Is Language a Practical Capacity

It is sometimes said, with an air of obvious truth, that mastery of one’s native language is a practical capacity.[1] The suggestion sounds reasonable enough, even somewhat illuminating: we do useful things with words, perform tasks, achieve stuff; we don’t speak just to broadcast propositions into the atmosphere. This is obviously correct for Wittgenstein’s builders, memorialized in section 2 of the Investigations, because they are engaged on a practical task. We also learn in section 11 that words are like tools, also things that are involved in practical tasks. But is it true that all uses of language are thus practical? Is all speaking and listening practically oriented? Some speech clearly serves a practical purpose, but does all of it? That depends on what is meant by “practical”. It can’t just mean “action involving” because many actions are not practical in the usual sense, e.g., writing philosophy or reciting poetry or kicking a ball around. It must mean a specific type of action. The OED gives us “the actual application or use of a plan or method, as opposed to the theories relating to it” for “practice”; hence “practical” means “not merely theoretical”. In speech we apply the rules of grammar and vocabulary, as opposed to theorizing about those rules (the job of the linguist). But to do what? You might say to communicate, or to aid thought: these are the practical purposes to which language is put. But communicate about what, or aid thinking about what? Suppose someone is discussing the puzzles of quantum physics: is that a practical task? If you are discussing theories, are you engaged in a practical activity? What about reciting poetry—is that a practical activity? If so, the slogan “language is a practical capacity” is reduced to a truism. No, we must mean something more limited than merely performing an action or else nothing we do will ever count as impractical. And, of course, that is not how the word “practical” is commonly understood: it is associated with such words and phrases as “useful”, “work-related”, “functional”, “businesslike”, “down-to-earth”, “pragmatic”, “effective”, “helpful”, “utilitarian”—not theoretical, intellectual, aesthetic, impractical, amusing, fun, enjoyable, devoid of all practical value. Plumbing and carpentry are practical, ballet and theoretical physics are not. Are your feet on the ground or your head in the clouds? Language used in practical contexts (“Hand me a slab!”) is practical, but not language used in theoretical or intellectual or aesthetic contexts (“Shakespeare is superior to Marlowe”). It is simply not true that all language use is practical in the proper sense.

Is seeing a practical capacity, or remembering, or thinking? Some types of seeing, remembering, and thinking are indeed practical, if they are employed about practical tasks; but not all types of seeing are thus practical. Seeing a picture in an art gallery is not practical, nor is remembering a concert, nor is thinking about the mind-body problem. Not every activity of mind is geared towards solving practical problems (abstract theorizing is not). That is the whole point of the concept—its contrast with the unpractical. It is the same with language: some language use is devoted to pursuing impractical ends—artistic, scientific, philosophical, playful. This is obvious and is only occluded by the platitudinous idea that language use is a type of action, as if that fact alone entails that language mastery is a “practical capacity”. You may as well say that all thinking is a practical capacity. Could it be maintained that impractical activity is necessarily derivative from practical activity? That doesn’t seem like a conceptual truth, and anyway it wouldn’t undermine the case against the doctrine at issue. That doctrine is just plain false. It is true that it is wrong to picture language mastery as consisting solely in the contemplation of abstract propositions or platonic essences (in the style of the Tractatus), but it is equally wrong to assimilate it to issuing orders on a building site. We can overintellectualize language use, but we can also underintellectualize it. If part of the intent of the doctrine at issue is to “naturalize” language use by comparing it to reflexes and habits, then that is a misguided project; language mastery is a complex cognitive achievement, even in the case of Wittgenstein’s builders. In fact, even this kind of “primitive” linguistic behavior embeds sophisticated cognitive processes (see psycholinguistics), and grammar itself is a mighty cognitive structure. We might better say that a deeply “intellectual” accomplishment is sometimes employed for mundane practical purposes: language is really “slumming it” when it descends to supplement a practical chore. Its real nature shows itself best in Shakespeare, the great philosophers, the poets. This is the essence of human language, with its infinite scope and power, not the monosyllabic “slab” and “block” (Wittgenstein himself describes such a language as “primitive”). We would do better to describe human language use as unpractical in its essential constitution; it raises us above nature, above brute practicality. We should not be language philistines.

