Dennett in The New Yorker

I wanted to bring to your attention a lovely profile piece on Dan Dennett, “Daniel Dennett’s Science of the Soul“.  It’s nice to see a philosopher as respected and well-known as Dennett come alive as a human being.

I’d also like to remind those of you interested in this sort of thing that Dennett has a new book out, From Bacteria to Bach And Back: The Evolution of Minds. The central project is to do what creationists are always saying can’t be done: use the explanatory resources of evolutionary theory to understand why we have the kinds of minds that we do. There are decent reviews here and here, as well as one by Thomas Nagel in New York Review of Books that I regard as deliberately misleading (“Is Consciousness an Illusion?“).

[Note: The profile and/or the Nagel review may be behind paywalls.]

 

211 thoughts on “Dennett in The New Yorker

  1. walto: Sounds like a cousin of the doubtful/dubious distinction to me.

    Ha! Hopefully more useful than that!

    I want to start off with a distinction between conceptual analysis and causal explanation.

    Roughly, in giving an analysis of a concept, we are interested not just in defining it but in giving a characterization of it. We’re interested in making explicit the role of that concept within a family of inferences — and in the case of empirical concepts, also the use of that concept in perception and/or action.

    By contrast, causal explanations are (generally speaking!) models of the underlying processes that give rise to observable phenomena. The models can be formal but needn’t be; in a loose sense even an account that simplifies and magnifies can still aid understanding (e.g. a focus on the economics of slavery will not tell us everything we might want to know about the causes of the Civil War, but it still advances understanding).

    The next step is to introduce Carnap’s distinction between analysis and explication. In his terms, conceptual analysis fully specifies the intension of a term through necessary and sufficient conditions. By contrast, explications aren’t complete and they aren’t cashed out in terms of necessary and sufficient conditions; an explication is a provisional specification of a term as already being in use within a community.

    In light of that contrast, ordinary-language philosophy is a philosophy of explications. What Ryle is doing in The Concept of Mind is explicating the use of terms like “remember,” “feel”, “look,” “decide”, and so on. In doing so a whole host of implicit inferences and background assumptions is made explicit.

    That’s not explanation, but explication.

    A great deal of Dennett’s project is about using cognitive science and evolutionary theory to get at the explanations necessary to supplement Ryle’s explications. (As well of those of Wittgenstein, Sellars, and Brandom.)

    So, we can say both that, as a matter of explication, thought is “simulating or modeling some potential behavior” and also that, as a matter of casual explanation, thought is information-processing in a neurocomputational hierarchy.

  2. Erik: If “user illusion” (a human using the mind) is a direct quote, then what objection can Dennett have when his theory of mind is summarized as “consciousness is an illusion”?

    First, it’s important to understand what the term “user illusion” literally means; secondly, it’s important to understand that Dennett is using this term as a metaphor. The term “user illusion” does not mean “a human using the mind”, and the metaphor of “user illusion” here does not mean that consciousness is literally an illusion.

  3. Kantian Naturalist: Ha! Hopefully more useful than that!

    I want to start off with a distinction between conceptual analysis and causal explanation.

    Roughly, in giving an analysis of a concept, we are interested not just in defining it but in giving a characterization of it. We’re interested in making explicit the role of that concept within a family of inferences — and in the case of empirical concepts, also the use of that concept in perception and/or action.

    By contrast, causal explanations are (generally speaking!) models of the underlying processes that give rise to observable phenomena. The models can be formal but needn’t be; in a loose sense even an account that simplifies and magnifies can still aid understanding (e.g. a focus on the economics of slavery will not tell us everything we might want to know about the causes of the Civil War, but it still advances understanding).

    The next step is to introduce Carnap’s distinction between analysis and explication. In his terms, conceptual analysis fully specifies the intension of a term through necessary and sufficient conditions. By contrast, explications aren’t complete and they aren’t cashed out in terms of necessary and sufficient conditions; an explication is a provisional specification of a term as already being in use within a community.

    In light of that contrast, ordinary-language philosophy is a philosophy of explications. What Ryle is doing in The Concept of Mind is explicating the use of terms like “remember,” “feel”, “look,” “decide”, and so on. In doing so a whole host of implicit inferences and background assumptions is made explicit.

    That’s not explanation, but explication.

    A great deal of Dennett’s project is about using cognitive science and evolutionary theory to get at the explanations necessary to supplement Ryle’s explications. (As well of those of Wittgenstein, Sellars, and Brandom.)

    So, we can say both that, as a matter of explication, thought is “simulating or modeling some potential behavior” and also that, as a matter of casual explanation, thought is information-processing in a neurocomputational hierarchy.

    That’s a little confusing. If I’m not mistaken you’ve actually got THREE things going in this post. Explanation (causal); Analysis (necessary and sufficient conditions); and Explication (incomplete analysis–discussion of common usage, etc.)

