Analyzing Knowledge

We are familiar with Gettier problems, which bring out the insufficiency of the classic analysis of knowledge as true justified belief. But there are other problems with that analysis, centering on circularity. They question the pretensions of such an analysis to provide conditions that don’t presuppose the concept of knowledge but together add up to it. The classic analysis is at best misleading about the constitutive structure of the concept of knowledge, its inner constituents. The first problem concerns the concept of belief, usually introduced without much fanfare and little reflection. Thus, we are baldly told that if x knows that p, then x believes that p—while the latter does not entail the former. But what is it to believe that p? Isn’t it to take oneself to know that p? If I believe that the Battle of Hastings took place in 1066, I take myself to know that—and in this case I am right so to take myself. In other words, I believe I know it—or else I wouldn’t believe it. Thus, the concept of knowledge enters the concept of belief. If we were interested in defining belief, we might well offer: x believes that p if and only if x believes that he knows that p. True, we are using belief as part of its own definition, but the definition is informative and not viciously circular. We might also try to eliminate the concept of belief from it and replace it with something less repetitious (taking oneself to believe, or a disposition to assent, or brain-based functional role). In any case, the definition seems perfectly correct, and it invokes the concept of knowledge—the very concept we are supposed to be analyzing. What else is belief but an attempt at knowledge? How could you have the concept of belief and not realize that belief is linked to knowledge? Belief is would-be knowledge. Belief doesn’t entail actual knowledge, to be sure, but it alludes to that concept—it entails knowledge aspirations. The classic analysis only looks non-circular because we don’t ask ourselves how belief is to be understood, i.e., what it is. It is hardly a satisfactory entry point in the quest to specify what adds up to knowledge but doesn’t presuppose it.

The same question recurs with respect to the next two conditions—truth and justification. What is justification? Isn’t it precisely information that leads to knowledge, or is apt to lead to knowledge in favorable conditions? You can’t have the concept of justification and not see that justification leads to knowledge; no one could possess that concept and have never heard of knowledge. At any rate, we need an argument to show that such a thing is possible, or else circularity looms. Thirdly, the concept of truth is also inextricably linked to the concept of knowledge: truth is what the search for knowledge is the search for. How could you have a grasp the concept of truth and not know that truth is linked to knowledge in this way? The circle of concepts is too tight for that. You couldn’t learn what knowledge is by consulting the classic definition, because you would already need to have the concept in order to understand the definition. Truth and justification don’t individually entail knowledge, to be sure, but they incorporate the concept at one remove. The classic theory is not wrong exactly, but it fails to provide the kind of illumination advertised on its behalf. The concept of knowledge doesn’t pop out of the three conditions like the proverbial rabbit from a hat, because it already lurks close to the surface of the concepts used to define it, particularly belief. In the matter of conceptual priority, the concept of knowledge seems to come first—as primitive not constructed (as has often been contended).

These reflections encourage a reformulation of the classic analysis. This reformulation doesn’t dispose of the circularity problems, but it does simplify the intent of the classical analysis. It brings out what is really going on when someone knows something propositionally. Let’s say that x knows that p if and only if (1) x believes that he has a true justified belief that p (the belief condition), and (2) this belief is true (the truth condition). For he knows what knowledge is (true justified belief) and it is true that he is in the state in question. Intuitively, he believes he is in a state of knowledge and he actually is. The classic analysis analyzes his belief according to its own theory, incorporating this into the belief condition, and then it says that that belief is true. If it is correct in its analysis, then x will (perhaps tacitly) know what knowledge is, and then it only remains to specify that the conditions in question are actually satisfied. Knowledge emerges as a combination of belief and truth, just as we might have expected; but the belief turns out to have more structure than we might have supposed. We upload the analysis into the belief component, so to speak. This is really what is going on when a person knows something: he has a belief about his epistemic status and that belief is true. He doesn’t just have a belief in the proposition in question (e.g., that the Battle of Hastings took place in 1066); he also has a belief about this belief, to the effect that it is true and justified. This gives us a more accurate and illuminating picture of what knowledge entails—at any rate, of the kind of knowledge in question (rational, reflective). The person doesn’t just believe the first-order proposition, but also has other beliefs of an epistemological nature, concerning his entitlement to believe. The structure of the situation is made more apparent in this reformulation, though the substance is much the same (knowledge as true justified belief). The knower doesn’t merely have the belief in the first-order proposition; he also has various attitudes about his state of belief. Animals, presumably, don’t have such attitudes, so according to this analysis they don’t have knowledge (in the way we do at least). To have knowledge in the way a typical adult human does, you need to have the attitudes in question—which is the normal condition of human knowers. Knowledge is thus more richly structured cognitively under this formulation than under the classic formulation.[1]

