally the same object. If you put a mark on one of the balls, the other is unaffected. If you smash one, you do not smash the other. If one is in one place, the other can’t be literally in the same place too. When we talk about the identity of objects over time, it is this quantitative identity that we are interested in. We want to know if the very same object has persisted over time, not merely whether two objects at different times are indistinguishable.

Quantitative identity is governed by a strict logic, summed up in what has become known as Leibniz’s law. In essence this says that if x and y are the same object, then what is true of x at any given time is also true of y at that same time. This is why two merely qualitatively identical bowling balls are not the same ball. If it is true of the first that it is on top of the second, then it cannot also be true of the second that it is on top of the first. If, however, I am told that ball x has the same size, weight, colour, composition and location as ball y, at the same time, then I know ball x is ball y. It’s the same logic that enables you to know that if Colonel Mustard was in the same room as the murderer, and that the only people in the room were the murderer and the victim, and that Colonel Mustard is not the victim, then Colonel Mustard is the murderer. The logic is unavoidable.

The problem is that this logically strict notion of sameness only works for reasonably solid, stable objects. In all sorts of real-world cases, notions of ‘sameness’ are vaguer and more context-dependent. Take Hobbes’s famous example of the ship of Theseus.21 This vessel is taken into dry dock for repairs. The masts are broken, so these are replaced. The hull is rotting, so that too is rebuilt with new timber. Then it’s noticed the deck is looking a bit worn, so that too is replaced. And so on, until no parts of the ship are the same as when it came in. So is it the same ship? And what if someone took all the old bits and put them back together? Would that have a stronger claim to being the original ship of Theseus?

As long as you know what aspect of sameness matters to the questioner, you can provide an answer. If you’re a forensic investigator searching for traces of DNA left before the ship went into dock, the boat you want is the one rebuilt from original parts. If you’re Theseus, your ship is the totally renovated one. If you know what aspect of sameness matters, you can provide a clear answer. What you can’t do is answer the general question, ‘but which one is really the same ship?’ It’s an empty question: it looks like a genuine question that should have an answer, but it has none. All sorts of questions of sameness in real life are of this kind.

As with the ship of Theseus, the important question is not about identity but about what matters. The same is true when we talk about people in the late stages of dementia. Are they they same persons we knew before the disease took its hold? In some ways yes, in some ways no. It depends. To expect the logic of identity to yield a clear answer is foolish.

One way of looking at this is that, rather than there being one question of personal identity, there are actually (at least) two different questions. There is a logical question about the identity of objects and the existential question about what matters to us about our survival and the survival of those we care about.

Not all philosophers have suffered from this misapprehension, of course. The best, to my mind, are precisely those who manage both to be no less precise than necessary, but no more precise than is possible. Paul Ricoeur is a rare example of a philosopher who seems to appreciate the unsuitability of applying logical identity to persons. His central idea is captured in the phrase ‘selfhood is not sameness’.22 Ricoeur uses the Latin terms idem and ipse to distinguish between sameness and selfhood. Sameness, idem, is unique and recurrent: one thing continuing to exist as exactly the same thing over time. In other words, idem is what we have called quantitative identity. But selves do not have this sameness over time. It is in their nature to change, never exactly the same from one day to the next. The trouble is that we tend to use the word ‘identity’ in relation to persons, unclear as to whether we mean sameness (idem) or selfhood (ipse). What we need to be clear about is that persons retain a sense of selfhood over time, but this is not a precise sameness.

Ludwig Wittgenstein is another thinker who can help explain why exactitude is not always possible. He saw that many words and concepts do not have precise rules for their use, and cannot have their meanings analysed in strict logical terms. Wittgenstein famously used the example of ‘game’. What is the precise definition of a game? Games are so varied that it is impossible to specify them without there being exceptions. Games are usually for fun, but war games aren’t, nor are all the games people play in relationships. Many games have winners and losers, but many playground games, for instance, are just fun activities.

Wittgenstein argued that meanings of words often cannot be specified in logically comprehensive terms. ‘For a large class of cases – though not for all – in which we employ the word “meaning” it can be defined thus: the meaning of a word is its use in the language.’23 (Ironically, this is often quoted without the initial qualification, making the definition appear to apply to all meaning. Wittgenstein’s defence of vagueness is thus more precise than is often believed.) In other words, to understand a word, sometimes you can do no better than learn by experience how it is used in different contexts. Everyone is familiar, I think, with the realisation that our competence to use a…

‘But is a blurred concept a concept at all?’ he imagined them objecting. ‘Is an indistinct photograph a picture of a person at all?’ he answered. ‘Is it even always an advantage to replace an indistinct picture by a sharp one? Isn’t the indistinct one often exactly what we need?’24 Fuzziness is not only a matter of concepts but of things. Custard cannot be nailed to a wall, not because the word ‘custard’ is vague, but because the sauce itself is not…

The philosophy of personal identity is, unfortunately, a prime example of this. It is fatally flawed by its goal of stating the strict logical conditions for the persistence of a person over time. Because this is how the whole debate is framed, it tends to underestimate the reality and significance of the vagueness inherent both in the concepts surrounding persons and in persons themselves. The theories thus say more about the logic of the concepts they employ than the phenomena they describe. And because many of the most important aspects of being a person do not yield…