What philosophy of language might the doctrine of practicality be ranged against? To whom and what does it react? And is this reaction justified? Frege provides a good example (the Tractatus descends partly from him). Frege depicts the language user as engaged in some pretty fancy mental gymnastics (the same is true of his philosophy of mathematics): the whole apparatus of sense and reference, function and object, is like a giant jigsaw puzzle that the mind must somehow solve. Just to understand a simple sentence, you have to be able to negotiate an elaborate construction of moving pieces slotted precariously together. You are said to “grasp a thought”, itself a quasi-platonic entity, that bears no essential relation to actual language use, let alone practical endeavors. One would never think this airy contraption might be useful in getting a house built or ordering lunch! It seems like a language suitable for gods not men. Frege seems quite unconcerned with psychological reality and the concrete activity of speaking and hearing. Those simple builders know nothing of sense and reference, thoughts and truth-values, functions and objects. Isn’t the practicality doctrine a useful corrective to this kind of intellectualist thinking? Actually no, and for a simple reason: Frege is primarily concerned with syntax and semantics not pragmatics. He could allow that pragmatics emphasize the practical side of language and stick to his elaborate theory of syntax and semantics. The mind must operate at all three levels in order to organize speech, and the pragmatic level enjoys a certain autonomy with respect to syntax and semantics. Nothing in Frege’s semantics denies the practical (and impractical) uses of language at the pragmatic level; and this can be as pragmatic as you like. Language use could be practical through and through without prejudice to semantics. Of course, the linguistic mind can’t just be a reflex mechanism if it grasps the Fregean structure—it can’t just be an input-output device. But it can still be as practically oriented as it pleases, as unconcerned with matters intellectual and theoretical. Practicality of result doesn’t preclude complexity of structure. These are orthogonal issues. Given that practical tasks come in an enormous variety, it is likely that a complex semantic system underlies their possibility: the builders might also be cooks, car mechanics, landscapers, and fishermen. All that is going to require a sophisticated machinery of moving parts, the full complement of human language as we know it. And anything above the simplest type of practical behavior calls for theoretical understanding of principles and means-end reasoning. Theoretically limited practice is limited indeed. Really, the whole idea of the noble savage speaker, always bent over the tools of his trade, is a myth; he will talk trash and nonsense (as well as poetry and astronomy) at least some of the time. Human language will always burst the bonds of the boringly practical. It will not wither away as life becomes less practically burdensome. Language is not essentially a practical device.[2]

[1] Michael Dummett used to say this a lot, but it is a common refrain, especially in ordinary language philosophy.

[2] I think the picture of human language here sketched is broadly Chomskian, what with his emphasis on grammatical structure and creativity. The practical activity conception is closer to Skinner-type behaviorist linguistics. I do wonder how philosophers of language, whose use of language is seldom practical, could have arrived at the view that the essence of language is practical, as if manual workers are the true speakers. Reverse snobbery? Language is above all flexible.

Memory and Expectation

How do memory and expectation differ? For example, I might remember going to the shops yesterday and expect to go to the shops tomorrow—how do these states of mind differ? They concern the same state of affairs, but they are evidently not the same; we never confuse one with the other (“Am I remembering going to the shops or expecting it?”). It might be thought that the answer is simple: memory is directed to the past while expectation is directed to the future—that is the essential and sole difference. This may be backed up by the claim that the same state of affairs that is now remembered was once expected and vice versa: time passes and the mind reflects how the world is at a given time, marking it as past or future. We simply have a different temporal perspective on the world analogous to our varying spatial perspective on things presently seen. You can believe that an event lies in the past or in the future, but the belief element is constant—it is just a belief state with a different temporal content, signified by tense. Beliefs about the past are the same as beliefs about the future except that they embed different propositions. Similarly, memory embeds the proposition that in the past p, while expectation embeds the proposition that in the future p. Memory is a belief about the past and expectation is a belief about the future.  We can call this the “symmetry theory”. If we turn to the dictionary, we receive some confirmation of the symmetry theory: “remember” is defined as “have in or be able to bring to one’s mind (someone or something from the past)”; “expectation” is defined as “a strong belief that something will happen or be the case” (OED). We could paraphrase the former as “be able to form an accurate belief about the past” and the latter as “have a belief about the future”. What distinguishes these beliefs is the occurrence of distinct types of temporal reference in their content, either to the past or the future. Crudely, remembering is thinking about the past and expecting is thinking about the future.

