The Non-Cognitive Mind

The so-called cognitive mind is said to be subject to the representational theory of mind (RTM).[1] This theory has a number of elements chief among which is that thoughts have compositional structure—roughly, their conceptual content mirrors the compositionality of language. That means that they have propositional content—the kind of content that sentences have. The general idea is that minds have different compartments, not all of which are subject to RTM. These don’t get discussed much, and are seldom even specified. But they exist and stand in contrast to the cognitive mind. Strictly, this latter mind excludes perception, because perceptual impressions are not structured like sentences; their content is not conceptually compositional (e.g., you don’t see that not all sugar is white or smell that milk doesn’t smell like roses, but you can think these things). So, the non-cognitive mind consists of everything else: sensations, emotions, needs, desires, images, after-images, personality traits, volitions, experiential memories. These mental goings-on are not deemed compositional in the indicated sense—not propositionally productive. Thus, RTM doesn’t apply to them. They may interact with the cognitive mind and even be infiltrated by it, but their essence (“architecture”) is different, markedly so. They must fall under some other theory; one wouldn’t want to adopt a linguistic theory of their nature—a language of emotion, say, comparable to a language of thought (LOT). Emotions exist way down the phylogenetic scale, well below the sophistications of logico-grammatical thought. This means that the vast majority of minds on planet Earth are not cognitive minds—most animal minds; and adult human minds are also largely non-cognitive. So, RTM is not a theory of mind but a theory of one part of the human mind, and quite a limited part at that. Most of psychology is concerned with non-cognitive mental phenomena; in fact, for most of evolutionary history there were no cognitive minds, and hence no compositional mental representations. RTM is a rather specialized subject, not representative of the mind in general. Cognitive science, so conceived, is rightly so called; it concerns only a narrowly defined aspect of the mind. If you want a name for the broader field, you might have to speak of “mental science” or “cognitive and non-cognitive science”. It would be conceded, indeed insisted upon, that RTM is not the right framework for discussing most minds and large sectors of the human mind. In scope we might compare it to the study of sensations or emotions or mental images—one part of the total subject matter of psychology.

But what is the right theory of the non-cognitive mind? What is the essential nature of the non-cognitive? Is it just a motley or can we unify it theoretically? Here we are apt to fall silent—nothing springs to mind. The cognitive mind is unified under the heading “representational”, but what label captures the non-cognitive mind? Fill in the blank in “the…mind”. This is not an easy question and our standard vocabulary is not up to the task, but I have a proposal. We naturally think of the items in my list as effects of something—stimuli (internal and external), triggers, prompts, pressures, elicitors. The mind is affected thus and so—caused to assume a certain color or shape (here our vocabulary runs out of steam). A visual sensation is elicited, an emotion is prompted, a desire is evoked, a visual image is brought to mind. This is the mind as receptive not pro-active: what is imposed upon us not initiated by us. It is reflexive not free—stimulus-dependent. Thought can go anywhere, belief is flexible to a fault; but much of our mental life follows fixed patterns, invariant laws of production. Accordingly, we could speak of “the affective mind”, with the word “affective” stripped of its narrowly emotional meaning. Clearly, the word is chosen to describe emotions because emotions are generally the result of impingements (sometimes internal); I am generalizing it to a wider range of mental phenomena. If we want to maintain verbal parity with “the representational mind”, we could speak of “the affectational mind” and swallow the neologism; in fact, I recommend this term as the label we need to accord the non-cognitive mind its nomenclatural due. Abbreviating, we get “ATM” to be set beside “RTM”. Then we can inquire into the proper form of the ATM: what is distinctive of this part of the mind (or type of mind) by contrast with RTM? We know it is not propositionally compositional, but what is it positively? Here we quickly draw a blank: we can try saying consciousness, but that will include conscious thoughts too; or we could appeal to old standbys like privacy, intentionality, privileged access, etc. I think the best structural feature to fasten onto is that these aspects of mind have an analogue character: they come in degrees and correlate with other quantities such as stimulus strength. There is a psychophysics associated with them—proportional mental responses to varying physical stimuli. The stronger the stimulus the more intense the response: the brighter the light the brighter the visual sensation, the emptier the stomach the more intense the hunger, the scarier the danger the greater the fear, etc. Thoughts are not like that; they don’t vary in intensity in correlation with an external stimulus or situation. Their subjective intentionality is not a function of the parameters of the eliciting stimulus; it is different in kind from the intentionality of thought. Their compositionality (and atomicity) is not of the propositional-grammatical type; it is more primitive, being possessed by all animal minds. It isn’t advanced intentionality. Their content is more phenomenal than propositional, visceral not cerebral. The affective mind is not about words or word-like mental representations.