[1] Here is a somewhat paradoxical result if Gettier is right and no one has an adequate analysis of knowledge: it will not be sufficient for knowledge that the subject’s belief about his epistemic state is true, since the true justified belief analysis is not sufficient. Accordingly, no one ever knows anything, or anything for which a Gettier case can be produced. In order to know, you need the correct analysis of knowledge; but according to Gettier no one knows the correct analysis, so no one knows anything. The strong belief condition rules out the possibility of knowledge, apparently. You can’t know without knowing what knowing is, but we don’t know that. Yet the reformulated classic analysis seems very plausible. One way out would be to say that we do know the correct Gettier-proof analysis, namely that knowledge is non-accidentally true justified belief—or whatever your favorite answer to Gettier is. This is an interesting wrinkle on the problem.

How to Take Your Cat to the Vet

I wish to impart some practical advice concerning the cat-vet problem. It comes from bitter experience that I have reason to believe is common. It is designed to spare you and your cat from distress and injury. How do you get the cat in the carrier? First, place the carrier in your bathroom (or other small sealed room if you have one) the day before you plan to visit the vet. Make sure you do it with the entrance part open at the top, so not resting horizontally on its bottom. I suggest covering it with a sheet so as not to make the cat suspicious. Wearing a pair of stout gloves, carry the cat into the bathroom at the appropriate time and shut the door; it shouldn’t be alarmed at this point. Now take the sheet and throw it over the cat. It will certainly be alarmed at this point and a struggle may ensue as you work to wrap the cat in the sheet. Keep a firm hand and ensure that the head and legs are enclosed in the sheet (the cat will probably be howling at this point). Now gently but firmly drop the enclosed cat hindlegs first into the box using the force of gravity. Make sure it doesn’t get out of the sheet on the way in (pay special attention to the legs). The hard part is now over. Close the door to the box and proceed to the vet. The cat will be moaning the whole way but will probably be free of the sheet in a few minutes or less. No harm will be done. The whole point of this method is to avoid the problem of the cat running away from you as you try to nab it—they are very quick and cunning. Also, it solves the problem of the cat fighting to stay out of the box, possibly scratching you and hurting itself, because the sheet immobilizes it. Notice that the whole operation can be done by one person. It is generally a bad idea to try to force the cat into the box head first and horizontally; dropping it in backwards and vertically is much better. Good luck![1]

[1] I described this method to the vet the other day and received his approval—apparently, it is a common problem.