    Is that right?

  4. Kantian Naturalist: secondly, it’s important to understand that Dennett is using this term as a metaphor.

    Metaphor for what? Metaphor as in “talking as if explaining something while really hand-waving it away”? This is my honest impression of Dennett when reading or listening to him.

  5. walto,

    Yes, that’s right. But then I don’t think there are necessary and sufficient conditions for empirical concepts, which means that emprical conceits can at best be explicated, not analyzed.

  6. Erik,

    A metaphor for the relationship between subjective experience and neural machinery. On Dennett’s view, subjective experience is the brain’s user illusion of itself.

    I don’t agree with Dennett on this, but it’s not as inchoerent as his detractors (like Nagel) make it out to be.

  7. Kantian Naturalist: A metaphor for the relationship between subjective experience and neural machinery.

    How can there be a relationship with something that has no locus (“medium”)?

    Kantian Naturalist: I don’t agree with Dennett on this, but it’s not as inchoerent as his detractors (like Nagel) make it out to be.

    Perhaps not as incoherent when we know about the particular background assumptions that make it less incoherent. Has Dennett been open about them? Neil thinks Dennett is an immaterialist. Keiths is positive he is a physicalist. You make something third out of it. Guess Dennett is neither open or clear.

  8. Erik: Neil thinks Dennett is an immaterialist.

    I should clarify that. Dennett claims to be a materialist. But I see him as effectively an immaterialist because I see his theory of mind as amounting to a fleshing out of Berkeley’s idealism (which is usually considered immaterialist).

    No doubt Dennett thinks that he is connecting his ideas to the material world. But what I have read so far (I’m currently in chapter 9), is mostly hand waving. He fails to actually make a connection.

    Dennett spends a lot of time arguing that the brain was designed by natural selection. And perhaps that’s where he thinks he is connecting it all to the material world. I’m a skeptic of Darwinism. I support evolution, but I don’t buy the “natural selection as designer” part.

  9. Erik: How can there be a relationship with something that has no locus (“medium”)?

    His point (as I understand it) is that consciousness is not some sort of medium or thing that stands in between the world and us. It’s not that the spike-trains across neurons that results from the bombardment of our sensory receptors is then taken up by some other thing and turned into subjective experience. That’s the dualistic picture (or at least Dennett’s version of it) that he means to dislodge.

    When I’m walking down the hallway, of course I see the doors, the lights, the carpet, and I sense my own walking.

    At the neurological level of description, photons are interacting with the electron shells in the rhodoposin molecules inside the rods of my retina, and that triggers spike-impulses that propagate across multiple pathways, subcortical and cortical, and that information is processed in the visual cortex.

    But it’s not as if there’s some other thing, the mind or self, that has to then look at the visual cortex to determine what I’m seeing. It’s just that the information processing (at the cortical and subcortical structures) is the seeing (at the level of personal experience).

    Perhaps not as incoherent when we know about the particular background assumptions that make it less incoherent. Has Dennett been open about them? Neil thinks Dennett is an immaterialist. Keiths is positive he is a physicalist. You make something third out of it. Guess Dennett is neither open or clear.

    I think Dennett has been perfectly open and clear about his background assumptions. It’s just a matter of reading him charitably.

    I haven’t gotten to the chapter on information yet so I have yet to find out why Neil thinks Dennett is an immaterialist. Keiths thinks Dennett is a physicalist because he reads more of the older Dennett (from Intentional Stance) and not the more recent work (esp “Real Patterns”).

    Don Ross, who draws heavily on Dennett, has a weird view in which what is really real is not any kind of “stuff” but patterns or structures. I’ve been told that Dennett has come to accept Ross’s interpretation, but I’m still hazy on the details.

  10. Neil Rickert,

    Yeah, I’m not happy with “natural selection as designer”, either.

    Mark Okrent (in Rational Animals) suggests that Dennett’s design stance readily invites a confusion between organisms and artifacts — in fact, the same confusion that IDists always make. He suggests the teleological stance as the right way of understanding biological explanations. I find that much better.

    It’s also worth noting that there’s a debate going on here, between Dennett and Alex Rosenberg, about whether Darwinism means explaining teleology (Dennett) or eliminating it (Rosenberg). I’m on Dennett’s side in that debate, but I’m not yet convinced he has a better handle on how to explain teleology that Millikan or Okrent do.

  11. “Metaphor for what? Metaphor as in ‘talking as if explaining something while really hand-waving it away’? This is my honest impression of Dennett when reading or listening to him.” – Erik

    Uh huh. A whole lot of hand-waving, and just so stories. Foolish posturing laughter. And MEMETICS (?!), well, but that KN won’t defend cuz he just secular Tom slices authors for his own empty-warm sophistry quilt, so unlike Joseph.

