A Dialog on The Self

What is the self? Do we have one? Do we need one? If you refer to “yourself,” are you referring to your body? Does the self last our whole life? Let’s talk things over with BQTA contributor Rick Lenon, M.D.

RL: I’ve been thinking about the “Self” for a number of years. Let’s start with this: the self is real, and everybody has one, installed by Darwin. I think that means that we all have the same one, as much as our sense of “mineness” insists to the contrary. My own sense of myself ranges from my best understanding of myself as a whole human being at one end, to what it feels like to be the recipient of conscious experience, moment to moment, at the opposite extreme.  

GNS: OK. But more needs to be said about what the self is. Sometimes you talk as if the self is not real. 

RL: I never intended to imply there was no such thing as a Self.  Only that it can’t be what we think it is.  Galen Strawson likes to say that we know what consciousness is, because we all have it.  No way of knowing what it is, better than that.  Easy to realize the difference between being conscious and not… as in deep anesthesia.   What you think about it, try to understand, is another matter.  Make sense to argue about what it is, but not whether it is.  

GNS: The Self is even tougher to figure out, if that’s possible.  

RL: In his book Selves (OUP 2009), Strawson refers to the whole human being as the “thick self.”  That thick self is continuously changing.  His “thin self” is the witness to consciousness, moment by moment.  Strawson does not see that thin self as constant over time.  He does not believe that there is a core self that remains the same over time, that witnesses all the changes, but is itself unchanging.  Strawson does not believe that I’ve always been me. In fact, I think he has to be right about that.  The only thing about humans that doesn’t change is the Social Security number.

GNS: You’re broaching the problem of personal identity.

RL: Most people feel like there is such a core self; that it was me in 5th grade, and it’s still me now.  That regardless of what I learn or forget, the self doesn’t change. That core “me” that’s witness to it all.  

GNS: Well, aren’t you the same person—in the sense of numerically identical— you were in the 5th grade? The problem is to understand what it is that makes you that same person, given all the changes you have undergone.

RL: The one thing that my core self always has is my conscious experience.  I’m still there if there is no input from the world; even if all memory is gone.  But if I am truly unconscious, that core self is absent for the duration.  A tautology really; I can’t be conscious of myself if I’m not conscious.  

I think that people feel like there is such a constant core self because we’re hard-wired to feel that; not to believe it, but to feel it.  

GNS: Are you saying you are not the same person you were when you were in the 5th grade?

Catherine Deneuve -1965 age 22 c57e5ba98b7623d509f118e0a8d2015a(1)
Catherine Deneuve in 1965

RL: Strawson isolates the thin self as the witness to conscious experience, moment to moment.  A “moment,” in his discussion, is somewhere between 0.2 and 5 seconds, as I recall.  And self-with-history doesn’t fit into 0.2 seconds. 

Love to ask Strawson about “mineness”; that’s the way conscious experience always seems to have the quality of being mine. That seems right, but I’m not sure that’s always true before I ask the question; pre-reflectively, as it were (love that word).  Not sure to what degree “mineness” has an owner.  

The fact that we feel like there is a constant Core Self in there is more interesting that the fact that nothing about us is actually constant.  

GNS: There is a useful distinction to keep in mind here between type and token. There are two senses of “same”. Let’s say you and I have copies of a brand new book. Our books are indistinguishable copies. Same type. But your book is in your backpack and mine is at home. Different tokens of the same type. As things age, they change. Let’s say your book stays in your pack for several years and gets scuffed up. Now your copy is distinguishable from a new copy. But it’s still the same token book you put in your pack years ago, even though it’s not the same as it was when it was new.

People change more than books. You are the same token you were in the 5th grade, even though you are not the same in the type sense. You have changed and now you are a not the same type of person. The point is simple and general. If you paint your blue house white, the type changes—white house rather than blue. But the token does not change— still your house. Painting your house does not alter its numerical identity.     

Simon Blackburn gives these examples:

How many words occur in the works of Shakespeare? The question may be asking how many types of word, or in other words how large Shakespeare’s vocabulary is, in which case the answer will be several thousand, or it may be asking how many tokens of those types, or words as they would be counted by a printer, in which case the answer will be many more. A car manufacturer may produce half a dozen cars in a year (types, here equivalent to models) but many thousand cars (instances, tokens).