Now that we have mapped out the geography, we can begin to investigate the geology. Which came first, which is more basic, how does the layering look? Clearly, the affective mind preceded the cognitive mind; logically structured thought appeared late in evolutionary history (and is still in its early stages). Sensation and emotion are more ancient. Somehow thought evolved from these beginnings, abetted by language in advanced species (we don’t know how). In the mental strata diagram, the non-cognitive mind is a deeper layer, forming the foundation of the cognitive mind. In it are fossils of earlier phases of life on Earth (e.g., temperature receptors, predator fears, etc.). Language is nowhere in sight and belongs in the upper layers. What part of the mind do we understand the best? Some might say we understand the cognitive part better than the non-cognitive part, because RTM is a well-developed explanatory theory. But this is to ignore all the mysteries entrained by that theory, and it assumes that RTM is a good theory within its domain.[2] In the bad old days of Pavlovian behaviorism people thought that the non-cognitive part of the mind was pretty much all tied up—it’s all just conditioned and unconditioned reflexes. Now we know better, what with the emergence of consciousness from the shadows and a greater respect for the inner. The truth appears to be that there are now scratches on the surface of both minds, but we still have a long road ahead of us (which might end in a town called Nowhere). At least we have a clearer idea of what we don’t understand. I would say this: it would be strange if we had a good theory of the cognitive mind but a poor theory of the non-cognitive mind; for the one depends on the other and interacts with it. More plausible is the suspicion that RTM is not as good as it sets out to be—less penetrating, more cobbled together. Isn’t it more likely that both aspects of mind are equally cloaked in mystery? My suspicion is that RTM carries a kind of spurious appeal, because it borrows from the study of language: we have a pretty decent idea of the structure of language (syntax) and some ideas about its pragmatics, so we apply this to thought in hopes of instant illumination (any light is welcome on a dark night). But actually, thought is quite remote from language and precedes it handily. RTM is really based on a loose analogy between thought and language not on any direct insights into thought as such (we can’t see it and hear it). If so, the two domains are on a par intelligibility-wise. The things we call “concepts” remain as elusive as ever (notice that Chomsky never bought into Fodor’s theory of cognition and had little to say about thought itself). A representational theory of language seems warranted (but see below), so we naturally transfer it to what language expresses, viz. thought. But this sleight of hand should not go unremarked; it is too clever by half. The linguistic model is all too tempting—a quick and easy fix for a big problem. I see the appeal, but I don’t commend the capitulation. The linguistic turn in cognitive psychology (LOT and its gang) is liable to be a turn into the wilderness, or just a trip around the parking lot. In any case, it is not clear that RTM is a step ahead of whatever we have to say about the affective mind (not a great deal, alas). What are these strange things—concepts, thoughts, beliefs? What is their inner nature?

I want to sketch a possible view of the thought-language interface that strikes me as having some plausibility (indeed to be obviously correct on reflection). It is simple: an ordinary natural language is a formal system with a determinate syntax; we are born with it coded into our genes (some mutation gave rise to it a couple of hundred thousand years ago). But it has no intrinsic reference; words in it (the lexicon) don’t have reference intrinsically—this is conferred on it by speakers in virtue of their thoughts. Thoughts do have reference intrinsically, but words don’t. Thoughts also have compositional structure in a way analogous to language. Thus, we can’t explain thought-reference in terms of word-reference; that gets things the wrong way round. It is an illusion to suppose that thought refers via language (internal or external); that would be circular. We are mistaking an effect (language) for a cause (thought). Thoughts are indeed about things, but not in virtue of being linguistic; language refers only because the thoughts of speakers do. If this is right, then RTM is a kind of pseudo-theory—not false exactly but misconceived. This resolves the puzzle of how cognitive psychology seems to be in better shape than non-cognitive psychology; it really isn’t, but it can give that appearance. Actually, the representational theory of language (RTL) is wrong, because language is not inherently referential; only thoughts have reference built into them. Language refers parasitically.

I return to the question of the non-cognitive mind. Some of the mind is propositional and some of it is not. We need a theory of both. We can’t give a cognitive theory of the non-cognitive part, and anyway it is doubtful that we have a theory of the cognitive part (only a kind of theory-imposter). The non-cognitive part seems no easier to understand than the cognitive part, though no harder either. We should not insist on a sharp bipartite division, as if you could do one and not do the other. The mind is more unified than that picture allows. “Cognitive science” is a bad label.[3]

[1] See the many works of J.A. Fodor, its chief architect and promoter (as well as comedian).

[2] See my “Fodor’s RTM”.