The false assumption can be put in another way. Concepts such as ‘person’ and ‘self’ are taken to be either ‘sortals’ or useless. A sortal concept is one which allows you to count the number of things to which that concept applies, and to specify the conditions for their identity at a time and over time. The distinction is similar to the grammatical one between count nouns and mass nouns, so ‘love’ and ‘water’ are not sortals, but ‘computer’ and ‘ball’ are. Given this distinction, isn’t it obvious that ‘person’ and ‘self’ must be sortal concepts? Most philosophers of personal identity accept this. But I think the animalists are right to say that, given this assumption, theirs becomes the only coherent view, because nothing else other than the human…

I’m happy to bite both Noonan’s bullets. In fact, I chew them with relish. ‘Person’ is not best thought of as a robust sortal concept and so ‘personal identity’ is strictly speaking non-existent. The idea that ‘person’ is not a sortal concept may seem bizarre. How could we possibly have a concept of person that did not allow us to count how many there are in the room? But this only seems radical if one assumes that the choice is between a strict sortal and a totally vague mass noun. This is just another false dichotomy created by putting logic first and the reality of language and the world second. Person is not a sortal in the strict sense for reasons that should by now be obvious. The boundaries of the self are not precise, and there may be times, real and imaginary, when it really is not possible to say how many there are or whether they are identical with past or future selves. But for practical purposes we can usually say how many persons are in a room, or whether I am the same person as I was yesterday.

The distinction I am making here is between strict logical identity and what we might call pragmatic identity. There are numerous examples of words and concepts that are not strict sortals yet which can usually be used as though they were. Take the act of ‘having sex’. I’m always a little puzzled by the ease with which people will report how many times they had sex over a given timescale. How exactly do they count? We don’t normally ask the question because we assume we know what is meant. But try to specify it and you can see it is actually not unambiguous. Some (men, mainly) will take it to mean the number of times they achieved orgasm. But would they say they didn’t have sex if they made love and didn’t climax? And what if between orgasms the lovemaking didn’t stop? Did the woman, or the man, have sex twice or once? It should be pretty obvious both that there is no right answer and that, for practical purposes, it usually doesn’t matter. ‘Having sex’ is not a true sortal, but it can usually function as though it were. There are plenty of other examples. Chop up an apple, a banana and a kiwi fruit and mix them together. Do you have one portion of fruit or three? Fill a pint glass with wine: is that one glass of wine or three, or four? Is Pink Floyd’s Dark Side of the Moon one piece of music or ten? To think that there must be one and only one correct answer to each of these questions is absurd. The best answer is the one that gives the information the questioner is looking for.

This claim is only disturbing if you think ‘strict’ is the same as ‘rigorous’. We don’t have to lapse into obfuscatory fog just because we reject the possibility of logical rigidity. It goes back to the idea that we can be as precise as possible without being totally precise.

We can remain the ‘same’ through great change because it is of the nature of persons that they change, and so our continued existence as the same person requires only that there is a steady enough process of change, not that we remain identical in any particular respect. It is because we are not pearls that we transform over time and can still be thought of as the same person over time.

These are the three central facts about ourselves that we have to accept, if we believe that the Ego Trick has done its job and created unified individuals out of a bundle of mental, neural and physical activity. First, the unity of the self is psychological. Second, we are no more than, but more than just, matter. And third, our identity is not what matters. Familiarity with this view has removed pretty much all the strangeness it might have had for me in the past. Oddly, however, the main question that remains is one posed by both critics and some supporters: if this is what the self is, then isn’t the self an illusion?

My inattentive acquaintance had failed to spot the difference between explaining something and explaining it away. The TV programme had done the latter: it had shown that what appears to be spontaneous human combustion is no such thing. There are those among both supporters and critics of bundle theories of the self who believe that such accounts do not so much explain what we are, as explain us away. Just as the wick theory explains the appearance of spontaneous human combustion but denies its reality, so bundle theories explain why it is we believe ourselves to be individual persons who exist over time, but deny that any such beings really exist. This radical idea has a history that goes back to the first bundle theory of all: that of the Buddha.

Like any other old and geographically dispersed belief system, Buddhism ceased to be a singular system of thought many centuries ago. It is therefore impossible to say what the Buddhist conception of the self is. Nevertheless, the central concept of anattā– traditionally translated as ‘no-self’ – is important to all schools,

what is being negated in the word attā? To understand that, argues Batchelor, you have to appreciate the context of the Buddha’s time, fourth-century BCE India, where attā or ātman was very much the central idea of the brahmanic tradition. That tradition thought of brahman as the impersonal idea of the deity, the ultimate reality, the transcendent, the unconditioned, absolute truth of things. There is a spark of that God within oneself. The true core of self is understood in terms of ātman, a unitary, partless, fundamental awareness or consciousness that is ultimately indistinguishable from the reality of brahman.

Ātman is a kind of depersonalised self, lacking individual personality and identity. So if ātman succeeds in becoming fully reintegrated with brahman, personhood is destroyed.