But this is too crude. In the first place, when we remember we think with both concepts: we think something was the case in the past but also that this something itself had a future, i.e., what followed the event remembered. We remember the event as occurring in time, with a past and future at that time. In the same way, when we think of a future event, we think of it as having had a past—as positioned in time. We employ this complex of concepts. Second, it is of course possible to have a belief about the past which is not a memory of the past, as with my belief that the earth is over three billion years old; not all knowledge of the past is remembering that which is known. Nor is memory always accompanied by belief: I might be remembering a real event and yet feel unsure about whether my memory is reliable in this instance (compare perceptual hallucinations). Expectation is much closer to belief, though even here I might find myself with an attitude of expectation but question whether it is well founded (some drug keeps causing me to have false expectations). So, what is the difference? What is the essential nature of memory and expectation? How should these concepts be analyzed? Now things become more difficult (as with everything about time). It won’t do to say that we can have knowledge of the past via memory but not knowledge of the future via expectation, because skepticism can be broached in both areas, and knowledge of the future is as extensive as knowledge of the past. We are on fairly safe ground if we observe that memories are caused by past events, while expectations are not caused by future events, though this observation by itself is less than explanatory. How does the attitude of remembering differ from the attitude of expecting, as opposed to the causal history of the attitudes in question? Causation is extrinsic to the inner nature of these attitudes. If in some remote possible world, the causal stories were inverted, would that change the nature of the attitudes? If the future caused expectation and the past failed to cause memory, would the attitudes be inverted, so that we remember the future and expect the past? I don’t think so. We need to find something more intrinsic, definitional, phenomenological.

I can think of two marked differences in the psychology of memory and expectation. The first is that we can fear the future but not the past. Remembering a past event does not evoke fear in us to the effect that the past might hurt us—the past is past, dead and gone. Recollecting a close call with a dangerous animal does not make us afraid of an attack by that animal now. But if we expect an attack, that certainly evokes fear. We are afraid of the future, as of the present, but not of the past. That is a big difference psychologically. Memories of the past might cause us to fear something similar in the future, but we are not afraid of the past as such (those events). The second point is that memory is saturated with detail and specificity, while expectation is etiolated and general. If I remember meeting someone yesterday, I can recall concrete details of our encounter drawn from my earlier experience; but if I expect to meet someone tomorrow, there is no experiential input to the expectation, just a schematic outline. Memory is experientially saturated, but expectation is not, since we have no experience of the future to draw upon. The future is conceived thinly, but the past is conceived thickly. Thus, the past seems to us more real than the future, less conjectural. To remember is to relive, but to expect is not. The past is part of me, but the future isn’t, not yet. My past makes me me, but my future doesn’t make me me (now). The past is bound up with personal identity; not so the future. This is why theorists of personal identity cite memory as crucial but not expectation and the future. I am what has happened to me, not what willhappen to me. I am not so much my plans for the future, which may or may not come to fruition, as I am my past experiences, which have definitely happened and shaped me. When I remember I am confronted with myself, but when I expect I am confronted with the world; I might not even survive into the future, but I sure as hell existed in the past. The past is fullness (Being, as Sartre would say); the future is emptiness (Nothingness). A human life is part Being and part Nothingness: an actual history and a speculative hope (or fear). Phenomenologically, memory is the repository of a lived life, while expectation is a receptacle waiting to be filled by life. Memory is replete with being, but expectation is striving after being. Non-being is the threat posed by the future; being is the accomplishment of the past. Fear is the emotion proper to the future, while regret is proper to the past (or pride and self-satisfaction). The difference between memory and expectation, at the phenomenological human level, is the difference between life as already lived and life as it might or might not be lived. Fact and supposition, reality and dream, the concrete and the conjectural. We remember as creatures of an experienced world; we expect as creatures of a world as yet unexperienced. These are very different attitudes. Human existence in time is suspended between the two attitudes. They are our constant companions. We are expectant memoirists.[1]

[1] I wrote this paper after having read Wittgenstein’s discussion of expectation in Philosophical Remarks (1930). In this work there is no trace of behaviorism. He is concerned with the logical nature of expectation, its specific mode of intentionality (he doesn’t use this term).