Moral Seeming

In moral philosophy we find distinctions between moral language, moral psychology, and moral reality. Interrelations between these areas are explored. Moral psychology tends to dwell on questions about moral motivation—does it consist in moral beliefs, moral desires, or moral sentiments? I wish to add a further topic: what I call moral seeming.[1] There are delicate questions of terminology here: how should we describe this moral seeming? We don’t normally speak of moral seemings (plural), so what noun would better serve our purposes? We might speak of moral impressions or appearances or presentations or perceptions or even sense-data, but for a variety of reasons I prefer sensations; so, I shall employ this (technical) term with the proviso that I mean a special class of psychological events in which something seems a certain way to the subject—as in “It seems to me that X is wrong”. Intuitively, the subject has the subjective feeling that X is wrong—even if he might go on to add, “But I don’t believe X is really wrong”. This is what people are getting at who use the phrase “gut feeling”: it is a kind of pre-reflective automatic response to a moral situation. My suggestion is that such sensations are ubiquitous in moral consciousness: they exist alongside moral judgments or beliefs; they belong to a more primitive level of moral awareness (perhaps young children have only such sensations). These sensations have an epistemic role in moral reasoning, providing reasons to form moral judgments, analogous to sense experience; thus, there is such a thing as moral epistemic phenomenology—moral sensations. The doctrine that these exist could be called “moral sensationalism”. It might be objected that such things don’t exist, because you would never say “It seems to me that murder is wrong”, since that would imply that you aren’t really sure that murder is wrong. But this objection conflates logical implication and conversational implicature. The case is just like saying “It seems to me there is a red bird over there”. I won’t go into this, as the distinction is well known and indisputable. I will take it that the existence of moral seeming (moral sensations) is hard to deny; in any case, I will accept it in what follows. My question concerns the role of such sensations in moral mentation—where they fit in, what they signify, how they relate to moral reality.

Let’s compare moral seeming to modal and grammatical seeming. In each of these areas it is common to speak of “intuitions”—we intuit the truth. For “intuition” the OED gives “the ability to understand something immediately, without the need for conscious reasoning”. I have no objection to intuitional psychology, but I think the term is slightly misleading in the present context, because it suggests something more intellectualist than the facts require. The word “sensation” captures these facts better—we need to bring out the primitiveness more clearly. Thus, we say that something seems necessary or that a string of words seems grammatical—this is how they strike us, present themselves to the mind. We don’t have a mere “intuition”, as if we are guessing or speculating. In any case, moral seeming is like modal and grammatical seeming—what we might call “first impressions”, correctable in the light of further thought. They are fallible and corrigible. You might have an impression that eating meat is morally permissible but later come to the conclusion that it is not, or that homosexuality is wrong but later come to realize that it is not. Moral consciousness operates on two levels, corresponding to immediate impressions and considered judgments, which do not always march together. Can you see how it mirrors ordinary visual perception and associated beliefs? Once moral seeming is recognized, the similarity to empirical thought becomes evident (though there are clear differences); in particular, there are such things as normative sensations, impressions of value. It starts to seem appropriate to speak of the “moral sense” and of “moral sensibility”.

How far can the analogy be pushed? Can there be moral illusions as there are perceptual illusions? I think there can be: you might be under the moral illusion that homosexual sex is morally wrong, or that masturbation is wrong, or sex before marriage, or atheism, or eating certain foods on certain days. These things seemwrong to you, but in reality, they are not—and they might continue to seem wrong even when you change your moral position on them. There might be moral equivalents to the Muller-Lyer illusion; moral sensation might be “encapsulated”, in Fodor’s term, unalterable by the “central system”.[2] Could there be “moral blindsight”? A person can make true moral judgments but has no moral sensations to go along with them—he just leaps to the judgment and skips the seeming stage. He might be a sensational moral blank but a competent moral judge. He is a sort of Kantian freak—no primitive moral awareness but competent moral rationality. Conversely, someone might be stuck at the childlike sensational level, full of moral impressions but lacking considered moral judgment (aren’t a lot of people like this?). Generally, moral sensation can be shaped and educated, to some degree, like perceptual sensation; but it would be wrong to assimilate moral sensation to moral thought—the distinction always exists. The relation between reason and sensation in the moral case is as complex as it is in the ordinary perceptual case, and strikingly similar.