RL: You are comparing the self to houses and cars. Are you going to treat the mind as an object and its contents like mental objects?

GNS: Undue reification?

RL: Among other problems.

GNS: Well, there are famous puzzles at the edges of the type-token distinction. In the ship of Theseus, the planks are gradually replaced over the years as needed to make repairs. So the ship changes in the type sense, but remains the same ship in the token sense. But what happens when all the planks in the ship are replaced? Is the ship with all its planks replaced still token identical to the original ship? What if someone gathers all the old planks and puts them back together? The ship built out of the old planks barely floats, but has a good claim to being token identical to the original ship. That would disqualify the ship made from the replaced planks from being “the same” in either sense.

Humans are like the ship of Theseus. Every seven years, according to conventional wisdom, every atom in our bodies gets replaced. So we are no longer the same in the type sense, but we remain the same in the token sense. Is Strawson really denying that you are same token you were five seconds ago?

When school children asked Nancy Reagan, “what should we say when kids ask, do you agree with Strawson’s account of the self”, Mrs. Reagan gave her iconic reply: “Just say no.”

RL: Alright, extending your ship of Theseus example to humans, suppose we gather all those atoms floating around which were once part of your body, and reassemble them to make a body type-identical to the one you have now. That body would have a better claim to be you than you do, since it consists of the original atoms which made up your body, rather the replacements. You would have to surrender your personal identity to the reassembly.

GNS: How does one surrender their personal identity? Some of these duplication problems were taken up in the BQTA series “Star Trek Transporter”  https://better-questions-than-answers.blog/2017/08/22/star-trek-transporter-part-1/

In any event, the problem is to figure out what being the numerically identical token person over time amounts to. Don’t we sometimes think of the self as the controller of our bodies? If we don’t control our bodies, who or what does?

RL: Control always seems to imply a controller, and we want to think it’s the conscious self.  

GNS: If the self is not the conscious self, what is it? You imply that the conscious self cannot be a controller, despite what we want to think.  

RL: Problem is, we think of it as a homunculus, and it’s not.

GNS: What’s a homunculus and why can’t the self be that? Apparently, the homunculus is not real and not installed by Darwin. Maybe it’s “the ghost in the machine” Gilbert Ryle argued we all believe in (The Concept of Mind 1949).

RL: It feels like there is a “me” in there that experiences consciousness, as if it were seeing through the eyes, hearing through the ears. And that raises the nesting Russian dolls problem. If a self inside us is needed to explain how we have experiences of the outside world, there needs to be another self inside the first to explain how the first self has experiences. And so on.

The conscious self cannot be in control, because everything it’s conscious of has to come into existence before we can be conscious of it.  Even if you are aware of factors leading up to a decision, those factors had to be constructed before they could be presented to you.  The choices of which factors will be presented have to be made before you can see them.  Some of the reasons those choices were made might be consciously available (after the fact), and many won’t be.  

Ideas and thoughts are as real as stones.  It’s just that you can’t see the mental stuff before it comes into existence.

GNS: You’re describing Libet’s epiphenomenalism. The mental is constructed by neural activity, which must happen first because the cause must precede the effect.

RL: I’m seeing self and consciousness both as hardwired, and consciousness as having value above all else, then if you have consciousness without self, the self doesn’t matter as much.

GNS: So self and consciousness are two different things?

RL: Consciousness matters to us more than anything else, and nothing matters to us without it. The philosopher’s zombie is a case in point.  But for Darwin, and for my omniscient alien, consciousness is only a means to an end, which is existence itself.

GNS: Zombies, in the philosophical literature, look and act just like us, but have no consciousness.

RL: Darwin would have no objection to the philosopher’s zombie, if it were better at staying around.  Why Darwin was so taken by the function of consciousness is a most interesting question.  Just because we don’t understand it is no reason to conclude that he was wrong.

GNS: Why should we be conscious at all if the zombie version of us could function just as well? Zombies would have the same adaptive advantages and disadvantages as us, because they act just like us in all situations.