    “Neil thinks Dennett is an immaterialist. Keiths is positive he is a physicalist. You [KN] make something third out of it. Guess Dennett is neither open or clear.”

    ROTFL. Dennett is crystal clear as only philosophistry requires. ; )

    “I think [Behe] has been perfectly open and clear about his background assumptions. It’s just a matter of reading him charitably.” – KN

    Shoe foot oops.

  12. Kantian Naturalist: But it’s not as if there’s some other thing, the mind or self, that has to then look at the visual cortex to determine what I’m seeing. It’s just that the information processing (at the cortical and subcortical structures) is the seeing (at the level of personal experience).

    So his philosophy of mind holds to the mind-brain identity theory. Whence the need for a metaphor for any sort of relationship then?

    If Dennett holds to mind-brain identity, then keiths is right.

    Kantian Naturalist: Don Ross, who draws heavily on Dennett, has a weird view in which what is really real is not any kind of “stuff” but patterns or structures.

    Aren’t patterns and structures always patterns and structures of some stuff? Which one has primacy, ocean or waves? Can there be any waves without the ocean?

  13. Erik: So his philosophy of mind holds to the mind-brain identity theory. Whence the need for a metaphor for any sort of relationship then?

    Because there’s a difference in these two different ways of understanding — the vocabulary of subjective experience and the vocabulary of neuroscience. The role of the metaphor is to help us understand how these two different ways of understanding are related to each other.

    Aren’t patterns and structures always patterns and structures of some stuff? Which one has primacy, ocean or waves? Can there be any waves without the ocean?

    First, the idea of structural realism isn’t about what’s real at the level of the stuff we see and hear — it’s a claim about what fundamental physics is all about. Secondly, it’s not really a view that I understand well enough to defend. I only mentioned it because it goes to the question of whether Dennett is a “materialist” in the traditional sense of the word. When it comes to the metaphysics of science, I’m with Nancy Cartwright: anti-realism about laws and realism about causal powers. The structural realists go in the opposite direction: realism about structures and anti-realism about causation. I find that, shall we say, baffling.

    Or, to put together Dennett and Deleuze, what’s real are dynamic processes, which we adopt different attitudes and interpretations towards as being mechanistic, teleological, and/or discursive.

  14. Kantian Naturalist: When it comes to the metaphysics of science, I’m with Nancy Cartwright: anti-realism about laws and realism about causal powers. The structural realists go in the opposite direction: realism about structures and anti-realism about causation. I find that, shall we say, baffling.

    Causation is not a law/structure in the fabric of the universe?

  15. Erik: Causation is not a law/structure in the fabric of the universe?

    Cartwright thinks that causal powers are real but laws are just our idealized descriptions of their effects.

  16. Kantian Naturalist: Because there’s a difference in these two different ways of understanding — the vocabulary of subjective experience and the vocabulary of neuroscience. The role of the metaphor is to help us understand how these two different ways of understanding are related to each other.

    “User illusion” of course explains this not one bit. E.g. third-person view versus first-person view is a much better metaphor – genuinely descriptive of how we truly experience reality. Anything wrong with it? Too much common sense? What does “user illusion” add in comparison? It only seems to take away.

  17. Neil Rickert: I don’t buy the “natural selection as designer” part.

    And

    Kantian Naturalist: Yeah, I’m not happy with “natural selection as designer”, either.

    But it’s what happens! What else is there to design life?

  18. Alan Fox: What else is there to design life?

    To oversimplify, I’m inclined to say that natural selection tends to favor systems that happen to be teleological. But then it’s the teleology that does the designing.

  19. walto: OTOH, when we merely think about a tree, we lose a lot about it too.Its color, texture, hardness, smell, etc.As Kant said, even if intuitions without concepts are blind, concepts without intuitions are EMPTY.

    The tree you merely perceive is objectively real in the same way that the letters in this post are objectively real, but we do not gain an understanding of the content by merely knowing of the reality of the letters. We need to understand the whole context of the words and sentences to know what is meant by them.

    Thinking doesn’t take anything away from the objective nature of the tree, it adds the concepts which gives us the higher reality of which the tree in front of us is but one snapshot. The concepts, growth and decay, organism, etc, are not something that only exist within the human mind, they belong to the tree. Or should I say the tree is part of them.

    Anyhow, for about the fifth time, nothing whatever follows about the non-reality of the material world from your Platonic musings.

    I would say that the material world is a very limited part of reality given through our senses. Our thinking process is in a way a higher organ of sense which combines for us that which our senses have given us the erroneous view of separation. Concepts are not something which we add to reality, the are objective entities which belong to the things themselves.

  20. CharlieM: The tree you merely perceive is objectively real in the same way that the letters in this post are objectively real, but we do not gain an understanding of the content by merely knowing of the reality of the letters. We need to understand the whole context of the words and sentences to know what is meant by them.