What is the intentional content of a moral sensation—how does it represent moral reality? In particular, is it absolutist or relativist? For example, does it represent the wrongness of needlessly causing pain as a cultural universal (a moral absolute) or as culturally relative (merely local)? That is, does it seem to you that causing pain is always wrong no matter the state of cultural belief, or does it seem to you that it can be sometimes wrong and sometimes right (or neutral) depending on the culture? This is not a question I have ever seen asked, presumably because the idea of moral seeming has not cropped up before. The answer to it, I believe, is that moral seeming is absolute—even if considered moral judgment is relativist. Accordingly, relativism can be revisionary of ordinary moral seeming: morality seems absolute but in fact it is relative. Moral sensations make an error about the nature of moral reality. Moral phenomenology suggests objectivity and absoluteness, but critical moral philosophy urges moral relativism. Our given moral sensations lead us into error. But in my opinion, this lack of convergence converts into an argument against relativism, because of the difficulty of explaining the alleged error. Without going into this in detail (it is familiar ground), I think that no plausible explanation can be given for why moral impressions are absolute except that moral reality is absolute. The reason moral sensations have the content they have is that moral reality has the nature it has. The imputation of error on that scale is just not credible. Moral impressions are absolutist because moral facts are—content follows truth. It is the same with non-moral impressions, and moral impressions simply conform to a general pattern. If they had a relativist content, we would never be prone to moral absolutism; but we are so prone (rightly in my view), so the odds are that our sensations suggest it. But if they do, then the best explanation of this fact is that moral absolutism is true. Even the convinced moral relativist must accept that we are all absolutist at the most basic level of moral awareness—we sense right and wrong as universal and fixed not as local and variable. It does not seem to people that their moral convictions are merely relative and optional; they have a strong impression of universality, rightly or wrongly. This is so regardless of their reflective philosophical views about morality.

Here is a difficult question: in cases where people have the wrong moral views, are their moral sensations different from their opinions? For example, do slave-holders tacitly sense that slavery is wrong, or meat-eaters tacitly sense that meat-eating is wrong, or capital punishment, or child labor? I think it is possible that they do: they sense it but they disregard the evidence of their senses (their moral sense in this case). For it is hard to see how they could fail to sense it, given the obviousness of the wrong. Granted, there is always self-deception, stupidity, willfulness; but at some level, aren’t people aware that what they are doing is pretty nasty, however much they try to excuse it? Didn’t children in the age of slavery have the distinct feeling that slavey was wrong, especially in its more violent aspects? They have to be indoctrinated out of these natural feelings. If this is true, then moral progress might not be as dramatic as we suppose—perhaps it largely consists in falling back on our natural moral reactions. Or am I being too kind—were people (even children) the absolute moral swine they appear to be back in the bad old days? Did they experience not even a trace of moral concern for those tortured and murdered in front of them? Was there no inner voice whispering, “This is wrong”? I like to think that their primitive moral sense was not as depraved as their actions (and beliefs). The same is true for contemporary atrocities.

Do some things seem worse than other things to our moral sensibility? I rather think so, just as our beliefs distinguish degrees of moral badness. It seems worse to murder than to steal, and stealing seems worse than promise breaking. If so, moral sensations vary in intensity—just like visual and auditory sensations. There could therefore be a kind of psychophysics of moral sensation: the greater the wrong, the stronger the sensation. Visual sensations can vary in brightness, and moral sensations can vary in intensity of condemnation. This seems phenomenologically correct: we can have mild or strong moral reactions to perceived wrongs (“That seems really, terribly, wrong to me!”). Some moral sensations may make as cry out and stamp our feet, while others elicit only a slight shake of the head. Here moral seeming touches moral feeling—the sensation is emotionally imbued. Emotions can vary in intensity too, and they combine with moral impressions to produce a scale of behavioral response. There might even be the analogue of quantitative psychophysical laws in the moral domain: we could scale moral sensations as a function of moral seriousness—the more atrocious the crime, the more intense the subjective response. There could be a new field: “psycho-ethics”—the study of the mathematical relations between ethical reality and ethical psychology. Its first law: moral sensations vary logarithmically with moral wrongness.