The zombie’s body is moved around by bio-chemical mechanisms installed by Darwin, just like us. Benjamin Libet’s experiments in neurobiology suggested that conscious pops up too late in the process to affect the movement of our bodies. That seems to make us zombies plus a witness. That Witness would be the Self. Maybe.

Next time, let’s take up zombies in more detail.


Catherine Deneuve at the 2011 César Awards




catherine deneuve-1960-22-peter-basch.jpg
Deneuve in 1960, photographed by Peter Basch


  1. I’m not going to state these things as well as they deserve, but I don’t think the type/token distinction gets us as far ontologically as we often take it. Is the “sameness of type” something brute, or merely provisional in the sense of “same because we treat it as the same”? That is, is the sameness a *conceptual* feature, something WE decide to gather under one roof, as it were, or something more *fundamental*?

    Human selves for instance. Do they differ in ‘type’ from the selves of other animals? If ‘no’, where do we draw the line between self-having and nonself-having entities? In what sense are we describing something ontologically *essential* or are we merely deciding criteria and making family resemblance assertions? If ‘yes’, where in our own evolutionary past did we suddenly acquire selves, because clearly there was a difference that Theseus’ ship cannot itself explain? If ‘yes’, at one point our human ancestors did not have selves and at another point we did? And this ‘self’ is somehow the same ‘type’ throughout the rest of our history? We have this foreshortened view of things when we do philosophy, as if our own present example sufficiently explained things. That breaks apart as soon as we see beyond our own example.

    The type/token distinction is a useful analogy for many things in ordinary life, but it *simplifies* the world in a way that our own experience defies at almost every turn. It is a *useful* fiction, to group some things as “the same”, merely for practical human purposes, but we continually make the mistake of giving these distinctions an independent ontological weight.

    The example of two copies of the ‘same’ book being tokens of a single ‘type’ is a good example of the confusion if only we take it a few steps further. The question came up that one copy may get worn, but that doesn’t change whether it is still that same token. Yes. But how far does that ontology go? What if for instance we put that book on a plane and send it to a land where not only do the people not have books, but they do not read and write? Well, *physically* it is the same thing, still, and if we are committed to physicalism of any sort to bolster our sense of ontology maybe that is enough.

    But in this other group of people it is not even what we would call ‘a book’. What if it gets used as a doorstop? Is that its own ‘type’, and is that relevant to this question? What if the purpose seems to be a supply of flammable materials for cooking a meal? Tear a ‘page’ and cook a meal? What if the ‘type’ invoked in this community puts it in a different class, groups it with things WE wouldn’t group it with? If we are committed to materialism then we answer one way, and the tokens simply suffer their fate and yet remain fundamentally ‘tokens’ of a ‘type’. Theseus’ ship.

    But the question then is what that type IS, in *material* terms? The author who wrote the book put it together from *ideas*, and the ideas themselves are not reducible to a material explanation. Are they? The type being invoked in the example of a book (or a ship, or selves) is a conditional construct of the human imagination. What things ARE seems to change depending on how we make sense of them in our lives. And the question should be to what extent we can build out an ontology from THAT. In what sense are we talking about natural KINDS, with strict identifiable boundaries, and in what sense are we inventing our terms with the language we use? Can we do science or metaphysics with the ordinary words of ordinary language?

    The website Edge.org has an annual question it asks leading scientists and philosophers. In 2014, the question was “What scientific idea is ready for retirement?” and were published in a book This Idea Must Die
    ed by John Brockman. Some of the responses suggest just how rampant our scientific misuse of ordinary words has become. Perhaps the ontological drive in mapping out ‘types’ and ‘tokens’ from our daily life is similarly hampered. “Self’ has an ordinary use, but is it scientifically viable? ‘Consciousness’? Here is what three authors in the book had to say.

    Pascal Boyer, Anthropologist and Psychologist, Washington University in St. Louis:

    Culture is like trees. Yes, there are trees around. But that does not mean that we can have a science of trees. Having some rough notion of ‘tree’ is useful for snakes that lurk and fall on their prey, for birds that build nests, for humans trying to escape from rabid dogs, and of course for landscape designers. But the notion is of no use to scientists. There is nothing much to find out, e.g. to explain growth, reproduction, evolution, that would apply to all and only those things human and snakes and birds think of as ‘trees’. Nothing much that would apply to both pines and oaks, to both baobabs and monstrous herbs like the banana tree.