    No. That’s a terrible analogy. Words are symbols–they have semantic content.

    CharlieM: I would say that the material world is a very limited part of reality given through our senses.

    Yes, you would say so and have. Good for you.

  21. I don’t mean to sound so harsh, Charlie, but what I’m getting at is that the fact that some particular position appeals to you is not a reason for its truth. You read it in Steiner and (except for the lamb’s bladder thing I guess), you take pretty much everything he wrote as gospel. The thing is, that’s not a reason for anybody else to believe it. The positions you are backing are just old chestnuts from the late 19th Century. They’ve been passed by long ago as largely indefensible. You don’t have any new defenses for these chestnuts, they just appeal to you.

    OK, so you ran across a philosopher you find congenial. That’s nice. I mean it. I do that sometimes too. Good for me. When that happens, it gives one a warm glow. But why do you repeat the chestnuts here? Hug yourself and go on happily. You’re not going to convince anybody with the slightest familiarity with philosophy of Platonism or Hegelianism here or anywhere else–never mind Steinerism! Steiner, as I’ve already told you was never taken seriously by anybody but cultists. An irrelevant backwater.

    It might be responded that some of the stuff and philosophers I like aren’t exactly at the top of the current academic pantheon either. That’s life.

  22. Neil,

    I should clarify that. Dennett claims to be a materialist. But I see him as effectively an immaterialist because I see his theory of mind as amounting to a fleshing out of Berkeley’s idealism (which is usually considered immaterialist).

    No doubt Dennett thinks that he is connecting his ideas to the material world. But what I have read so far (I’m currently in chapter 9), is mostly hand waving. He fails to actually make a connection.

    Dennett obviously and straightforwardly connects his ideas to the material world. How you’re managing to miss that is a mystery.

  23. KN,

    Keiths thinks Dennett is a physicalist because he reads more of the older Dennett (from Intentional Stance) and not the more recent work (esp “Real Patterns”).

    “Real Patterns” was written over a quarter-century ago, KN.

    Dennett was a physicalist then and he remains one now. From his new book:

    Francis Crick, the recently deceased co-discoverer of the structure of DNA, was another of history’s greatest scientists, and his last major piece of writing was The Astonishing Hypothesis: The Scientific Search for the Soul (1994), in which he argued that dualism is false; the mind just is the brain, a material organ with no mysterious extra properties not found in other living organisms. He was by no means the first to put forward this denial of dualism; it has been the prevailing—but not unanimous—opinion of both scientists and philosophers for the better part of a century. In fact, many of us in the field objected to his title. There was nothing astonishing about this hypothesis; it had been our working assumption for decades! Its denial would be astonishing, like being told that gold was not composed of atoms or that the law of gravity didn’t hold on Mars. Why should anyone expect that consciousness would bifurcate the universe dramatically, when even life and reproduction could be accounted for in physico-chemical terms? But Crick wasn’t writing his book for scientists and philosophers, and he knew that among laypeople, the appeal of dualism was still quite overpowering.

  24. keiths: Dennett obviously and straightforwardly connects his ideas to the material world.

    Yes, he does. But the “material world” that he connects it to is what he calls a “user-illusion” built by cultural memes with natural selection. It’s a modernized version of Berkeley’s idealism.

  25. Neil,

    Yes, he does. But the “material world” that he connects it to is what he calls a “user-illusion” built by cultural memes with natural selection.

    Absolutely not.

    He thinks our “manifest image” of the world is a user-illusion shaped by evolution, but he doesn’t think the physical world itself is such an illusion.

    The physical world came first, and “users” evolved within it.

  26. “The physical world came first, and ‘users’ evolved within it.”

    Garden variety ideological physicalism. Therefore must be ‘antagonistic’ towards *all* (smells like Marx!) religion and spirituality – Native American included – which holds Mind &/or Spirit ‘came first’ too; not over-against Physical World, but together with it.

  27. Gregory,

    I am sincerely looking forward to the day when you finally, bravely make an actual argument.

    Will today be that day, by any chance?

  28. keiths: He thinks our “manifest image” of the world is a user-illusion shaped by evolution, but he doesn’t think the physical world itself is such an illusion.

    That’s where I see an inconsistency. If he applied his same arguing strategy, he ought to see the scientific image as a “user-illusion” optimized for scientific users. But, instead, he seems to take the scientific image as if it were the world in itself.

  29. Neil Rickert: That’s where I see an inconsistency.If he applied his same arguing strategy, he ought to see the scientific image as a “user-illusion” optimized for scientific users.But, instead, he seems to take the scientific image as if it were the world in itself.

    Now I see the problem you’re pointing to. If one is an “idealist” about the manifest image, then how can one justifiably be a “realist” about the scientific image?