It isn’t surprising, from an evolutionary point of view, to find that the human moral faculty has the two-tiered structure I am suggesting. As a social species, becoming sensitive to moral distinctions, we would need a fast-response reflexive system to guide our behavior towards others. This is the system of moral seeming—moral impressions based on initial appearances and employing general rules (“He steal—bad!”). Superimposed on this, we have the slow, holistic system commonly known as moral reason, which takes immediate moral impressions as input and delivers considered moral verdicts as outputs (“We find you guilty of the crime of stealing and sentence you to twenty lashes”). This is the general form of all our thought, all our knowledge, and moral thought is just a special case. I have called the lower-level system “sensation” with some misgivings, given the connotations of that term, intending to point to the psychological role played by the designated mental elements. The idea is to indicate the immediate quasi-sensory nature of the relevant psychological processes—the way they give rise to episodes of seeming. This needs to be added to our overall moral psychology. There is good seeming and bad seeming, right and wrong seeming, as well as associated beliefs, desires, emotions, and acts of will. The moral faculty has many departments and interactions between them. It is quite a bit richer than has commonly been supposed. No doubt it will continue to evolve.[3]

[1] For background on seeming, see my “Seeming” and “A Philosophy of Seeming” on this blog.

[2] See Fodor’s The Modularity of Mind (1983).

[3] It is odd that philosophers (and psychologists) have tended to have oversimplified views of the nature of moral psychology—from simple sentimentalists to simple rationalists. Perhaps this results from a tendency to regard moral psychology as a minor department of the human mind, one that runs on simple principles. Emotivism is the extreme case: moral psychology is nothing but surges of emotion without structure or liaisons. And this in turn might result from an inability to recognize how complex moral reality itself is. We need to recognize that morality is as complex and multifaceted as reality in general—plural, mind-taxing, elusive, inscrutable. It’s hard for the human mind to get itself around morality; it needs all the help it can get.

Language As Thought

I am going to present a new theory of language—a new philosophy of language. Or perhaps I should say newish because elements of it are already out there, particularly in Chomsky.[1] The question is what language is—what fundamentally constitutes it, what its primary mode of being consists in. And the idea is that language is a type of thought: it is identical to thought of a certain type—the linguistic type (not the imagistic type). So, this is to be an identity theory: language is identical to thought, reducible to it, nothing more than thought. In particular, it is not speech: speech is not language at all but its expression or externalization. Language is psychological not acoustic; it is a kind of category mistake to identify language with speech, like identifying pain with pain behavior. The theory is mentalistic not behavioristic—internalist not externalist. Language is essentially inner, though it can have outer manifestations. It’s entirely in the head never in the body (not in the mouth or ears). Ultimately, it is in the brain: it is a cognitive structure that is realized in the brain. Linguistic competence is a mental system located in the brain; linguistic performance (speech) is located in the body. Language is soundless and imperceptible; speech is noisy and perceived (the ears are involved). Internal cognition is not sensorimotor activity. The language faculty is a faculty of thought not of speech; it is an ability to think in a certain way not to make sounds in a certain way (or gestures). If you want empirical evidence for this theory, you need look no further than ordinary experience: language is always going through our minds even when speaking isn’t at issue.[2] We use language inwardly all the time but we speak intermittently. In principle, we could use language inwardly and not speak at all: the language faculty can exist in the absence of the speech faculty. In any given day, you may be a hive of linguistic activity but not make a sound; this activity is a type of thinking, neither more nor less. Speech acts are irrelevant to language, mere offshoots of it; they are not of the essence. There cannot be language without thought, but there can be language without speech (by sound or sign).

Four views of the nature of language may be distinguished: language as a means of communication, language as speech, language as a medium or vehicle or instrument of thought, and language as thought itself (or one type of thought). The first two views are distinct because there could be speech without communication, as in soliloquy; and the second two are distinct because only the last proposes an identity theory, as opposed to a correlation theory. Language could be an instrument of thought and yet not be reducible to thought—it could be a distinct existence, yet correlated. It could indeed be maintained that speech provides the instrument of thought—as I think is commonly assumed. X can be an instrument of Y without X being Y. According to the view I am describing, however, language is no more a medium of thought than the brain is the medium of the mind; the brain is the mind. It doesn’t stand apart from the mind but constitutes the mind (this is not to be a “physicalist”). When we think in images they constitute the thought—they are not just its vehicle or carrier or clothing or correlate. When we think in language the language forms the essence of the thought—it isn’t just thought’s accomplice or helper. So, the mentalism I am defending is stronger than merely the claim that the essence of language is to be found in its thought-assisting role (as opposed to its communicative role); it is the claim that language is nothing other than a species of thought. We have an innately given language faculty and its primary output is thoughts of a certain type—not speech behavior or acts of communication or even thought vehicles (tools, crutches, bearers).