    Why do we think there is such a thing as culture? Like ‘tree’, it is a pretty convenient term. We use it to designate all sorts of things we feel need a general term, like the enormous amount of information that humans acquire from other humans, or the set of idiosyncratic concepts or norms we find in some human groups but not others. There is no evidence that either of these domains corresponds to a proper set of things that science could study and about which it could offer general hypotheses or describe mechanisms…

    Is the idea of culture really a Bad Thing? Yes, a belief in culture as a domain of phenomena has hindered the development of a proper science of human behavior in groups—what ought to be the domain of social sciences…

    First, if you believe that there is such a thing as ‘culture’, you naturally tend to think that it is a special domain of reality with its own laws. But it turns out that you cannot find the unifying causal principles (because there aren’t any). So you marvel at the many-splendored variety and diversity of culture. But culture is splendidly diverse only because it is not a domain at all, just like there is a marvelous variety in the domain of white objects or in the domain of people younger than Socrates…

    Third, if you believe in culture you end up believing in magic. You will say that some people behave in a particular way because of “Chinese culture” or “Muslim culture”. In other words you will be trying to explain material phenomena— representations and behaviors—in terms of a non-material entity, a statistical fact about similarity. But a similarity does not cause anything. What causes behaviors are mental states.

    Some of us aim to contribute to a natural science of human beings as they interact and form groups. We have no need for that social scientific equivalent of phlogiston, the notion of culture.” (Pascal Boyer: https://edge.org/response-detail/25388)

    John Tooby, Professor of Anthropology, UC Santa Barbara, Co-director, Center for Evolutionary Psychology:

    Worst of all, the flow of discoveries and better theories through institutional choke points is clogged by ideas that are so muddled that they are—in Paul Dirac’s telling phrase—not even wrong. Two of the worst offenders are learning, and its partner in crime, culture, a pair of deeply established, infectiously misleading, yet (seemingly) self-evidently true theories.

    What alternative to them could there be except an easily falsified, robotic genetic determinism?

    Yet countless obviously true scientific beliefs have had to be discarded—a stationary earth, (absolute) space, the solidity of objects, no action at a distance, etc. Like these others, learning and culture seem so compelling because they map closely to automatic, built-in features of how our minds evolved to interpret the world (e.g., learning is a built-in concept in the theory of mind system). But learning and culture are not scientific explanations for anything. Instead, they are phenomena that themselves require explanation.

    All “learning” operationally means is that something about the organism’s interaction with the environment caused a change in the information states of the brain, by mechanisms unexplained. All “culture” means is that some information states in one person’s brain somehow cause, by mechanisms unexplained, “similar” information states to be reconstructed in another’s brain. The assumption is that because supposed instances of “culture” (or equally, “learning”) are referred to with the same name, they are the same kind of thing. Instead, each masks an enormous array of thoroughly dissimilar things. Attempting to construct a science built around culture (or learning) as a unitary concept is as misguided as attempting to develop a robust science of white things (egg shells, clouds, O-type stars, Pat Boone, human scleras, bones, first generation MacBooks, dandelion sap, lilies…).

    Consider buildings and the things that allow them to influence each other: roads, power lines, water lines, sewage lines, mail, roads, phone landlines, sound, wireless phone service, cable, insect vectors, cats, rodents, termites, dog to dog barking, fire spread, odors, line of sight communication with neighbors, cars and delivery trucks, trash service, door to door salesmen, heating oil delivery, and so on. A science whose core concept was building-to-building influence (“building-culture”) would be largely gibberish, just as our “science” of culture as person to person influence has turned out to be.” (John Tooby: https://edge.org/response-detail/25343)

    Richard Dawkins detailing another misuse of labeling:

