Incorrigibility and Imageless Thought: Reply to Judith Economos

(Reply to Judith Economos)

That I know what I am thinking is setting me apart from a computer or a zombie that does not know what it is thinking and therefore is only “thinking”. You see, I do understand that much. It just seems to me important not to confuse this with feeling, but all the words I might use, like “aware” and “conscious”, are [you say] bad words, dishonest and weaselly. It is true that those words are vague, but so is Feeling, as you use it, and the problem is that what goes on in our minds is private to us and not really shareable. We have to communicate it to each other with words whose meanings to the other can at best be guessed at by analogy with states we think are comparable, working from the most obvious behavioristic criteria to the more delicate shades of inward state, events, processes, efforts, and, er, well, awarenesses. Since we cannot directly compare your mental target with mine, we just can’t know if we are really on about exactly the same thing, or even if there is such a thing in my mind as there is in yours. That is why I am comfortable in my intransigence about what I feel or don’t feel, and why I think argument is futile. Each of us is (as it were) sitting in a closed room considering objects for which we have never had a common name, nor common adjectives, and trying to see if the other has similar objects in his room.”

I think the debate about whether there are unfelt thoughts may be a throwback to the introspectionists’ debates about whether there is “imageless thought.”

The big difference is that the introspectionists thought that an empirical methodology was at stake, one that would allow them to explain the mind. We now know that there was no such empirical methodology, and no explanation of the mind was forthcoming, not just because introspection is not objective, hence not testable and verifiable, but because it does not reveal how the mind works.

(This conclusion was partly reached because it was not possible to resolve, objectively, the rival claims of introspective “experimentalists” as to whether imageless thought did or did not exist. The outcome was an abandonment of introspectionism in favor of behaviorism, until it turned out that that could not explain how the mind works either, which led to neuroscience and cognitive science.)

The difference here, is that we are debating about whether or not there are unfelt thoughts while at the same time agreeing that introspection is neither objective nor explanatory.

The disagreement is, specifically, about what it means to say that we are conscious of something, when that being-conscious-of something does not feel like anything.

The reason I keep calling “consciousness” and “being conscious of” something weasel-words is that if things are going on in my mind that somehow are different yet do not feel different then it is not at all clear how I am telling them apart, or even how or what I am privy to when I say I am “conscious of” them despite the fact that I don’t feel them in any way.

It seems to me that the notion of “unconscious thought” and “unconscious mind” is as arbitrary and uninformative as “unfelt thought” — or, put in a less weaselly way that bares the incoherence: unfelt feelings.

Among other things, this makes it more obvious that “thought” and “thinking” are themselves weasel-words (or, at best, phenomenological place-holders). For apart from the fact that they are going on in my head, it is not really clear what “thoughts” are: we are waiting for cognitive science to tell us! (We keep forgetting that “cognition” just means thinking, knowing, mentation — weasel words, all, or at least vague if not vacuous until we have a functional — i.e., causal — explanation of what generates and constitutes both them and, more importantly, what we can do with them, apart from thinking itself.)

Now explaining how we can do what we can do is these days called the “easy” problem. What, then, is the “hard problem,” and what is it that makes it hard?

My own modest contribution is just to suggest that (1) the hard problem is to explain how and why thought (or anything at all) is felt, and that (2) it is hard because there is no causal room for feeling in a complete explanation of doing.

So where is the disagreement here? Let us assume we can agree that explaining unconscious thought — i.e., those things going on inside the brain that play a causal role in generating our capacity to do what we can do, but that we are not conscious of — are not hard problems but easy problems. They may as well be going on inside a toaster. The rest is just about how they generate the toaster’s capacity to do whatever it can do. And of course we are not privileged authorities on whatever unconscious “thoughts” may be going on in our heads, because we are not conscious of them.

So all I ask is: What on earth do we really mean when we speak of having a thought that we are conscious of having, but that it does not feel like anything to have!

You (rightly) invoke “incorrigibility” if I venture to doubt your introspections, but, to me, that justified insistence that only you are in a position to judge what’s going on in your mind [as opposed to your head] derives from the fact that only the feeler can feel (hence know) what he is feeling: what the feeling feels like. But apart from that, it seems to me, there are no further 1st-person privileges. One can’t say: I don’t feel a thing, yet I know it. On what is that privileged testimony based, if it is not the usual eye-witness report? “I didn’t see the crime, but I know it was committed?” How do you “know”? In the case of thought, what does that “knowing” consist, if not in the fact that you feel it (and it feels true)?