According to Stephen Batchelor, ‘What the Buddha did was to reject that whole model altogether and declare that such an ātman effectively was a fiction, an illusion. So when he says anattā, he is rejecting the idea that there is an ātman, that there is a brahman, and focusing attention therefore on the phenomenal world. The Buddha’s teaching is really about how we come to terms with the world of appearances. For him there is nothing behind the veil of appearances, there is simply an open field of impermanent and contingent and very often tragic suffering and painful events.’

Batchelor points to a very famous passage in the Dhammapada (verse 80), in which the Buddha says: ‘Well-makers lead the water (wherever they like); fletchers bend the arrow; carpenters bend a log of wood; wise people fashion themselves.’2 ‘In that usage,’ says Batchelor, ‘attā, which is the same word, is not seen as problematic at all. In fact what the Buddha is pointing to is the notion of self as something like a project to be realised rather than something that inheres within you in some sort of transcendent way. So like a field to be cultivated, like an arrow to be fashioned, like a block of wood to be sculpted, so the person through their actions creates themselves. The Buddha’s idea of self therefore is something that we create.

Your identity, your sense of being a person is formed through your actions, and that is only possible because there is not a fixed self. There is no unchanging essence or substance to which those attributes are then attached at all.’

the most coherent readings of its teachings on not-self are indeed remarkably congruent with more recent bundle theories. The self is not an illusion. What is illusory is an idea of self which sees it as an unchanging, immortal essence.

Strip that away and you are left, in Buddhism, with the ‘five aggregates’: body, feelings, perception, mental formations and consciousness. The self, as Batchelor puts it, ‘is neither reducible to them nor can it be understood as existing independently of them.’ No more than, but not just.

that is what the Buddha was saying – not that there’s no such thing as a self, because in many contexts he would say there is, but that the self is not what it seems to be.’

The philosopher Daniel Dennett takes a similar line, calling the perception of a unitary controlling mind a ‘veridical illusion’. ‘It’s an illusion in the same way the desktop on your computer is an illusion,’ he told me. ‘There aren’t any little yellow files on your hard disc and in fact files are distributed, scattered all over your hard disk. All of the icons stand in for real and quite messy and elaborate processes – you really don’t want to know anything about them. It’s called the user illusion and that’s a good term for it. I won’t go so far as to say we’re the only species with any kind of user illusion, but that’s pretty close to the truth, I think.’

we must avoid what Paul Ricoeur called ‘the great oscillation that causes the “I” of the “I think” to appear, by turns, to be elevated inordinately to the heights of first truth and then cast down to the depths of a vast illusion’.

Similarly, if we are to see ourselves as we really are, we will always see a self there. Ego-less experience is not somehow more veridical than ego experience.

example of this given by Douglas Hofstadter, another writer who is keen to talk of illusions, describing the self as ‘a mirage that perceived itself, and of course it didn’t believe that it was perceiving a mirage’.6 He recounts an anecdote in which he pushed down on a pile of envelopes to pick them up and found, to his surprise, ‘there was a marble sitting (or floating?) right in the middle of that flimsy cardboard box!’7 Of course, it wasn’t a marble at all. As he explains, ‘for each envelope, at the vertex of the “V” made by its flap, there is a triple layer of paper as well as a thin layer of glue. An unintended consequence of this innocent design decision is that when you squeeze down on a hundred such envelopes all precisely aligned with each other, you can’t compress that little zone as much as the other zones.’ And that is a metaphor for the self: the feeling of a solid core which is created by nothing of the kind.

The feeling (resistance to compression) is entirely veridical: only the interpretation of the feeling (there is a marble there) is mistaken. He only goes wrong when he misinterprets what it is

‘There’s a stability and it’s actually like the stability of everything else in this world. It’s like the stability of the ship of Theseus. There’s no core of the ship of Theseus, but there are some parts that last longer than others and there’s an overall structure and if you keep replacing parts, you can come back in forty years and still recognise it, even though most of the parts have been replaced. What you’re recognising is not just the parts that are still the very parts that you experienced before but the overall organisation of those parts, and even if every part was replaced, you’d still have the ship of Theseus.’

Perhaps the simplest analogy is with a cloud. From a distance it looks like an object with fairly clear edges, but the closer you get to it, the more indistinct it becomes. Get really close and you can see it’s just a collection of water droplets.

The solidity of self is an illusion; the self itself is not. The Ego Trick is not to persuade us that we exist when we do not, but to make us believe we are more substantial and enduring than we really are. There may be an illusion as to what we really are, but not that we really are.

Pearl-like assumptions about the self are perhaps most stubborn when it comes to the idea of character. Almost everyone seems to believe that their personality is a consistent combination of traits which is, paradoxically, both unique and classifiable in any number of different schema.

Inspired largely by Aristotle, virtue theorists believe that inculcating certain habits and dispositions is central to being a good person. Ethics is not so much about following rules but becoming the kind of person who behaves well.

The belief that character predicts action can be so strong that people sometimes confidently say they are certain how someone acted, even when they cannot possible know what they actually did.