    I think the problem needs to be turned around, though: it’s actually realism about the scientific image that leads to idealism about the manifest image. That is, the dialectic here would go something like this:

    1. realism about the manifest image (our default, ordinary starting point of inquiry), which leads to
    2. realism about the scientific image (over the course of inquiry from antiquity to today), which leads to
    3. idealism about the manifest image (when we explain the manifest image in terms of the scientific image).

    This is a narrative that should be pretty familiar to those who have studied where Descartes fits into the history of Western philosophy: he ‘mentalized’ experience in order to fit subjectivity into the world as explained by physics.

    Years ago, Dennett coined the term “Cartesian materialism” to refer to people who think that we can identify the res cogitans with the brain without having to rethink the whole enterprise. (John Searle is a paradigm of this strategy.) It would be deeply ironic if Dennett himself turned out to be a Cartesian materialist by another route!

  30. Kantian Naturalist: I think the problem needs to be turned around, though: it’s actually realism about the scientific image that leads to idealism about the manifest image.

    Yes, I suppose that’s Dennett’s reasoning.

    For myself, I have a pretty good idea about how we come up with both the manifest image and the scientific image. And it is really the same basic methodology. My preferred view is to be take both to be pragmatic constructs.

    Dennett probably thinks he is taking the manifest image to be something like what I would consider a pragmatic construct. But all of his hand waving argumentation about memetics and natural selection leaves me seeing it as closer to idealism.

    Perhaps I should explain the difference there. I see us building both the manifest image and the scientific image based on our interactions with the world. The scientific image is different from the manifest image, because the scientists interactions are very different from the ordinary person’s interactions.

    Berkeley’s idealism was that we receive percepts and invent an ideal world out of them. That seems to be derived from a Cartesian view of perception as passive. Dennett also seems to take perception as passive, rather than interactive, though he sees natural selection in a world of memetics as part of the picture. So Dennett’s view seems to me to be closer to idealism.

  31. Neil Rickert,

    I still haven’t gotten very far, since I’m trying to finish a paper that’s due next week. But I was told that Dennett tries to make use of Gibson in talking about the manifest image. I think I take Gibson to be a weak or pragmatic realist about the manifest image. (It’s not as if animals create or invent their affordances!) So if Dennett is committed to an idealism about the manifest image, there might be a tension with his use of Gibson.

    (Sellars was also an idealist about the manifest image, also due to his scientific realism. It’s one of the major flaws in his system.)

  32. Kantian Naturalist: But I was told that Dennett tries to make use of Gibson in talking about the manifest image.

    Yes, he does. But not very effectively (in my opinion).

    I think I take Gibson to be a weak or pragmatic realist about the manifest image.

    That’s probably about right. Gibson is on record as saying that he’s a naive realist. But what a scientist means by “realism” isn’t always the same as what a philosopher means. I think “pragmatic realist” is probably a more appropriate term for Gibson.

    The main flaw that I see in Dennett’s argument, is his Cartesian view of perception. That is to say, his view of perception as passive. This is actually a very common problem (or assumption). People look around and seem to see information everywhere. So they assume that the world is full of information waiting to be used. But I think that’s wrong.

    Yes, they see information everywhere, but that’s because their perception is doing a great job of giving them information. Actually getting information about the world is hard work, and it is a mistake to assume the information is free for the taking.

    Gibson understood this. His wife wrote a book on “perceptual learning”. I take it that, for Gibson, an affordance is something that arises from perceptual learning. Dennett, instead, credits natural selection as providing affordances.

  33. Neil Rickert,

    It certainly would be a disaster for his view if Dennett were trying to be both a Cartesian and a Gibsonian about perception!

  34. keiths:

    He thinks our “manifest image” of the world is a user-illusion shaped by evolution, but he doesn’t think the physical world itself is such an illusion.

    Neil:

    That’s where I see an inconsistency. If he applied his same arguing strategy, he ought to see the scientific image as a “user-illusion” optimized for scientific users. But, instead, he seems to take the scientific image as if it were the world in itself.

    No, he doesn’t, and neither did Sellars. Hence the word ‘image’ in ‘scientific image’.

  35. KN,

    Years ago, Dennett coined the term “Cartesian materialism” to refer to people who think that we can identify the res cogitans with the brain without having to rethink the whole enterprise.

    No, “Cartesian materialism” refers to the idea that there’s a place within the brain where it all comes together — a “Cartesian theater” in which the contents of consciousness are projected on a “screen” for viewing by an inner homunculus. This creates an obvious regress.

    The mere idea that the brain is the site of consciousness does not suffer from that problem.

  36. Neil:

    Berkeley’s idealism was that we receive percepts and invent an ideal world out of them. That seems to be derived from a Cartesian view of perception as passive. Dennett also seems to take perception as passive, rather than interactive, though he sees natural selection in a world of memetics as part of the picture. So Dennett’s view seems to me to be closer to idealism.