There is no point referring to objective evidence here. I can “know” it’s raining in the sense that I say it’s raining and it really is raining. I have then made a true statement, just as a robot or a meteorological instrument could do, but that has nothing to do with the mind or the hard problem (and, as the Gettierists will point out, it’s not really “knowing” either!). It’s just back to the easy problem of doing. What makes it hard is not just that that “thought” is going on inside your head, but that it is going on in your mind, which is what makes it mental (another weasel-word).

Hence (by my lights) the only thing left to invoke to justify calling such a thing mental, and hence privileged, is that it feels like something to think it, and you are feeling that thing, and you’re the only one in the position to attest to that fact.

Yet the fact you are attesting to here (as an eye-witness) is the fact that it doesn’t feel like anything at all to think something! And that’s why I think I am entitled to ask: Well then what is a thought, and how do you know you are thinking it? In what does your “consciousness of” it consist, if not that it feels like something to think it? (How can you be an eye-witness if your testimony is that you didn’t see a thing?)

Galen Strawson has invoked the less weaselly (but nonetheless vague and hence still somewhat weaselly) notion of “experiencing” something, as opposed to “feeling it.” But that just raises the same question: What is an unfelt “experience” (an experience it doesn’t feel like anything to have)? Galen invokes “experiential character” or “phenomenal quality” etc. But (to my ears) that’s either more weasel-words or just euphemisms because — for some reason I really can’t fathom — one can’t quite bring oneself to call a spade a spade.

To be conscious of something, to experience something, to sense something, to think something, to “access” something — all of those are simply easy, toaster-like “information”-processing functions (“information” can be yet another weasel-word, if used for anything other than data, bits) — except if they are felt, in which case all functional, causal bets are off, and we are smack in the middle of the hard problem: why and how do feel? (Ned Block‘s unfortunate distinction between “phenomenal consciousness” and “access consciousness” is incoherent precisely because unfelt “access” is precisely what toasters have — hence “access” too is of the family Mustelidae, and the PC/AC distinction is bared as the attempt to distinguish felt feelings vs. unfelt “feelings”…)

Amen — but with no illusions of having over-ridden anyone’s (felt) privilege to insist that they do have unfelt [imageless?] thoughts — thoughts of which they are in some (unspecified) sense “conscious” even though it does not feel like anything at all to be conscious of them…

Linguistic Non Sequiturs

(1) The Dunn et al article in Nature is not about language evolution (in the Darwinian sense); it is about language history.

(2) Universal grammar (UG) is a complex set of rules, discovered by Chomsky and his co-workers. UG turns out to be universal (i.e., all known language are governed by its rules) and its rules turn out to be unlearnable on the basis of what the child says and hears, so they must be inborn in the human brain and genome.

(3) Although UG itself is universal, it has some free parameters that are set by learning. Word-order (subject-object vs. object-subject) is one of those learned parameters. The parameter-settings themselves differ for different language families, and are hence, of course, not universal, but cultural.

(4) Hence the Dunn et al results on the history of word-order are not, as claimed, refutations of UG.

Harnad, S. (2008) Why and How the Problem of the Evolution of Universal Grammar (UG) is Hard. Behavioral and Brain Sciences 31: 524-525

Dawkins on Miracles

Re: http://richarddawkins.net/articles/91-to-live-at-all-is-miracle-enough

Can’t quite agree with Dawkins.

Feeling lucky, like feeling happy, is a mental state. Nonexistent entities (and indeed existent but nonliving entities, like teapots) are neither lucky nor unlucky, nor happy nor unhappy. So how can an existent, living entity be luckier than they? It’s like saying I’m greener than F# Major, or that smells are more concave (sic) than abstract…

And, setting aside the euphoria, the lucky ones are the ones who didn’t end up in Auschwitz (except if their loved ones ended up in Auschwitz).

Lucky to be alive? If I take an average over all the sentient creatures alive on the planet today, I’d say most of them are not so lucky; the lucky ones are maybe the ones that eat rather than get eaten, but then the numbers are beginning to dwindle for Dawkins’s ecstatic numerology…

What’s left that can be said and makes any sense is just a rather banal tautology: When you’re having a good day, you’re having a good day… Enjoy it while it lasts (and don’t think too hard!).

As for Dawkins’s hopes of posthumous pleasure from the people who are still partaking of his prose in 2111 — ask him again how he’s feeling about that in 2111…


PS and whether you prefer sci or sci-fi is really down to a matter of taste…

Well-meaning time

Well-meaning time
and insouciant memory
have compounded
age and mortality’s
matricide,
making off with
even the immediacy
and immanence
of her loss,
the ubiquity
of her absence,
sparing only
its perpetuity,
immutable,
immune
to appeal.