What type? What distinguishes linguistic thoughts from other types of thought (assuming there are other types)? We might not have the full answer in the current state of knowledge, but we have enough to give the idea some substance: recursiveness, hierarchy, finite basis, infinite potential, digital discreteness, logical form, propositionality, meaning. Such thoughts have just the properties commonly attributed to language—because they are what language really is. Speech doesn’t have these properties, not intrinsically anyway, but thought does, in virtue of the innate language faculty. To think linguistically is to think in accordance with these features. Language is defined by these features, and thought has them in abundance. It might be said that thought isn’t the only thing to have these features—don’t acts of communication have them too? And isn’t there such a thing as body language and whale language and the language of the genes and computer language? All this is irrelevant, however, just so much metaphor and extended usage. The word “language” as it exists in ordinary parlance has many meanings; I am concerned with the correct scientific account of a specific ability we observe in human nature. I am concerned with what the primary reality of language consists in—what language intrinsically is not what might be described derivatively as language (e.g., speech behavior). It aids clarity to insist on a degree of terminological purism: only inner thought is language in its primary meaning—original language not derivative language (compare original and derivative intentionality). We could write “Language” to indicate the domain in question and allow other uses of “language”, but I think in scientific and philosophical writing we do better to insist on strictness—it’s good for the soul to exercise some terminological discipline. Thus, there is no language but thought language–literally, scientifically. We might hear the word as meaning “mind-language” or “internal-language” or “brain-language”—so as to exclude such irrelevancies as “body language” or “the language of love” or “computer language”. All language in the restricted scientific sense is a language of thought—with the proviso that this language isn’t just of thought but thought’s inner structure. It is thought-language, brain-language. When Descartes said, “I think, therefore I am” he could have said, “I cogitate linguistically, therefore I am”. Man is a thinking being, i.e., a being with linguistic modes of thought (defined as above). The primary (biological) function of language is to form a specific type of thought, presumably because this type of thought has adaptive advantages. Speech came later and might not have come at all. It might be called the externalization of language, but it isn’t the genuine article, the heart of the matter. Speech is a by-product of language proper not its pith.

The study of language is therefore the study of this internal system. Linguistics and philosophy of language are primarily concerned with this region of reality. They might study other things too (speech, communication), but it invites confusion to say that these are studies of language. We need to clean terminological house and keep distinct things distinct, not use the word “language” loosely and ambiguously. This means that a lot of philosophy of language (and maybe some linguistics) is confused: it is really about communication and speech not language proper (i.e., what goes through our heads all day when we are not actually speaking). The work of Austin, Searle, Grice, Quine, Davidson, Dummett, later Wittgenstein, and many others is not about languageat all, and indeed officially repudiates what I am calling language. So-called linguistic behavior isn’t language at all, just the external sensorimotor expression of language (an effect of language)—speech acts, illocutionary force, language games, speaker-meaning, verbal assent, assertion, command, communication, etc. None of this is about language at its core, though it may be perfectly worthwhile in itself. It is notable that earlier work in philosophy of language was about language in the strict scientific sense I am advocating—work by Frege, Russell, and early Wittgenstein. For these philosophers were not concerned with language as a social institution but with language construed as a type of mental representation confined to the individual (and later maligned for this stance). They viewed language simply as a form of thinking: hence the theories of sense and reference, function and object, acquaintance and description, sense-data and inference, objects and pictures, saying and showing. None of this has anything to do with speech or communication; it is purely inner—language as a mental phenomenon. It is about symbolic thought as such, which is what language fundamentally is. Language is an abstract structure, syntactic and semantic, formally defined, existing in the mind. Speech maps onto this structure in various ways, but it isn’t thus structured intrinsically: it’s just noise, atmospheric perturbations, flexing of the larynx and tongue. I would say it is a category mistake to suppose that such physical phenomena can be accurately described as forming sentences or phrases or words, still less as having meaning, reference, logical form, and entailments; that is sheer projection. The self-same noises could exist and have none of these properties. Language is not a physical thing (except in so far as the brain is physical).[3]