    Paleontologists will argue passionately about whether a particular fossil is, say, Australopithecus or Homo.But any evolutionist knows there must have existed individuals who were exactly intermediate. It’s essentialist folly to insist on the necessity of shoehorning your fossil into one genus or the other. There never was an Australopithecus mother who gave birth to a Homo child, for every child ever born belonged to the same species as its mother. The whole system of labelling species with discontinuous names is geared to a time slice, the present, in which ancestors have been conveniently expunged from our awareness (and “ring species” tactfully ignored). If by some miracle every ancestor were preserved as a fossil, discontinuous naming would be impossible. Creationists are misguidedly fond of citing “gaps” as embarrassing for evolutionists, but gaps are a fortuitous boon for taxonomists who, with good reason, want to give species discrete names. Quarrelling about whether a fossil is “really” Australopithecus or Homo is like quarreling over whether George should be called “tall”. He’s five foot ten, doesn’t that tell you what you need to know?” (Richard Dawkins: https://edge.org/response-detail/25366)

    Back to Carter:
    My point is that we too often expect miraculous explanatory powers from the ordinary words of our ordinary language, and that we get carried far too far downstream pursuing these quests to easily turn about and head back to the way these words are used ordinarily in our ordinary language, the home they properly have. ‘Self’ is not a scientific term but a useful practical word for humans to refer to one another and them’selves’. And to be honest, ‘consciousness’ seems another.


  2. The type/token distinction explains the different meanings of the word “same”. So it can be true that “book A and book B are the same but not same”. A and B are the same token but not type if the book you bought five years ago (A) is now scuffed up (B). If A and B are two copies of the same book, they are the same type, but not the same token. Either way, same but not the same.

    This distinction is not controversial. It is not “a useful analogy” which “simplifies the world in a way that our own experience defies at almost every turn.” It confirms our experience at every turn. It explains how it can be true that you are the same (token) person as you were yesterday but not the same (type) because you got a haircut. Pointing out that “nothing about us is actually constant [type]” does not rob us of our personal (token) identity. Nothing mysterious or metaphysical about this. The same could be said of a book the owner is constantly writing in. Whether someone uses the book as a doorstop is irrelevant.

    Trouble could brew if token identity were treated as a separate object; if we tried to think of the thing that’s the same about a changing object as itself another object. Is that what “the Self” amounts to? Turning the token identity of people into a separate object? Maybe, but there seems more involved than that.

    Type/token might break down when dealing with abstractions. Suppose you and I both have the same thought: 7 + 5 = 12. Our thoughts bear an unmistakeable similarity. Same type, different tokens? Well, your thought is in your head and mine is in mine, I guess. But it might also be true that we are thinking about the same token arithmetic principle, not merely the same type of arithmetic. Here the distinction maybe is not too useful, unless we want to worry about the ontological status of numbers.

    Should we worry about the ontological status of “types” in general? Not all the time, surely. Any predicate we could apply to anything could be considered a type. If we can’t use predicates, we can’t talk about anything at all.

    Trying to figure out what types exist in the mind-independent world is tough. That’s the “natural kinds” problem. The authorities cited argue that “tree” “culture” and “learning” are muddled unscientific ideas. I remain unconvinced, but they have written about their theories much more elsewhere, I’m sure.

    Richard Dawkins makes some cogent points about the concept “species”. First of all, “every child ever born belonged to the same species as its mother.” The puzzle is that the transitivity of species must eventually break down. If A is the same species as B (mother/daughter), and B is the same species as C (mother/daughter again), then A must be the same species as C (grandmother/grand-kid). However, repeated through millions of generations, A will not be the same species as Z1183956. See BTQA “The Greatest Story Ever Told” https://better-questions-than-answers.blog/2017/10/03/the-greatest-story-ever-told/. It’s mind-boggling. The simplest idea in evolution is that you have ancestors who were not human beings.

    Does this mean that species do not really exist? That the distinctions between species are entirely arbitrary? Dawkins claims that “It’s essentialist folly to insist on the necessity of shoehorning your fossil into one genus or the other”, because “there must have existed individuals who were exactly intermediate.” But that does not prove that every distinction between species is useless or arbitrary. There is no scientific difference between Homo sapiens and Saccharomyces cerevisiae (the species of yeast involved in winemaking)? That’s “essentialist folly”? Of course not.

    The individuals exactly intermediate between Homo sapiens and Saccharomyces cerevisiae would be the most recent common ancestor, as Dawkins explains beautifully in The Ancestor’s Tale: A Pilgrimage to the Dawn of Evolution (new edition with Yan Wong 2016).