Not his defiant

No, it was not his defiant
but unmistakably wounded
words that revealed at once
what an unspeakable travesty
this all had been.

It was his consort
and coeval
shuffling out laboriously
with her walker.

Languages, Kinds and Kinds of Language

Commentary on: Savage-Rumbaugh, S. (2011) Human Language — Human Consciousness On the Human. January 2011

There is much to agree with in Sue Savage-Rumbaugh’s reflections on human and nonhuman primates. Sue has probably spent more real time rearing and observing our closest hominoid cousins than any other human being has done. Bonobos are indeed astonishingly intelligent and capable, and become still more human-like when reared in daily contact with humans.

But there is one radical inference Sue makes that it will be hard for most people to agree with: Bonobos have acquired a (“kind of”) language: “the kind of language they have acquired — even if they have not manifested all major components yet — is human language as you and I speak it and know it.”

Let us reflect for a moment on languages and kinds: Humans have many kinds of languages, but there is one thing all those languages have in common: Anything you can say in one of them, you can say in any of the others. And anything and everything that can be said at all can be said in any one of them. Not necessarily in the same number of words (and you might have to define a few new ones); not necessarily equally elegantly; but anything and everything.

(Some readers may find the foregoing assertion as hard to agree with as Sue’s that bonobos have language. I suggest they test their intuitions by finding a counter-example: either a human language in which you can say this, but not that; or something you cannot say in any human language. Until someone comes up with such a counter-example, I will provisionally take it to be a true property of language — not human language, but language itself — that if you have it, you can say anything and everything that can be said [or gestured or written, as the modality need not, of course, be vocal], and if not, not.)

Neither Kanzi nor his kin or kind can say everything (or anything faintly near everything). I accordingly conclude that they cannot say anything. They can do a lot — far more than anyone ever imagined nonhuman primates could do. And what they can do includes an astonishing amount of intelligent, purposive communication with humans, using some of the same components to communicate that humans use for language: They can communicate purposively by sending and receiving computer images as well as by responding to human spoken sounds. But the undeniable fact is that — no matter how much linguistic understanding we attribute to them — they cannot enter into this “conversation” we are having in this Forum, not even into a rudimentary approximation to it, whereas any speaking human being, using any (spoken or gestural or written) language, can; even a child.

And the most likely reason for that is that bonobos cannot express or understand propositions as propositions (statements with a truth-value: true or false), otherwise they could express and understand any and every proposition; and what they do understand and express when we think they are understanding propositions is not what we think it is. The “narrative” gloss that we project on it is more like the sound-track of a silent movie — one generated by our own language-prepared brains, irresistibly “narratizing” (as Julian Jaynes dubbed it) every scene we see, but especially every communicative interaction with another mind (and sometimes even, frustrated, with malfunctioning machines). We are inadvertently projecting propositionality even where it is absent.

(This is not merely about “aboutness” in the sense Sue intends it — not just about the intended object or “referent” of attention, shared attention, pointing, gesturing, or miming; it is about making and meaning subject/predicate assertions with truth values. For that is what gives language its unbounded expressive power, allowing us to express any and every proposition. Nor does that have anything to do with “consciousness,” i.e., feelings, which bonobos, and of course most — probably all — animals have; nor with the “self/other” distinction, which many species can make, to varying degrees, in the practical, sensorimotor sense, but none but ourselves can make in the linguistic sense.)

It is hard to understand why creatures as stunningly intelligent and capable as bonobos cannot acquire language. I’d say that that inability was a more remarkable and puzzling fact — begging to be understood and explained — than even the remarkable intellectual and communicative feats that bononbos have indeed proved capable of mastering; for of course it is precisely how very much they can do that makes what they can’t do all the more perplexing: Why can’t they say anything and everything, given what they can demonstrably do, if it’s really language?

Sue’s reply is: “cultural differences”; and with Teco she’s hoping to close the cultural gap. But with any human child, the gap is closed almost immediately, in infancy, once the child acquires (any) natural language. (Some unnatural languages can be designed that defy the child’s language-learning capacities, but that’s another matter; even those artificial languages still have the full expressive power of any natural language.)

So until Teco can join this conversation, I will assume that what is going on is a good deal of hopeful, irresistible propositional over-interpretation (by humans) of some remarkable cognitive and communicative capabilities and performance (by bonobos) — but not a conversation, not propositions, and hence not language.

Harnad, S, (2010) From Sensorimotor Categories and Pantomime to Grounded Symbols and Propositions. In: Handbook of Language Evolution, Oxford University Press.