What there is plenty of evidence for is that this confidence in our ability to know the characters of others, and ourselves, is misplaced.

important is it in determining what we do? It is a truism to say that how we behave is a result of two factors: our own natures and the situations we find ourselves in. Many are now claiming, however, that it is situation, not character, which often makes the biggest difference to what we do. And if that is true, character is not as powerful a determinant of behaviour as most of us take it to be. John Doris has assembled the most comprehensive case for this view in his book Lack of Character. Many of the experiments he cites have become widely known. In one, an actor drops a folder full of papers outside a phone booth, just as an unwitting experimental subject is stepping out. Some of these guinea pigs, however, have just had a pleasant but trivial surprise: the phone box was rigged so that an unexpected dime appeared in the coin-return slot at the end of the call. Of the 16 people who found the dime, 14 helped the actor pick up her papers. Of the 25 who didn’t, only one did.3 It would be an amazing fluke if the 14 who found the dime just happened to be more altruistic people.

More likely is that, as other experiments have suggested, being put in a good mood, even by something small, makes people temporarily nicer. Character has nothing to do with it.

More worryingly, our willingness to cause terrible harms also seems to be highly situation-dependent. Two classic studies stand out. In his notorious ‘obedience experiments’, Stanley Milgram asked subjects to deliver increasing levels of electric shock to someone in an adjoining room.

The other infamous case is the Stanford Prison Experiment, in which Philip Zimbardo and his team simulated conditions in a prison with student volunteers taking on the roles of prisoners and guards. The two-week experiment was abandoned after six days, because the guards had descended into terrible cruelty, making prisoners clean out toilets with their bare hands and using fire extinguishers to hose down an insurrection. These were ordinary, decent people, turned sour by being placed in a position of power they could abuse.

‘There’s the myth about human dignity and human character,’ he told me over lunch. ‘So we really want to believe that our actions all stem from some internal set of motives, of goodness, of morality, and courage, and they get perverted in certain situations for certain people. I start with a more simple model, to say that the human mind gives us templates or potentials to be anything at any time, and in fact, any time that somebody does something extraordinarily good or bad, it becomes imaginable for us.

It had taken Zimbardo five days to fall into this mentality, and a furious hour-long argument to change his mind back. But, as is usually the case, having stepped back, he was in no sense corrupted for life. ‘Most people, including all the research I know on Nazi guards, bounce back. The guards in the Stanford Prison study, who did horrendous things to other students, there was no carry over.’

It’s worth stressing that although Zimbardo argues we have underestimated the power of social situations because we overestimate the power of individual dispositions, that does not mean he thinks people are not accountable for their behaviour. All the situational approach says is that we must take account of the mitigating circumstances.

‘situational factors are often better predictors of behaviour than personal factors … In very many situations it looks as though personality is less than robustly determinative of behaviour. To put things crudely, people typically lack character.’

The central truth is not that character is a myth, but that it varies more according to situation than we tend to assume. This means that, even if we take the findings at face value, a credible notion of character remains. Consider, for instance, a person’s tendency to be generous or mean. The Doris critique says you can’t just say that a person is mean or generous. The best you can do is to consider all the different kinds of situations a person would be in and ask if they would be mean or generous in each case.

corollary: the same person, with the same character, if now put into a life where a different range of situations was normal, could have the following cloud: Now the overall description of the exact same person would be ‘mean’, not ‘generous’. What this shows is that there is some contingency as to which descriptions of our characters are true. In other places at other times, a mean person could be generous, a brave one cowardly and so forth. This fits rather than contradicts common experience. Many German citizens, for instance, would have led blameless lives had they not lived under the Third Reich and been placed in situations which brought out their worse selves. Conversely, many people are able to live good, moral lives only because circumstances have not tested them.

If character doesn’t mean the tendency to display the same kind of traits across different situations, then it means nothing. Character without consistency is not character at all. There is clearly something to this, but one can still ask how much consistency character requires. As I’ve said, no one expects complete consistency.

Several attempts to analyse data on behaviour over a range of situations have suggested that there is significant stability of traits over different situations.7 This alone is enough to show that the idea that character is just an illusion is far too simplistic.

perhaps the most important limitations of character scepticism arise from a failure to distinguish between two senses of character, what I’ll call passive and active. Passive character is the set of dispositions we just happen to have as a result of our genes, upbringing and experience, without any particular effort on our own part. Active character is the set of dispositions that we have because we’ve worked on developing them. To use a philosophical term of art, dispositions are active if they are the result of an individual’s attempts at self-constitution, to build who they are.

it should be clear that, in real life, traits are never purely active or passive. Rather, they tend to greater activeness the more they reflect deliberate efforts to mould our own behaviours.

For the sake of simplicity, we can crudely schematise any given individual’s character into a hierarchy of types. The weakest kind of character traits are those which are passive and variable, neither self-chosen nor constant. Slightly more robust are passive, constant traits, which at least exhibit a kind of consistency. Next are active, variable traits, which, being self-chosen, are more fully ‘ours’ but lack constancy. The traits which are most ‘characterful’ are those which are both active and constant, ones we have in part chosen for ourselves and we display with a measure of consistency.

If people are indeed variable in this way, then what would that predict for the kinds of studies that Doris cites? We would expect to find that only a minority of people are actually of consistent good character, and, likewise, only a few of consistent bad character. We would expect that character is often under-developed and that people would indeed behave differently according to circumstances. And that is precisely what the evidence does suggest. In almost all the examples given, it is not the case that the situation entirely determines the action. There is a small but significant percentage of people who always do the right (or wrong) thing, regardless of the changes in situation. Some of these do so purely because they happen to be so disposed, others because they have developed their characters in good or bad ways.