Why don’t we see this more clearly? Surely, it’s because language as a mental faculty is hidden to normal perception, even though it is perfectly evident to introspection. Speech, however, is perceptible, whether it takes the form of sounds, marks, or gestures. We thus gravitate towards behaviorist conceptions, attributing to behavior what properly belongs to the mind. It is really quite obvious that there is a sharp distinction between animal minds and animal communication systems—the latter greatly under-express the former. The two capacities evolved for different reasons. The animal has a cognitive life as well as a behavioral life; its thought processes, such as they may be, are not constrained by its ability to express itself in external acts of communication. To adopt a communication-based view of animal cognition would be absurd (though not unheard of). We are essentially the same: our speech is a poor guide to our thought. Speech serves one sort of purpose, thought serves another. It is a mistake to think that language is all about speech and communication; its original function lies in what it provides in the way thought. We evolved a new way of thinking, augmenting imagination-based thought, and then subsequently deployed it to produce external speech—not vice versa. That is the simple truth. It just so happens that the basic facts of language are hidden from observation, silent and invisible. This prompts us to assign its properties to external facts, and then to project speech into the mind. The phrase “inner speech” is a misnomer: we don’t internalize what is essentially external; we externalize what is essentially internal. We commit a kind of “object-expression” fallacy, assigning to the mere expression of internal language the inner nature of the object expressed—treating the expression as if it were the object. Yet at the same time we recognize—what is quite obvious—that language has an inner psychological reality, because we experience it every day in our conscious mental life. Language courses through the conscious (and unconscious) mind. It may be that as children we acquire the ability to think linguistically well before we learn to vocalize linguistically; the cognitive language faculty develops before the speech faculty. Certainly, the child does not first learn to speak and only subsequently learns to think linguistically in the interludes between acts of speech. What is called “language acquisition” should really be called “speech acquisition”, because these two things by no means coincide and speech is clearly what is meant. Actually, there isn’t much research about language acquisition in the proper internalist sense—how exactly might this be studied? Speech is much easier to observe than inner monologue (or dialogue). We will never get a good psychology of language, or a good philosophy of language, or a good linguistics until we recognize that language is an inner mental phenomenon—speech being just an imperfect and adventitious sign or symptom of it. If we want to take a linguistic turn, it needs to be an internal linguistic turn. If that should prove impossible, we are shit out of luck.[4]

1. See, for example, chapter 1 of What Kind of Creatures Are We? (2018). Also, Why Only Us: Language and Evolution (2016).

[2] Chomsky makes this point on p.14 of Creatures.

[3] Strictly speaking, there is no “phonological component” in the study of language, though there is in the study of speech. No single thing has both a syntax and semantics and a phonetics.

[4]Prim reader, please excuse the vulgarity, but there exists no polite way to express the underlying thought. We might see the reluctance to accept the mental nature of language as a fear of natural mystery: if language is essentially a hidden psychological structure, lurking obscurely in the brain, then the study of language could be prone to eternal ignorance, or at least methodological intractability. We can’t even examine it under a microscope. By contrast, a behaviorist approach gives us hope of progress. This is a case of changing the subject in order to find something to say. A cruel dilemma. I should note that introspection is hardly stellar when it comes to discerning the workings of mental language. The question of whether we think in the language we speak is notoriously maddening and obscure. My sense is that we do not, except around the edges.