[3] It is an interesting question why this label caught on: what is “cognitive” being contrasted with? Overtly, with “behavioral”—we are (at last) doing internal psychology (“intervening variables”). But is there also a hint of snobbery vis-à-vis the non-cognitive parts of mind? Are we looking down on the other parts of the mind, especially the emotional? Is Freud lurking in the background somewhere with his lurid sexual psychology (“erotic science”)? We just don’t hear much about the psychology of the passions these days—it’s all computations and ratiocinations. I would like to see a serious emotional science alongside the cognitive kind. Fear psychology would be a good place to start. What is the role of mental images in the generation of fear? How do phobias develop? What is the nature of emotional intentionality? We need a Fodor of the feelings, a Chomsky of the sentiments.

Do Bats Know What it is Like to be Human?

One answer would be that we don’t know whether they do or can know what it’s like to be human, because of the problem of bat other minds. That is a boring answer and, I think, incorrect; in any case, I will ignore it. We know enough about bat minds to know whether or not they know what it’s like to be human. And I believe the simple answer is that they do know. This is because we don’t have any senses they don’t have. They have one sense we don’t have (echolocation), but we don’t have a sense they don’t have. Their senses may not be as acute as ours (e.g., sight), but they have the basic phenomenology associated with our five senses; they are not totally in the dark about what it’s like to be human from a sensory point of view. Someone might say bats can’t have such knowledge, because their cognitive abilities are too limited to know anything of the kind. They don’t even know what it’s like to be a mouse. Again, this answer is boring and, I think, incorrect: they know what’s it’s like to be them, and other animals share that subjectivity to a sufficient degree to ground knowledge of other animal minds. Bats know what it is like to see and hear (etc.) generally. Basically, all mammals know what it is like to be a mammal, so long as their sense modalities overlap. They know each other’s mode of sensory consciousness. Indeed, the same point applies to reptiles and birds, fish and octopuses. So, perhaps surprisingly, animals have good knowledge of what it’s like to be most other animals, including humans. Even mice know what it’s like to be human!

But that point only applies to sensory consciousness, and it is not the only kind. What about what may be called a priori consciousness—logical, mathematical, ethical, philosophical? Assuming that bats don’t have that kind of consciousness, do they know what it’s like to be human? I think not. In fact, I am doubtful that any animal knows what human a priori consciousness is like, even our closest relatives. Of course, if they have a kind of closeted a priori mental life, then they can know what it’s like to be us; but there is no evidence for this and it seems clearly false. So, there is a large part of what it’s like to be human that bats and other animals don’t and can’t know.  Nor can humans who lack rational faculties, say because of brain abnormality. Human intelligence (of an advanced kind) is not knowable by animals in general; they don’t know what it’s like to have the kind of intelligence we have. We can know what it’s like to have bat intelligence, but they can’t know what it’s like to have human intelligence—though they do know what it’s like to have human sensations. If a bat were to write a paper called “What is it Like to be Human?”, the answer would have to be “We don’t know”, or “We know some but not all”. The topic of animal knowledge of other minds is a neglected area, but part of psychological zoology. I am just laying the groundwork here.[1]

[1] I think we know exactly what it’s like to be a dinosaur, sensorily and intellectually, and I think they know what it’s like to be a human sensorily—but not intellectually. We humans are distinctive in having a developed a priori form of consciousness. It is like having another sense.

Asleep and Awake

Listen, do you want to know a secret? The government has been withholding it from us. When we are asleep, we are awake, and when we are awake, we are asleep. We have been bamboozled into thinking that in sleep we are not awake and when awake not asleep. Sheer propaganda and lazy ordinary language (ordinary language propaganda). The best way to describe our mental state when in the condition we call “sleep” is that we are partially asleep and partially awake (very awake actually); and similarly, for being awake (we are partially asleep). This is not hard to demonstrate and requires no sleep science or fanciful hypotheses; it is common knowledge. We are always both asleep and awake. The OED defines “sleep” thus: “a regularly occurring condition of body and mind in which the nervous system is inactive, the eyes closed, the postural muscles relaxed, and consciousness practically suspended”. This definition leaves a lot to be desired: people don’t always have the luxury of sleeping regularly, the nervous system is not inactive in sleep, the eyes are not invariably closed (you might not even have eyelids), the postural muscles are no more relaxed than they are when lying down awake, and what does it mean to say that consciousness is “practically suspended”? Clearly, the editors are having trouble defining the word (and the state itself). The basic idea is that the sleeping person (or animal) is (largely) unaware of his surroundings: he is not awake to his immediate environment. He is not tuned into what is going on around him, not conscious of it, oblivious to it. His mind is blank relative to his environment; his senses are cut from their usual perceptual function. He is selectively “blind”.