    You’ve packed a lot of confusion into one paragraph.

    First, you’re badly misunderstanding Berkeley’s idealism here:

    Berkeley’s idealism was that we receive percepts and invent an ideal world out of them.

    His idealism was quite different and much stronger than that. I can only suggest that you do some reading on the topic.

    Second, Dennett does not see perception as passive, but as an active process of extracting relevant information from our sensory inputs:

    It is often noted that the brain’s job in perception is to filter out, discard, and ignore all but the noteworthy features of the flux of energy striking one’s sensory organs. Keep and refine the ore of (useful) information, and leave all the noise out. Any nonrandomness in the flux is a real pattern that is potentially useful information for some possible creature or agent to exploit in anticipating the future. A tiny subset of the real patterns in the world of any agent comprise the agent’s Umwelt, the set of its affordances. These patterns are the things that agent should have in its ontology, the things that should be attended to, tracked, distinguished, studied. The rest of the real patterns in the flux are just noise as far as that agent is concerned. From our Olympian standpoint (we are not gods, but we are cognitively head and shoulders above the rest of the creatures), we can often see that there is semantic information in the world that is intensely relevant to the welfare of creatures who are just unequipped to detect it. The information is indeed in the light but not for them.

    Third, Dennett’s views could hardly differ more from Berkeley’s. Dennett is an atheist; Berkeley was a theist. Dennett is a physicalist; Berkeley was an immaterialist. Dennett thinks that matter gives rise to mind: Berkeley thought that mind gave rise to (the illusion of) matter.

  37. It’s no wonder you scoff at Dennett. If I misunderstood him as badly as you do, I’d scoff too.

    Dennett-as-interpreted-by-Neil is a hopeless mess, but the problem lies with the interpreter, not with the source material.

  38. “Dennett’s views could hardly differ more from Berkeley’s. Dennett is an atheist; Berkeley was a theist. Dennett is a physicalist; Berkeley was an immaterialist. Dennett thinks that matter gives rise to mind: Berkeley thought that mind gave rise to (the illusion of) matter.”

    Berkeley & Descartes were both Christians, i.e. Abrahamic monotheists.

    Sellars sold his soul to _________ (scientism, materialism, spiritual agnosticism, etc.) – a bad bargain. KN likes that price. But if you folks can’t even count the obvious sheep & goats, it’s kinda hard to think you know what kool aid you’re even drinking.

    It’s clear most atheists can’t distinguish how worldview even matters *at all* in the (philosophy of life, science, mind, etc.) ideas these people had! ‘Oh, that was Descartes’ idealism!’ claims the agnostic. Actually, it was his theism, his worldview, the source of meaning and purpose in his life, his (swerving) belief, not just his (empirical) knowledge.

    “(Sellars was also an idealist about the manifest image, also due to his scientific realism. It’s one of the major flaws in his system.)”

    This is richly bankrupt; a pretzelled philosophist in the USA who basically idolises Sellars, now notes “one of the major flaws in his system” and now criticises his master saying he was ‘also an idealist.’ KN wants to get rid of ideals and idealism ENTIRELY; he wants to be more materialistic, more naturalistic, more atheistic (or with KN, likely just more agnostic), more un-religious, more Science-worshipping, more heathen, further away from the beliefs of his own ancestors, more obnoxious & crude … even than Sellars! It is obvious now that is KN’s goal; not to come closer to anything Ideal, but simply to run away from ideality and G_d as far as possible, and the ‘running away’ is the lynchpin of his philosophistic life strategy, disguised as ‘honestly just seeking naturalistic gnosis’.

    Yes of course claims the philosophist, one can believe in ideals, in ‘fairy tales’ (‘oh brilliant Dawkins, we laugh with thee, at believing subhuman idiots for their faith’) and in anything that isn’t actually ‘physically/materially/naturally real’ that one wants. Ideals themselves are not ‘unreal’ (except for in the really bad moments!) according to agnostics. But turning ideals into idealism is wrong, while turning nature into naturalism is just fine, according to the poisonous philosophy teacher at TSZ. What a set of confused ideologies!

    It’s come to be obvious that one of KN’s favorite games in life is to identify points (only material, physical or natural ones) and then to draw targets around them, claiming that all he hits with his philosophistry is bullseyes! Sellars was a sellout and so is anyone who follows his despair trail into ideological ignominy.

    Help these idolaters of Nature improve their awareness that Sophia takes a different home than in merely natural caves of biological substance and physical matter. They are idolising the created logos, unaware that they too are written in the Narrative of Earth’s history, (being) written together with Logos. Why they want out of the Story, other than just wanting to steal the pen from the Author?

  39. I’m now about 1/3 through From Bacteria to Bach and Back, and I’m beginning to see Neil’s worries about “semantic information”. I’m not sure I agree.