Virtue ethics is not a theory about how most people actually make their moral choices, but a proposal for how we should set about making ourselves able to make better moral choices. Building moral character is a challenge, so the fact that the majority don’t seem to have risen to it is not surprising.

This does not mean there is no value in trying to cultivate a generous personality: indeed, there may be even more. The evidence is that unless you try to cultivate traits, whether you display them will depend more on circumstances than on you. So consistency in character is something to be created; it does not arise organically.

Our knowledge of situationism should be used to build character more wisely, not to give up on it. For example, John Doris takes as one lesson the idea that we should avoid tempting situations rather than rely on character to enable us to resist temptation. The father of virtue ethics, Aristotle, agreed. He advised against getting drunk, for instance, for precisely the reason that we cannot rely on our inherent goodness to stop us from behaving badly if we’re intoxicated. Likewise, a faithful spouse should avoid getting into situations that might tempt them into infidelity. To become the kind of person who avoids drunkenness or temptations of infidelity is precisely to develop virtuous character traits. Being of good character requires self-monitoring and sensitivity to context, not just relying on automatic behaviours.

If character has to be built in important ways, could one go further and say that identity itself is not just given, but is something we in some way create ourselves, by our actions? One of the most persistent recent advocates of this view is the philosopher Christine Korsgaard.

Agency – the individual’s capacity to act – can perhaps explain some of the unity of self over time that a standard bundle view makes problematic. Take a long-term commitment like marriage. On a simple bundle view, wedding vows make no sense: I cannot bind my future self if that self is in a real sense not the same self as I am. But to be a person does not just mean to be whatever series of psychological connections and continuities endure; it is to commit oneself to making certain connections and continuities happen. In that sense, we can perhaps see the wedding vows as being even deeper than usually thought: it is not just that I promise to have and to hold, I promise to try to build future selves that will be able to maintain this vow. We do not so much promise to be good spouses as promise to make ourselves into good spouses.

‘We are responsible for our actions not because they are our products but because they are us, because we are what we do.’9 However, her view is not an excessively romantic one. Self-creation is not heroic but necessary. ‘Every human being must make himself into someone in particular, in order to have reasons to act and live. Carving out a personal identity for which we are responsible is one of the inescapable tasks of human life.’10 Indeed, in order to do this, we have to adopt ‘practical identities’ which will define us socially, and these may be radically contingent.

Husserl wrote: ‘The ego constitutes himself for himself in, so to speak, the unity of a history,’ and ‘Since, by his own active generating, the Ego constitutes himself as identical substrate of Ego properties, he constitutes himself also as a “fixed and abiding” personal Ego.’

agency is certainly something that is often missing from bundle accounts of the self. Just as there is no point in the brain where the self comes together, so there does not appear to be any one thing that unifies the self. Memory plays an important role, as can the creation of an autobiographical narrative. But action and agency do critical work too. For example, the pro ject of writing this book arguably is more important for the unity of my self than the set of beliefs and ideas it contains and which I endorse.

When I sat down to write it, my own beliefs were not held on file, permanently ready for inspection. I often don’t remember my view about something: it is as though I have to think about the subject again in order to find out what I believe.

No matter how jumbled my brain is in some ways, unity of agency only requires a reasonably stable set of intentions, and the ability to carry them through. By a combination of retrieval, reconstruction, adaptation and thinking anew, I am creating this book, and that has helped sustain and create the person I am.

Strawson’s counter points to what is perhaps the most important consequence of the bundle view. If the self is not a concrete, singular, stable thing, then we should not expect there to be only one way of being a self.

If that’s right, then virtually every theory of self and personal identity has inevitably gone wrong, because they have sought a description of selves which neatly fits all cases. But bundles can be tied together differently, more tightly and more loosely. This does not mean there are no general truths about selves at all. The nature of our bundles means there are constraints on what kinds of persons there can be. These constraints are sufficiently tight that we never come across other persons who are entirely alien to us.

Strawson is perhaps best known for his articulation of one such axis of variability in persons, between short- and long-termers, or what he calls episodics and diachronics. ‘Here what I think you find is that people fall along a spectrum,’ he told me. ‘At one end of the spectrum there are the long-termers, who I call the diachronics, and they just naturally think that they’re the same selves throughout their lives, and they really identify with themselves when they were children, say. And then there are these people who whenever they look back more than about five years, feel that it’s not really them. And then moving to the other extreme, there are people who, like me, or how it seems to me, people who just feel that they’re effectively in the moment and they don’t really think that the self that they are in the moment was there even a minute ago.’

He has a sense of his self, but he doesn’t identify this with this persisting human animal. He’s relatively indifferent to its past and future, and this is temperamental, it’s not philosophical. ‘I found these things in my diaries saying things like “I have no narrative sense of self at all”, so I’ve got some nice evidence that this was [what] it was like to me before I might have been corrupted by philosophical thinking about it.’