Common Sense and Philosophy

Philosophers often assume that there is something called common sense, or commonsense belief, with which their theories may agree or disagree. This gives rise to the idea that there is such a thing as commonsense philosophy, or anti-commonsense philosophy. I think this is a mistake for a number of reasons. The first reason is that the phrase “common sense” doesn’t mean what it is taken to mean: it doesn’t describe a category of beliefs but a way of approaching problems.[1] The OED gives us “good sense and sound judgment in practical matters”. To have common sense is to have the ability to solve practical problems in a sensible manner, i.e., not to do things in a silly or inefficient way. It is not common sense to believe that the earth goes round the sun, or that we see material objects, or that we know we have two hands. It is common sense to wash a car with soapy water not beer, or to look before you leap. The phrase “common sense” is not synonymous with “common beliefs”: the latter phrase means something like “widely held beliefs”. Nor does it mean “beliefs people hold without knowing any science or philosophy”. The phrase “common sense beliefs” is nonsensical: there are common sense methods but not commonsense beliefs. You can have beliefs that don’t require any specialized knowledge, but these are not instances of common sense: you are not exercising your common sense if you believe that the sun rises in the morning or that there are a great many material things in the world. So, the phrase “common sense (beliefs)” does not identify any class of beliefs that might conflict with philosophical theories or theses. It is a misnomer, a kind of category mistake. Its cash-value appears to be something like “beliefs that don’t derive from science or philosophy”, though even that formulation is riddled with difficulty (how about the belief that the post office is closed on Sunday?). So far, then, we have not identified a class of beliefs that are philosophically significant and in conflict or agreement with philosophical theories, but not the province of philosophy as an academic discipline. We have no general characterization of such a class of beliefs. Having good practical judgment is not the same as having a specific set of beliefs with which a philosopher might agree or disagree.

Part of the idea behind the offending phrase is that of beliefs that everyone shares. Some beliefs are held by a select few, often specially trained, but some are held by everyone just in virtue of being human. So, at any rate, it is supposed. We thus have the idea of universal beliefs—what the human species believes as a matter of course. We can then oppose these species-wide beliefs to the beliefs of specially trained philosophers (or scientists). Just as there are linguistic universals and conceptual universals and perceptual universals, so there are doxastic universals—things we all believe just by being human. Indeed, these beliefs may be shared by creatures other than the human kind, just in virtue of their being believers in anything (animals, aliens, non-conformists of various stripes). Let us then substitute the phrase “universal beliefs” for “commonsense beliefs” so that we can say that there are universal beliefs with which the philosopher might agree or disagree. For example, it might be said that we all believe that we see material objects and that philosophers of perception might agree or disagree with that proposition. The trouble with this formulation is that it is false: there are no such universal beliefs. Not everyone believes that we see material objects—some may be agnostic or skeptical or in open dissent or just uncomprehending. The thing about beliefs is that they are friable, labile, and mercurial: they vary tremendously between people and over time, and without much effort. People naturally disagree; and their beliefs can change at the drop of a hat. We don’t have fixed beliefs as we have fixed eye color: that is why it is not difficult to persuade people that we do not see material objects (but only sense-data). We can’t easily change how we see or smell things, but we can effortlessly change what we believe. Universality is not the natural condition of belief; diversity is. That is the joy of belief, but also its curse: you can believe virtually anything, which sets you free, but it also sets you up for error and credulity. Some people are constitutionally contrary, making a point of believing what others don’t believe. Beliefs are plastic and malleable; hence they are not naturally or nomologically universal. And the more contestable they are the more disagreement there will be; thus, the closer to philosophical disputes the less universal. Take free will: the belief in free will is not humanly universal, precisely because it is open to philosophical dispute. I do not think it is possible to find a single belief that even approximates to universality that is of philosophical relevance. Maybe everyone believes that the earth beneath their feet is solid not gaseous, but that is not a belief with which a philosopher might disagree. Similarly for the belief that one is human or has a body or goes to sleep regularly: some eccentrics might not believe even these things (the nature of belief will not prevent them), but their widespread acceptance is of no relevance to philosophy, since no philosopher disputes them. There is no philosophical school whose main contention is that people are not human or have no body or never sleep.