    Neil seems to think that, according to Dennett, semantic information is something that is out there any way, and the organism just has to passively detect it.

    I don’t think that’s Dennett’s view at all.

    I think he’s introducing semantic information as a way of characterizing the species-specific manifest images, or what the ontology is for each kind of critter. The ontology of beavers will be very different from the ontology of free frogs, etc. Beavers and tree frogs have very different ways of engaging with their environments, and those engagements will involve different ways of finding things meaningful. A tree is meaningful for a beaver (as a potential source of wood for a dam) in a very different way in which it is meaningful for a tree frog (as a high place from which to sing).

    I don’t see there’s any problem with Dennett allowing that semantic information is partially constituted by what an organism brings to the table, so to speak. And that means he’s not committed to a passive uptake model of perception, either. In short, I don’t see any commitments that interfere with his use of Gibson.

    It is true that Dennett, unlike Gibson, thinks that we can appeal to natural selection to explain why organisms are responsive to the affordances to which they are responsive. But that’s just a different question than the ones that interested Gibson. Gibson was interested in explaining what it is to perceive. Dennett is interested in explaining how organisms have come to perceive what it is that they do. He’s giving ecological psychology an evolutionary twist.

    That said, I’m not happy with “semantic information” as a term here. It seems to equivocate between linguistic and non-linguistic ways of interpreting the world, and it’s not always clear if Dennett is talking about representings or representeds (this is where Neil got confused, and understandably). I’d be happier talking about neurodynamic states as representing affordances, and I’m not sure where ‘semantic information’ would fall there — would it be the representing relation, or the representing, or the represented?

    In short, I don’t think he has the passive view of perception or the quasi-idealist commitment (the world is made of information) that Neil thinks he does, but I do think he’s not being as clear as he needs to be.

  40. KN,

    It is true that Dennett, unlike Gibson, thinks that we can appeal to natural selection to explain why organisms are responsive to the affordances to which they are responsive.

    You’re saying that Gibson denies the role of natural selection in explaining our responses to affordances?

  41. KN,

    That said, I’m not happy with “semantic information” as a term here. It seems to equivocate between linguistic and non-linguistic ways of interpreting the world, and it’s not always clear if Dennett is talking about representings or representeds (this is where Neil got confused, and understandably).

    “Semantic information” is the standard term, going at least back to Carnap if not beyond. Dennett is simply following convention.

  42. Kantian Naturalist: That said, I’m not happy with “semantic information” as a term here. It seems to equivocate between linguistic and non-linguistic ways of interpreting the world, and it’s not always clear if Dennett is talking about representings or representeds (this is where Neil got confused, and understandably).

    No, I’m not confused.

    There’s no such thing as “semantic information.” I used to think that there was such a thing, and that it was what was important for understanding cognition. Then I read Dretske’s “Knowledge and the flow of information”. And that was when I realized that it was unworkable — which I’m sure is the opposite effect to what Dretske was trying to achieve.

    I settled on Shannon information as what is important. To his credit, Dennett seems to accept that.

    The point, though, is that Shannon’s theory of information does not deal with semantics. Information is ink marks on paper or pixels on the screen or electric or magnetic charges or holes in paper, etc. There isn’t any semantics in the information. The semantics are at the end points. The sender of information has a semantic story. But he encodes that as syntax. The information stream is only syntax. The receiver of the information puts the semantics back. The semantics can be “found” in the conventions that establish the means of communication. The semantics is in how the information is to be used — which fits Wittgenstein’s “meaning is use”.

    All information is intended to be used. So it makes no sense to single out some information as semantic. If the information is not intended to be used, then on what basis can we call it “information”?

    To summarize: Information is a syntactic stream. Semantics is not present in that stream. The semantics can be found at the end points — the sender and receiver of the information. And the semantics depends heavily on the conventions that establish the communication channel. To understand meaning, one must understand the role of convention. Unfortunately, there’s a tendency for philosophers to be allergic to the idea of conventions.

  43. Neil Rickert,

    I see your point that there’s no meaning in Shannon information, but isn’t that just why Dennett doesn’t use that concept?

    What seems to be important to Dennett is that animals can find things as meaningful to them. I have no commitment to the concept of “semantic information” as a way of talking about this. I myself would rather put this point in terms of intentionality, but Dennett can’t use that word because he is committed to a pragmatic anti-realism about intentionality.

    One worry that I do have about Dennett’s general project here is that he doesn’t seem clear about when he’s talking about representings and when he’s talking about representeds. (Sellars somewhere remarks that Berkeley’s theory of perception depends on this conflation, which is why Berkeley thinks that what we directly perceive are our own sensations.) Is “semantic information” what is represented by a brain? Or is how a brain represents its Umwelt? I don’t find a clear answer to this, and it bothers me!