I have found Buddhist philosophy to be a rich intellectual tradition. Yet its often profoundly rational thought sits alongside other ideas which seem to be no more than outdated superstitions. Conceptions of self are a case in point. Like many, I have been struck by the similarities between western bundle theories of the self and Buddhism’s doctrine of anattā, which we looked at in Chapter Eight. Nevertheless, Buddhists also believe in reincarnation. But how can a rational view of the self as a kind of temporary construction be compatible with the supernatural view of the self as something that can transfer from one body to another?

I wanted to find out what it really meant for someone who sincerely believed he was a reincarnation.

‘In Buddhism, everybody is a reincarnation. But there are different levels of reincarnation. Some choose what kind of home they want to be reincarnated in, in order to help others. Some are reincarnated without their own wishes because of their past karmic accumulations.’ In the case of tulkus – ‘highly realised’ lamas who are able to choose their own reincarnation – some leave letters or other instructions to tell their followers where to find their reincarnation. Others don’t, and it is for the disciples to go to a ‘high person who has the capability to know where to look for the child’.

Revealingly, however, there is an absence of any semblance of a personal relationship between Akong II and Akong I. Most obviously, Akong says of his previous life, ‘I don’t remember anything.’

For him, all that matters is that everybody says he is the first reincarnation of Akong, and that means he must have the same qualities. ‘Then I have to serve the best that I can for the benefit of people whenever I can. I’m not here to investigate the first Akong.’ Nor does he give any thought to who his successor will be: ‘That is up to the others.’ All he is interested in doing is ‘to try to develop loving kindness and compassion as much as possible, and try to meditate as much as you can’. That way, death will not be suffering but like ‘trying to change house’.

Ringu says that the whole idea of reincarnation is ‘just Tibetan culture’. As he explains, the idea of lines of reincarnated lamas is not something that goes back more than two millennia to the life of the Buddha, but one that actually only started in the twelfth century with the Karmapa lama (nearly two centuries, incidentally, before the first Dalai Lama). ‘The first Karmapa, when he died, said, “I have to come back, because my teacher asked me to build this monastery.” So after some time a boy came and said “I am Karmapa.”’

‘It became like a system that if your teacher is supposed to be a highly realised being then his students would try to find his reincarnation, and Tibet became full of tulkus,’ said Ringu. ‘So some of them may be true reincarnations, some of them maybe not so much, so we don’t know exactly.’

Akong and Ringu’s lack of any felt connection with their previous incarnations, the absence of any clear evidence that someone even is a reincarnation, and the no-self doctrine of anattā: all these suggest that, whatever reincarnation is supposed to be, it is not about the survival of the self in any standard sense.

Akong’s explanation starts with the assertion that on the Buddhist view of ultimate reality, there is no distinction between self and other. ‘Each individual is part of everybody,’ he tells me. ‘We are all the same mind, part of the same thing. We are not divided.’ How is it then that individuals are reincarnated? There must be some sense in which there is a distinction between self and other, other wise there could be no rebirth, nor any karmic consequences for individuals of their actions.

Akong offers a better explanation of what might survive death by use of an analogy. Ultimate reality is like an empty space. Individuals are like buildings occupying a part of this space. The end of a life is like the demolishing of a building, but then we build another – our next life – around the same emptiness. And if you do that, you have the same land and same atmosphere, which is why the karma follows you. The habits you follow in one life follow you to the next. The analogy neatly draws out a fundamental distinction between the Buddhist view of the self and the dominant one in the western tradition. For in a sense, I think of myself as the building, not the space. The building is what gives shape and coherence to a particular life: personality, memory, body and the largely conscious connections of experience.

But for Akong, these aren’t the most important things at all. What really matters are the habits that are practised inside the building, and which give the space a strange kind of character that persists even once the walls have come tumbling down.

That which continues from one life to the next is not the contents of consciousness, such as memories or knowledge of people and places, but the mind’s ‘habitual tendencies’. So if, for example, you have practised mindfulness enough to have extreme mental clarity, then that clarity will be there in your reincarnation. ‘If anything continues, I think what would continue are my habitual tendencies, the way I react, the way I think. That is what would probably continue, not necessarily all the memories, because lots of things we forget even now.’ What we seem to have here is a threefold distinction between the ultimate unified reality; the relative reality of individuals in each mortal lifespan (skandhas or ‘aggregates’); and the intermediate reality of the ‘mindstream’, persisting mainly as habits, which are the subjects of karmic laws. To put it another way, if we accept the Tibetan Buddhist view, we have three kinds of selves: the building, the space occupied by a series of buildings, and the entire space of which the building only occupies a part.

Buddhism needs to deny the personal nature of reincarnation in order to make it coherent, but assert its personal nature in order to make it ethical.

In practice, Buddhists seem to be ambivalent about the extent to which reincarnation is personal. For instance, many insist that even the terms ‘rebirth’ and ‘reincarnation’ are misleading, and that ‘mindstream emanation’ is better.

Rather than being a rebirth of self, it is as if you inherit the habits and karma of your predecessor and you try to leave a decent legacy to the next person in the chain. But, in other ways, Akong does talk as though it will really be him in the next life.

‘Death is trying to change another house. If I want a good house then I have to accumulate good karma, positive things.’