So, is there nothing to the alleged clash between what people ordinarily believe and what philosophers have contended? Not quite, because some of our beliefs seem to clash with certain philosophical doctrines. I have already mentioned one such case, which is the standard example of the alleged collision between “common sense” and philosophy, viz. the question of whether we (really) see material objects. Is it true that some ordinary people (though not all) believe that we see material objects, while philosophers have denied that? Of course, no one with any knowledge of visual perception thinks that we always see material objects, since we can suffer visual illusions and hallucinations; the question is whether we ever see material objects. It is thought that a great many people believe that we do sometimes see material objects, while some philosophers of perception have denied this. Isn’t that an example of disagreement between ordinary people and (some) trained philosophers? Aren’t many people naïve realists and hence in disagreement with what certain philosophers have maintained? Here is where things get interesting: for it is really not clear that there is any genuine disagreement between the parties involved. Does your average perceiver, who has had no exposure to philosophy, believe that he sees material objects and not surfaces of objects or sense-data of them? Highly doubtful. It would be different if he went around saying as much in so many words, but he doesn’t. Isn’t it just not plausible that he has any such convoluted technical-sounding belief? So, what does he believe about perception? He says things like “I saw the cat a couple of hours ago” in reply to the question “Have you seen the cat recently?” Normally he doesn’t make perceptual statements at all, and if he does it is in a very non-analytical way. If you ask him about sense-data, he will look at you blankly—he has no views about such things. He doesn’t even use general phrases like “material object”—he just says he saw the cat or a car in the driveway. If you press him about sense-data, he might reply “I’ve never thought about it, but now that you bring it up, I do rather think that I see those things more directly than cats and cars”. He isn’t rejecting such notions in his ordinary statements (and corresponding beliefs), so he isn’t committing himself to the doctrine of naïve realism. Nor is his casual talk of seeing cats and cars to be understood as evincing the belief that he sees such things “directly”; he might well allow that he sees them only “indirectly”—yet he still can be said to see them. This is obviously true for surfaces: he might quickly allow that what he really sees directly are surfaces of objects, but point out that this is consistent with seeing whole objects via seeing their surfaces. Do children who employ the verb “see” in conjunction with terms for cats and cars really believe that they see material objects and not surfaces or sense-data? Hardly. The truth is that the philosophical theory of sense-data is not in contradiction to what the ordinary perceiver thinks about perception, if he thinks anything; it is only in contradiction to the philosophical theory of naïve realism, which is a product of the philosophical mind not the mind of the ordinary man, woman, or child. Perhaps such a theory could emerge in the mind of the ordinary person without any formal instruction in philosophy, but it is still a philosophical theory, not a piece of ordinary thought (“common sense”). We may thus venture a generalization: philosophical theories conflict only with other philosophical theories not with what we might call ordinary thinking. Naïve realism is therefore not part of pre-theoretical thought; nor is ordinary thought inconsistent with what philosophers have claimed about perception (they might well concede that in a loose sense we do see cats and cars—they are not invisible like atoms or remote galaxies). This means that there is no such thing as “common sense philosophy” that can agree or conflict with professional philosophy. Ordinary thought and philosophical thought proceed at different levels, the latter being at a meta level with respect to the former. Your ordinary man on the Clapham omnibus is not a closet philosopher, equipped with beliefs that are in contradiction with philosophical theories, whether good or bad. There is really no such thing as ordinary bloke philosophical belief.