But let’s notice two things about this state of oblivion: it is not complete, and the sleeper is normally conscious of other things. Outside stimuli are still getting in to some degree, as experiments have shown; and, of course, he is often dreaming. He is quite well aware of his dreams—they are occurring in his sleeping consciousness. In dreaming we are aware of our dream objects—the intentional objects of our dreams (people and things). We are awake to these. We are not detached from our dreams; we know about them perfectly well, as our subsequent memory shows. We are not awake (much) to our environment, but we are awake to our dream world. We are not perceptually awake, but we are introspectively awake. Our perceptual consciousness is blank, but our dream consciousness is brimming with content—just as our day-dreaming consciousness is. We are not asleep relative to this, i.e., unresponsive to it, inactive with respect to it. Psychologically, we are very much awake: aware, conscious, affected, emotionally moved. We are not in an unconscious coma. Our nervous system is firing on all cylinders. Thus, we are partially awake during sleep (unless sleeping dreamlessly).

And there is a further point: we are also aware of the condition of our body—our posture, temperature, the state of our bladder, etc. We are awake where they are concerned. To be asleep is really best understood as being unaware perceptually (though even this is not total). Our senses are asleep, but not the rest of us. But we lazily describe this as if the whole conscious mind has shut down, which is far from the truth. Our normal concept of sleep disguises the psychological facts from us. Ordinary language misleads us. It was never designed to be pedantically correct psychologically. It isn’t meant to be scientific psychology. We can imagine speakers who say things like, “I was wide awake all night, dreaming of my mother-in-law”.  Might it not also turn out that our so-called dreams are really reality, so we never had them while asleep? It is epistemically possible that you are really awake when you think you are asleep dreaming. Qualitatively, dreams are like waking experience. The OED gives the phrase “awake to” the definition “aware of”, and we are certainly aware of our dreams (not to mention the state of our bladder). In short, you are awake when you are asleep—awake to.

The next claim is less obvious: that we are asleep when awake. But the reasoning is much the same: we are unaware of certain things that we are aware of during sleep. Here it is convenient to invoke Freud, at least as a thought experiment: in Freud’s psychology we are closed off in waking life to our unconscious, but we are aware of it in dreams—for example, you might dream of having sex with your mother. Suppose there exists some sorry species that actually has a mind as Freud described our mind: they dream at night of acting on their repressed sexual desires, but during waking hours no such content enters their heads. Then they are not aware of (awake to) the facts of their own mind, as revealed in their dreams; they are asleep relative to those facts. They are unconscious of their inner (psychological) environment when awake. They are awake to it at night, but not awake to it during the day—it’s as if they are asleep where that is concerned. They are in a detached sleep-like state when it comes to their unconscious desires. They have no consciousness of these while awake. They know them by night, but not in the light of day. Sometimes people are described as sleep-walking through life, meaning that they are oblivious to obvious facts, closed off to reality. It’s as they are asleep cognitively. The question then becomes whether we partake of any of this: are we at least somewhat Freudian? I think this is not an unreasonable conjecture: our dreams are tuned into facts about ourselves that never reach the surface in our waking life, except perhaps in a crisis situation. We gleefully go through life thinking we are decent intelligent human beings, say, while all along we are stupid bastards. We are asleep relative to the truth about ourselves, while awake relative to other things—unaware, incognizant. We are in a state like that of not seeing or hearing what is going on around us when asleep. The same natural psychological kind occurs in both cases.

We need a word for what we do typically at night when we close our eyes and don’t want to be disturbed, and we call that “sleep”. We also need a word that signifies that we are up and about and ready to face the day, and we call this being “awake”. We then think these are mutually exclusive, so it sounds funny to speak of sleeping wakefulness and waking sleep. But the psychological facts underlying these phases of human (and animal) existence are more complex than the simple binary distinction recognizes; the two states are more intertwined than we suppose. The fact is that we are in both states nearly all the time—blind to this but not to that, awake to this but not to that. A sleeping man might be extremely awake to certain things, and a waking man might be fast asleep with respect to his own self. Your dreaming life may be more awake than your waking life, and your waking life may be sleepier than your dreaming life, strange as it sounds. So much for the ordinary language of sleep and wakefulness.[1]

[1] Is it a matter of conversational implicature? If I say of a sleeping man “He is awake”, do I speak the literal truth, though the implicature that he is in perceptual touch with his surroundings is clearly false? If so, that would be a striking example of Grice’s invaluable distinction. Try it with “He is conscious” said of a sleeping (but dreaming) man; surely that statement is quite true, given that dreaming is a state of consciousness—implicatures notwithstanding. Is sleep, in reality, one kind of wakefulness and wakefulness one kind of sleep, though we don’t talk that way for implicature reasons? The human mind is asleep-awake all the time.