    Dawkins is wrong if he thinks that ordinary words lack meaning because their meaning might depend on context, or if there are exactly intermediate cases. His example is “tall”. Often science needs to strive for precision and “tall” is not precise enough for science. But “tall” remains a useful and meaningful concept. Green and blue are meaningful concepts even though there are exactly intermediate cases.


  3. You state, “This distinction is not controversial. It is not “a useful analogy” which “simplifies the world in a way that our own experience defies at almost every turn.” It confirms our experience at every turn.” Of course this is a whole other can of worms, what actually “confirms our experience”, but I will grant you that we DO make good use of the distinction and that it IS an important feature of our daily life. Does that mean it offers a clue as to what “really exists”, as you rightly question later on in your response? That too is worth looking into!

    My suggestion is that we sometimes look at the type/token distinction as a LITERAL expression of how things stand, and that this is a mistake. That it helps us navigate the world is unmistakable. That it EXPLAINS the world is quite another thing. And you acknowledge this in a seeming attempt in parts to defend the ordinary language view of things. As Wittgenstein so often points out, when our explanations run their course, often what we are left with is merely a way of *describing* things. The foundation is simply in what-it-is-we-do. And the truth is that we contradict ourselves quite often. Same/not-same isn’t an insight into the metaphysical firmament as much as it is an insight into the muddles that humans get themselves into in making sense of things.

    So the question is whether the type/token distinction that gets used so often so productively is a way of explaining or a way of describing. Because that DOES make a difference. I think you rightly point out that we are not always engaged in doing ontology, and in that case we are not really *explaining* what the metaphysically concerned would phrase as what “really exists”. Rather, we are describing how humans divide up the world, and different practices do it differently. The status of these non-metaphysical ‘things’ is merely what has come to our attention and how we make sense of the world. Generally not much more, except that the world does *not* contradict us. As you say, our conception of things “confirms our experience”. That is, the way we divide the world MAKES SENSE. But not being contradicted is not *proof* of much at all…..

    The issue is not that these are ‘arbitrary’ as you rightly point out, but simply that they are contingent. Conventional. Different people do it differently. That is why it makes sense to look at the form of life in which any such distinction is made. Our slicing of the world is different from how others do it. We can’t pretend to speak universally on much at all outside the empirical sciences (physics, but perhaps not biology?). Ordinary language across communities fractures into differences much more readily than it combines into generalities. In daily life our ordinary words are not quite the same instrument as the things we expect from hard science. We can’t expect scientific precision universally from our ordinary conception of things. Rather than science we need a *conceptual* investigation.

    Your conclusion about the “essentialist folly” Dawkins derides has it upside down, unfortunately. To point out legitimate differences BETWEEN our categories doesn’t ‘prove’ or support essentialism. Of course it doesn’t! The point against essentialism isn’t that we can’t tell things apart, but that the things we sort out as the *same* don’t always have an essential unity. What did Wittgenstein have to say about family resemblances? Was his point that ‘games’ were NOT different from ‘houses’?

    Our inclination to think in essentialist terms brings me to the question you raised in our private conversation: “We can’t have types?” I think this is a GOOD question. It is one we have to ask, rather than assume is answered, because ordinary language presents the world in what can only be described as types, and yet almost everything (not everything) so described is a matter of convention rather than fundamental ontology. There is a convenience factor in what we describe as types. They have practical uses for us. But other people using other languages and having other practices may carve the world up differently. And yet, it makes perfect sense to say that we carry on our lives AS IF they were established reality.

    That has consequences. Philosophers are too often too at home in their own languages to notice this difference. We don’t inspect our own use of words enough to see what actually goes on. We carry on investigating from our arm chairs rather than a bit of digging into the world outside our ken…. Enter Wittgenstein.

    What we need to be more clear on is how deeply these types extend, versus how contextual and humanly contingent they are. How DEEP are they? Philosophers make their hay on generalizing, so we need to be clear in what sense our categories ‘represent’ the world. Are we doing science and ontology? Or are we doing something human and conceptual? (And I’m not saying science can’t make some profit from how we ordinarily describe the world, simply that we can’t take for granted that every hairbrained notion we derive from ordinary words has an ontological implication) This is precisely why your follow up question NEEDS to be asked.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s