Are there any areas in which we can detect a genuine conflict between philosophy and what many people ordinarily believe or assume? I will briefly mention three possible areas intended to illustrate sharp disagreements between philosophical theories and ordinary beliefs. The first is free will: don’t people generally believe in free will, and yet some philosophers have denied its existence? Those philosophers think that determinism rules out free will but many people seem to believe in free will. The obvious reply is that the people who believe in free will define it as freedom from external constraint, in which case it is not in conflict with determinism. Thus, there is no contradiction between what they believe and what incompatibilist philosophers maintain. Those who define free will as requiring indeterminism don’t tend to believe in it (a minority). So, it’s not clear there is anyone left who disagrees with the philosophers. Probably many ordinary people have an inchoate philosophical theory of what free will involves, so they are not counterexamples to the claim that all supposed disagreements with ordinary belief are really disagreements between philosophical theories. The second example is knowledge: don’t the common folk think we have knowledge in cases in which philosophical skeptics think we don’t? Isn’t there a clear conflict here? But again, there are nuances to be noted and distinctions to be drawn: the ordinary chap may well be only too ready to agree that many of our claims to knowledge fall short of absolute certainty, and hence are not strictly knowledge, but insist nevertheless that in a loose sense uncertain propositions can be said to be known—and doesn’t the skeptic agree with this? Not all beliefs are equally unjustified, so we use the word “know” to distinguish the better ones; this is pragmatically sensible. The alleged disagreement is therefore not as sharp as one might initially suppose. The ordinary person doesn’t hold that there are no differences in degree of justification, and he accepts that very few propositions admit of conclusive justification; how we decide to use the word “know”, however, is a matter of pragmatics. We ordinarily describe a room as empty when there is no chunky stuff in it, but it is not as if we deny that it has air or dust in it. It is not clear that ordinary people explicitly or implicitly reject the basic epistemology of the skeptic, though they speak loosely and pragmatically of knowing things. Generally, they have no philosophical opinions at all, so none that are contradicted by other philosophical opinions. They may come to have such opinions as a result of philosophical reflection, but they don’t harbor them in their bosom all along (do young children believe an epistemology that conflicts with that of the skeptic?). Third, and least easily disposed of, we have Zeno’s paradox: surely, it will be said, the folk walking down the street believe that things move! Isn’t it a “commonsense” belief that objects travel from A to B, and isn’t this denied by Zeno? But again, we have to ask what exactly the folk believe (including what children and animals believe regarding motion). Do people believe that objects traverse space in such a way as to occupy successive intermediate places in a continuous manner? That sounds pretty fancy, more like a theory. It is true that people and animals believe they are at different places over time, but that isn’t in conflict with Zeno; he just denies there was continuous spatial transposition over time. To what extent is our understanding of motion conditioned by a theory of how things get to be in different places? Is Zeno really denying what people actually hold as a matter of their basic psychology? There surely could be a being that only thought of motion in the bare-bones terms I just sketched (“I was at home this morning but now I’m at the shops”), in which case Zeno’s paradox would not contradict anything this being actually believes. How do we distinguish between having a theory of something and merely accepting its existence? At the very least it is much harder to detect conflict between philosophy and basic belief than many philosophers have supposed. Are we reading more into ordinary belief than is really warranted? So-called common sense is not proto-philosophy. Philosophers have a tendency to over-intellectualize the thoughts of the non-philosophical thinker.[2]

[1] I believe the first person to make this point was Norman Malcolm, commenting on G.E. Moore.

[2] I plead guilty to this: I never questioned the rightness of the view here criticized for decades. I accepted that ordinary people think things that philosophers sometimes deny. I thought of people as mini philosophers. Many, no doubt, are, but not because it is built into what they already think. If I am right in what I say here, large tracts of philosophy are quite misguided. Philosophy is neither revisionary of ordinary belief nor in accordance with it; the two stand cognitively apart.

Qualities of Mind

Galileo, Descartes and Locke divided the physical world into primary and secondary qualities, thus carving out a place for the science of physics (the study of the primary qualities of matter). But they said nothing about the primary and secondary qualities of the mind, in the hope of carving out a place for the science of psychology (the study of the primary qualities of mind). What should we say about this question? The answer is not immediately clear: what would such qualities be? Someone might say that the answer is obvious: since secondary qualities are mental and primary qualities are non-mental, all the qualities of mind are secondary, none primary. The indicated conclusion might then be that there is no science of the mind comparable to physics, on account of a lack of proper subject matter for such a science. You can’t make science out of the merely subjective. Alternatively, and less drastically, it might be said that the relevant qualities belong to the body and are simply a special case of the familiar primary qualities of physical things. Bodies have shape, size, weight, number, and motion, so that we can identify bodily behavior as the proper subject matter of the science of psychology. The so-called mental attributes are all secondary qualities and therefore do not belong to the science of the mind, i.e., behavioral science. In either case, what we naively think of as the mind is consigned to the realm of the merely subjective—like color, sound, smell, etc. This realm contains no primary qualities for the simple reason that the mind does not have shape, size, weight, number, and motion—only the body does. Psychology can thus never be the same kind of science as physics (it might confine itself to introspective reports). It’s either behaviorism or nothing.