This is where I think the doctrine of reincarnation gets slippery. On the one hand, for it to make any sense, it clearly can’t involve the straightforward transference of one life to another body. You have a set of memories, plans, projects, friends and family. All of these things do not continue in whatever the next life is. So there is something which is obliterated at death and does not continue, something which seems to be very important to who we are. People like Akong seem to accept this, which is why they talk about spaces and buildings. Yet it is not difficult to elicit certain trains of thought – such as the karmic consequences on future incarnations of our present actions – which seem to make these spaces much more building-like after all.

Here we move from conceptual coherence to empirical evidence, and despite the numerous books and testimonies Akong and Ringu cite, as the old saying goes, the plural of anecdote is not data. If someone had really remembered a detail from a past life which could be verified and they could not have otherwise known, that would be an astonishing discovery, on a par with making contact with extraterrestrial life. And if people did routinely recall past lives, it shouldn’t be hard to find numerous verifiable instances of such memories.

Without necessarily adopting a deliberate strategy, a child could sense from the reactions of the group which object should be his settled choice. This is not remotely fanciful: even horses can do as much.

In the early twentieth century, one called Clever Hans appeared to be able to do some basic arithmetic, repeatedly and reliably tapping his hoof the same number of times as the right answer to a sum set for him to solve. When investigated, it was discovered that Hans was actually responding to changes in the posture and expression of onlookers, who visibly relaxed when he got to the right number. What psychologists call the Clever Hans Effect might well be what enables a child to correctly identify the lama’s possessions.

It is often considered disrespectful to dismiss any aspects of religion as mere superstitions, but if that is what they are, there is no other truthful way of putting it.

Reincarnation in the context of anattā, even if it takes place, is certainly nothing like survival of the self in any ordinary sense – and that is not in itself a criticism of Buddhism, but simply a description of its claims.

Having dismissed the idea that we have immaterial souls, you might assume that there is nothing left to be said for the Christian idea of a life to come. However, I was surprised to discover that many Christians, including Pope Benedict XVI, do not believe in the dualist soul of Descartes or Richard Swinburne. Far from being a core creed of Christianity, it is arguably not authentically Christian at all. Rather, as the Christian psychologist Warren S. Brown puts it, many Christians see the person ‘as a unitary physical entity without a separate nonphysical soul, but not reducible to “nothing but” the physiology of cells of the chemistry of molecules’.1 This is very much like the kind of material person I described in Chapter Seven: we are no more than, but more than just, matter.

Aquinas, for instance, believed in the existence of immaterial souls but wrote, ‘Soul is not the whole human being, only part of one: my Soul is not me.’2 Little wonder that the future Pope Benedict, then Cardinal Ratzinger, could write nearly 800 years later: ‘We have discovered anew the indivisibility of man. We live our corporeality with a new intensity and experience it as the indispensable way of realizing the one being of man.’3

the word ‘soul’ in the New Testament is usually a translation of the Greek word psuchê, which it is generally agreed does not necessarily apply to what Thacker describes as ‘a kind of detachable non-material part of me’. Aristotle, for instance, talks about psuchê and many scholars believe that he did not think it referred to an immaterial object, but a kind of organisation of matter which makes it function as a living thing. It’s somewhat similar to the way we now think about life. Many used to think that life was a kind of separate force that had to be ‘breathed into’ inanimate matter. Now we accept that life occurs simply when matter becomes sufficiently ordered and self-replicating.

‘Scripture contains no word denoting only the body (separated and distinguished from the soul),’ wrote Ratzinger, ‘while conversely in the vast majority of cases the word soul, too, means the whole corporeally existing man.’4

One corollary of this view is that, as Thacker puts it, ‘it is not remotely the case that the soul has its own inherent immortality. Basically God makes the call after death whether we survive or not. Immortality is a gift of God, and it’s a gift of God after death, it’s not a gift of God at birth.’ Ratzinger again is in accord. ‘The Greek doctrine of the immortality of the soul … is wrongly regarded as a Christian idea, too.’

If we reject dualism and believe that bodily resurrection is necessary for an afterlife, do we have any good reasons to think it will happen? The idea is clearly a hard sell in the contemporary world. You might well ask, ‘Is this not all completely absurd, quite contrary to our understanding of matter and its modes of behaviour, and therefore hopelessly mythological?’ If you’re Joseph Ratzinger, you have indeed asked just that, in exactly those words.8 It would seem to be good news for Christians if their religion didn’t commit them to a dualistic view of the mind, but fully embraced the necessity of embodiment. If you are going to believe in the life of the self after death, it’s a good idea to start with a conception of the self that is in step with our best knowledge about what people are.

A more common, and more plausible, strategy has been to accept that there will be some mixing of the basic matter out of which we are made, and to accept the consequence that God would have to use new atoms when resurrecting us. As Brown puts it, ‘our eternal existence would not be the continuation of some aspect of our current embodiment but would be a new creation, outside of the time and space we currently occupy’.10 This invites the question: would the resulting person be a resurrection or a replica?