    On the one hand, his use of Bayesian hierarchical architectures as a theory of how brains function suggests that what these architectures do is represent the local affordances. And that seems basically right to me. But then, if affordances are semantic information, then the information is just out there in the world, waiting to be detected. And that seems really problematic.

    The problem here is that Gibson himself was vague on the ontology of affordances. As he put it, by ‘affordance’ “I mean by it something that refers to both the environment and the animal in a way that no existing term does. It implies the complementarity of the animal and the environment” (Ecological Approach to Visual Perception, p. 119 in my edition (second paragraph of Chapter 8)).

    Subsequent ecological psychologists have tried to figure out if affordances are what is perceived, or if they are how the animal perceives it. Tony Chemero (in Radical Embodied Cognitive Science) suggests that affordances are relations between sensorimotor abilities and environmental features. But Chemero is far more hostile towards the concept of ‘representation’ than I am.

    I’m more comfortable with the idea that brains do represent their environments, that those representations are basically affordance detecting and action guiding, and that representations are implemented in (roughly) Bayesian architectures of the sort described by Howhy, Frith, and others. And I think (but am not sure!) that this will put me at odds with Dennett here.

    I’ll have a better sense of where I stand vis-a-vis Dennett after I’ve finished the book and had a chance to talk about it with bona fide philosophers of cognitive science (of which I am not one).

  44. Kantian Naturalist: I see your point that there’s no meaning in Shannon information, but isn’t that just why Dennett doesn’t use that concept?

    But he does use that, probably because his computationalism requires it.

    Checking the index, there are lots of reference to “Shannon” and “Shannon information”. Unfortunately the references give page numbers, but my kindle copy only has kindle locations. This is a huge frustration with the way Amazon does kindle versions.

    What seems to be important to Dennett is that animals can find things as meaningful to them.

    Fair enough. But Dennett wants that to be “semantic information” in the world. By contrast, I see that as coming from perception. The idea that the world is full of semantic information is mostly an illusion because perception does its job so effectively. But it is perception that constructs information about the world.

    One worry that I do have about Dennett’s general project here is that he doesn’t seem clear about when he’s talking about representings and when he’s talking about representeds.

    I’m taking that you be your way of wording the same kind of objection that I had.

    What it really all boils down to, is that a theory of consciousness or a theory of cognition requires a good theory of perception as a prerequisite. And Dennett does not have one. He pays lip service to Gibson, but it seems to me that he misses much that’s important in Gibson’s account.

    The problem here is that Gibson himself was vague on the ontology of affordances.

    Yes, this is indeed a problem with Gibson. It seems to me that Gibson did not actually have an adequate theory of perception. He pushed us in the right direction. But we need to go further than he did.

  45. Neil Rickert: Checking the index, there are lots of reference to “Shannon” and “Shannon information”. Unfortunately the references give page numbers, but my kindle copy only has kindle locations. This is a huge frustration with the way Amazon does kindle versions.

    Then you just do a word search for ‘Shannon’ no?

  46. Neil Rickert: But he does use that, probably because his computationalism requires it.

    Yes, I should have said that he doesn’t use Shannon information to characterize the relation that animals have with their environments. I’m not sure if Bayesian theories of the brain require that neurons or neuronal assemblies deal in Shannon information. So I’m not quite sure if his neurocomputatioinalism requires it. But you might be right.

    But Dennett wants that to be “semantic information” in the world. By contrast, I see that as coming from perception. The idea that the world is full of semantic information is mostly an illusion because perception does its job so effectively. But it is perception that constructs information about the world.

    I see your point, and I think it’s ambiguous as to whether Dennett allows for some degree of ‘construction’ of semantic information or if he has more of a passive detection model. His interest in Bayseian theories of neurocomputation would certainly push him much more in the former direction, but it’s not clear if he realizes that. By contrast, Clark’s Surfing Uncertainty is very clearly a theory of perception in which the brain actively constructs what is perceived, with constraint from sensory transducers on what can be constructed.

    I’m taking that you be your way of wording the same kind of objection that I had.

    Yes, I think so.

    What it really all boils down to, is that a theory of consciousness or a theory of cognition requires a good theory of perception as a prerequisite. And Dennett does not have one. He pays lip service to Gibson, but it seems to me that he misses much that’s important in Gibson’s account.

    That all seems right to me. We might quibble on how Gibsonian he is, but there’s a lot more in Gibson that Dennett could use and doesn’t.

    Yes, this is indeed a problem with Gibson. It seems to me that Gibson did not actually have an adequate theory of perception. He pushed us in the right direction. But we need to go further than he did.

    Do you have any suggestions? Are there psychologists working in a post-Gibsonian tradition that you recommend? I’m always looking for new things to feel guilty about not reading!

  47. walto: Then you just do a word search for ‘Shannon’ no?

    Yes, that works. But it would be better if the numbers shown in the index were also available while reading.

Leave a Reply