It seems to me that this is the exact opposite of the truth. It completely misconceives the concepts of primary and secondary qualities. True, the examples given by the originators of these concepts are as listed, but that is not the same as the underlying conception. That conception invites us to consider the qualities of things that constitute their intrinsic nature as opposed to qualities conferred on them from outside. When we switch subject-matter we must expect a different array of qualities. If we consider the primary qualities of numbers, we don’t expect to find the same qualities that apply to physical objects; we expect to find, and do find, mathematical qualities (being even, prime, motionless, weightless, imperceptible). In the case of the mind, then, the primary qualities will include whatever characterizes the intrinsic nature of the mind, as opposed to how we might subjectively represent it. And there is nothing to prevent us identifying these primary qualities as irreducibly mental. Thus, we might count as primary qualities of mind both phenomenological and propositional qualities (possibly also functional and computational qualities). These are what the mind inherently is, its inner constitution, its objective nature. The science of psychology will then study these kinds of qualities (properties, processes). Just as the standard primary qualities of physical objects comprise how objects are in themselves, independent of subjective viewpoint, so these primary qualities of minds (phenomenological and propositional) fix how minds are in themselves, independent of subjective viewpoint. The qualities that were declared extraneous to physics turn out to be central to psychology: colors, say, were transferred to the mind away from physical objects, but now these mental qualities become the primary qualities of psychology—what that science is really about. They are primary for psychology.

But this raises an interesting question: does the mind also have secondary qualities? What would that mean? It would mean that our ordinary ways of representing the mind introduce one perspective among many possible perspectives on the mind itself. Martians might see what we see as red objects as green, or taste our sweet things as bitter—might we likewise experience our own minds in a particular way possibly not shared by Martians? Might we introspect our minds via subjective representations of those minds? For example, might we introspect our experiences of red as experiences of green? This would be like experiencing the same wavelengths of light in two different subjective ways—as red or as green. The same mental primary quality (seeing red) might be experienced in two different ways—as an experience of red or as an experience of green. The mental quality might have two different dispositions to appear to a subject depending on his psychological make-up. That would be the analogue of the same physical object appearing in different ways to perceivers. If this were so, then that would be a genuine mental secondary quality. There would be a real distinction between mental primary qualities and mental secondary qualities. But surely there is no such thing, nor could there be. We are not at that kind of remove from our own minds; we don’t project qualities onto our minds in that way. I don’t have a subjective viewpoint onto my own mind, which might differ from other viewpoints onto the same (type of) mind. I project colors onto objects, but I don’t project experiences of red onto my own mind. Accordingly, there are no mental secondary qualities; all mental qualities are primary. The mind is not a composite of primary and secondary qualities, as physical objects are, but a collection of uniformly primary qualities, objectively possessed. The qualities of mind are never merely powers to produce impressions of those qualities in that mind or in other minds.

The absolute conception of the physical world supplied by physics involves abstracting away from the human perspective on the physical world; it is subtractive, invidious. But the absolute conception of the mind is not like that: it treats all attributes of mind as part of its subject matter. So far from ignoring subjectivity, it embraces it. In psychology the subjective is what is objective (real, absolute). And of course, this is true: what belongs to the mind objectively exists—though it may not exist “for physics”. If we try to mimic the formation of physics, carving out a subset of the qualities normally attributed to things, we end up with no proper subject matter for psychology (only behaviorism); but we should not attempt to duplicate this kind of movement of thought. There is really no primary/secondary distinction in psychology. Much the same is true in mathematics: there isn’t a collection of attributes of numbers found in common sense mathematics that needs to be jettisoned in order to put mathematics on a scientific footing—as it were, the colors and smells of numbers. There are just primary mathematical qualities. This is the right model for the science of psychology not physics. The metaphysical apparatus of primary and secondary qualities developed by Galileo, Descartes and Locke to put physics on a sound basis does not carry over to the psychological case. In particular, the subtractive impulse is out of place.[1]

[1] It is an indication of how undeveloped psychology was as a science in the seventeenth century that no one raised the question of whether psychology might admit of a primary/secondary distinction analogous to physics. As far as I know, the question has not been raised till now.