For anyone who lacks the sure faith to simply trust that things must turn out as they are described in the Bible, no matter how incoherent they may seem to us, the traditional Christian idea of the physical resurrection of the dead seems implausible, if not entirely impossible. The dilemma is simple: the price you pay for accepting the essentially embodied nature of human beings is that such bodies are not and could not be reconstructed perhaps centuries after death and made immortal. The price you pay for saying that they are like our present bodies, only radically different in certain key respects, is that you place a huge mystery where an explanation should be. If I wanted eternal life, I would not be happy to bet on an eschatological resurrection.

‘In my forty years of being an evangelical, probably thirty years of remembering sermons, I don’t think I’ve heard a single sermon on the soul.’ Preachers tend to steer clear because it is all rather murky and complicated. This may seem reasonable, but in some ways I think it’s quite damning. Christianity’s great promise is of a life after death, reconciled with our maker. So if it can’t give a clear account of what that life will be like, it’s guilty of avoiding a defence of its most important central claim.

Let’s take the resurrection of Jesus Christ. If science somehow – and I can’t even imagine how – but if it told me that the resurrection of Jesus Christ was just categorically impossible, could not happen, I would disbelieve that and continue to believe what the Bible teaches about the resurrection of Jesus Christ, because if you take away the resurrection there is no Christian faith, it just doesn’t exist.’ It’s worth stressing this point because this is not just Thacker’s view. Many, if not most, people of faith start from that faith even when thinking about issues which can be approached in an entirely secular way. And if you want to understand why people believe what they do, you first have to identify what beliefs act as their bedrocks. For many Christians, that bedrock is the Bible, a personal experience of God or Christ, or some combination of the two. And if that is the case, that means they do not approach issues such as the nature of soul and self as objective enquirers after truth. When they reason about the issue, reason is in the service of faith and does not simply follow the argument wherever it may lead.

These dystopias are readily imaginable only because at some level it is obvious that human nature need not be set in stone, and that there is no reason in principle why creatures like ourselves might not become radically different.

‘Everything’s exploded in the last twenty years. You will get people who, even if they still live in the same community and a very stable family situation, are subject to a thousand different cultures every time they walk out of the door. Every one of those impacts produces a slither of new personality because you have to respond to the world and that’s how personalities get made.’ Carter does not see this as a bad thing. ‘Those very integrated people, I think, are becoming dinosaurs. I don’t think they’re very well equipped to deal with a multicultural, very fast moving, very dynamic sort of culture they might have to move in now.’

Greenfield has received a lot of flak from colleagues for her self-confessed speculative musings on how the very nature of human identity might change.

‘Although most neuroscientists are familiar with the concept of what we call plasticity,’ she told me, ‘it might be that most people haven’t really grasped just how malleable their own brains are, how sensitive their brains are to the environment.’ Given that, she is fascinated by ‘how the unquestionably different environment of computer games and sustained exposure to a screen might change the brain in ways that have not happened before, because we’ve never had this kind of experience over sustained periods of time before.’

‘When you’re born, you’re born with pretty much all the brain cells you’ll ever have, but it’s the growth of the connections between those brain cells that accounts for the growth of the brain after birth. So, even if you’re a clone – that means an identical twin – you will have a unique configuration of brain-cell connections because they

Where Greenfield gets a little more speculative is when she imagines the variations in types of selves which might emerge if different kinds of connections become dominant.

a quite useful way of simplifying three forms of subjective experience. The first she calls the ‘Someone’ scenario, which is the normal case of ‘you being different from someone else’. This is how we almost all, almost always, feel ourselves to be. ‘In brain terms, the Someone scenario could be reflected in the forging of idiosyncratic connections, and associations that are unique to your particular experiences, and also leave you open to modification by those.’ Although we tend to think of ourselves as Someones, we are on occasion also ‘Anyones’, the second of her scenarios. ‘Sometimes in our everyday life, we’re all parts of a team, it’s where we subsume our own interests, our own individuality, we sacrifice what we personally want to do in favour of the group.’ Instead of ‘living out an individual life story’, Anyone lives out a collective life story.

We also sometimes live life under the third scenario, as Nobodies. ‘Human beings have stalked this planet for 100,000 years, [and] we have from time to time indulged in wine, women and song; or, drugs and sex and rock and roll, the modern equivalent. There are moments in most of our lives when we want to “let ourselves go”, “blow our minds”. The very word “ecstasy” in Greek means to stand outside of yourself. These are moments when we abrogate our sense of self, when we are no longer self-conscious, we’re just conscious.’ This is what psychologists call depersonalisation. Dancing is one of Greenfield’s favourite examples. ‘The whole point of dancing is that you let yourself go. You’re having a sensational time, that is to say you’re putting a premium once again on the senses, as you would as a small child, you’re back in the booming, buzzing confusion.’ What’s interesting about the feeling of being Nobody is that it can be wonderful and dreadful. For instance, most people who take marijuana report depersonalisation effects, sometimes quite strongly, and find them pleasant.

Greenfield’s worry is that the future is Nobody’s.

The route from here to this dystopia seems very unclear. Greenfield bases her worries on the idea that screen culture is essentially ‘sensational’, that it bombards us with discrete experiences which do not forge the kinds of connections that make us into Someones. She seems to think that sitting in front of a computer renders the user passive. But actually it is much more demanding than the television, its predecessor as the destroyer of western civilisation. Even