______. (2010) Symbol Grounding and the Origin of Language: From Show to Tell. In: Origins of Language. Cognitive Sciences Institute. Université du Québec à Montréal, June 2010.

Morals, Mores and Mood: On Saying and Doing What Feels Right

Commentary on: Kobe, Joshua (2010) Do People Actually Believe In Objective Moral Truths? On the Human. December 2010

Neither logical judgments nor moral judgments would have much power if they were not underwritten by feelings. The proposition “P & not-P” is not only formally wrong, but it (usually) feels wrong. Same with the proposition “lying is wrong.” If it did not feel wrong to lie (mostly, for most people, in most circumstances), we’d probably lie a good deal more (as sociopaths do) — and information conveyed through language would become a lot less reliable: Perhaps even the adaptive value of language itself — the advantages it conferred on our ancestors: the advantages that gave birth to language (hence culture) and embedded it in our genomes and hence our brains — would have been nonexistent or insufficient if telling hadn’t been coupled with a sufficiently strong propensity toward telling the truth rather than lying (as CL Dodgson‘s Tortoise tried to explain to Achilles).

So we believe certain things are true because they feel true (not just because they are true). And we believe certain things are wrong because they feel wrong. This is not unlike the reason we like sweets (evolutionary biology’s favourite example of proximal causation): not because they raise our blood-sugar and give us energy but because they taste good. And they taste good because the tendency to seek and eat sugar was adaptive for our ancestors (when sugar was still rare: unlike now, when it is available everywhere and excess causes caries and diabetes.)

And what feels right and wrong is no doubt more influenced by our experiences and culture than it is by logic, or even by empirical evidence (another taste that is partly wired-in and partly acquired).

So it is unsurprising that our moral tastes coincide more with our own culture than someone else’s. It is also unsurprising that in hypothetical or even counterfactual mode — “Imagine extraterrestrials with feelings and practises very different from our own…” — we make different “moral judgments” on others’ behalf. It’s rather like asking “If candy tasted bitter rather than sweet, would one feel like eating it?” The prevailing feeling in such hypotheticals is the more abstract one, of formal logical consistency with the premises rather than about what you yourself find tasty.

One might have added (hypothetically) that in a hypothetical, insentient-zombie world, anything goes, morally speaking: nothing feels (hence is) wrong or right, since nothing feels like anything at all. (But universal sociopathy does not seem to be an evolutionarily stable strategy in the real world, even if behavioural propensities replace feelings.)

So much for “objective moral truths.” The rest is just about either (1) practical rules for people to agree to abide by and enforce compliance with or (2) ways to influence or manipulate people’s feelings about what’s right and wrong.

Carroll, L. (1895) What the Tortoise said to Achilles Mind 4(14): 278-280

Harnad, S, (2010) From Sensorimotor Categories and Pantomime to Grounded Symbols and Propositions. In: Handbook of Language Evolution, Oxford University Press.

________ & Scherzer, P. (2008) First, Scale Up to the Robotic Turing Test, Then Worry About Feeling. Artificial Intelligence in Medicine 44(2): 83-89


Footnote: I am a vegan, and the image of the mutilated rabbit, which I display very reluctantly, is in no way intended to be facetious or flippant: It’s dead serious. 

Providence

Mit van mit kivánni még
Ily áldott időben? –
Adjon Isten, ami nincs,
Ez uj esztendőben…

What’s there,
This blessèd time,
To want?
May God
What’s not
This year ordain…

Arany János (1853)

Who’s there,
This blessèd time,
No more?

May God
(Who’s not)
His “gifts” retain…

The King’s English

Apart from not capturing the King’s English — either then or now — The King’s Speech does rather simplify and even trivialize speech defects, speech therapy, and, no doubt, George VI’s struggle. But the two principal male (and female) roles are well (if inauthentically) played. (Colin Firth mastered the royal mispronunciation of “r,” but not the Windsor accent.) Derek Jacobi, however, is simply dreadful as the A of C, and Timothy Spall’s face and facial expressions were terrible as Churchill. The anachronisms — e.g., I rather doubt that the royal family’s locution “the firm” dated as early as the 1930’s, but the urge to slip it in prevailed — are sometimes intrusive, and I’d certainly hate to be one of the portrayed parties having to view this. Nevertheless, overall, the film works.

But two obvious strategies to make the task of public speaking (and especially public broadcasting) easier for the king were never tried. And leaving us to wonder why cannot but reduce the drama of the struggle:

(1) Why insist that broadcasts be done live, rather than recorded in advance (with multiple takes and edits)?

(2) If playing loud music in headphones while reading a speech inhibited the stammer, why not use that during the speech-making, rather than only as therapy?