rather than to logical ideals of what it ought to be. Like other forms of purposive activity, thinking is primarily undertaken as a means to an end, and especially the end of economy. It is often easier and always quicker to manipulate ideas than to manipulate real things; to the common mind the thoughtful man is one who “uses his head to save his heels.” In all the arts of life, in the growth of language and institutions, in scientific explanation, and even in the speculations of philosophy, we may remark a steady simplification in the steps to a given end or conclusion, or—what is for our present inquiry the same thing—the attainment of better results with the same means. The earliest machines are the most cumbrous and clumsy, the earliest speculations the most fanciful and anthropomorphic. Gradually imitation yields to invention, the natural fallacy of post hoc, ergo propter hoc to methodical induction, till what is essential and effective is realized and appreciated and what is accidental and inert is discarded and falls out of sight. In this way man advances in the construction of a complete mental clue or master key to the intricacies of the real world, but this key is still the counterpart of the world it enables us to control and explain.
To describe the process by which such insight is attained as a mere matter of abstraction deserves the stigma of “soulless blunder” which Hegel applied to it. Of course if attention is concentrated on X it must pro tanto be abstracted from Y, and such command of attention may require “some pains and skill.” But to see in this invariable accompaniment of thinking its essential feature is much like the schoolboy's saying that engraving consists in cutting fine shavings out of a hard block. The great thing is to find out what are the light-bearing and fruit-bearing combinations. Moreover, thinking does not begin with a conscious abstraction of attention from recognized differences in the way logicians describe. The actual process of generalization, for the most part at all events, is much simpler. The same name is applied to different things or events because only their more salient features are perceived at all. Their differences, so far from being consciously and with effort left out of account, often cannot be observed when attention is directed to them: to the inexperienced all is gold that glitters. Thus, and as an instance of the principle of progressive differentiation already noted (§ 6), we find genera recognized before species, and the species obtained by adding on differences, not the genus by abstracting from them. Of course such vague and indefinite concepts are not at first logically general: they only become so when certain common elements are consciously noted as pertaining to presentations in other respects qualitatively different, as well as numerically distinct. But actually thinking starts from such more potential generality as is secured by the association of a generic image with a name. So far the material of thought is always general—is freed, that is, from the local and temporal and other defining marks of percepts.
38. The process of thinking itself is psychologically much better described as (1) an analysis and (2) a re-synthesis of Thought as Analysis. this material already furnished by the ideational trains. The logical resolution of thought into hierarchies of concepts arranged like Porphyry's tree, into judgments uniting such concepts by means of a logical copula, &c., is the outcome of later reflection—mainly for technical purposes—upon thought as a completed product, and entirely presupposes all that psychology has to explain. The logical theory of the formation of concepts by generalization (or abstraction) and by determination (or concretion)—i.e. by the removal or addition of defining marks—assumes the previous existence of the very things to be formed, for these marks or attributes—X's and Y's, A's and B's—are themselves already concepts. Moreover, the act of generalizing or determining is really an act of judgment, so that the logician's account of conception presupposes judgment, while at the same time his account of judgment presupposes conception. But this is no evil; for logic does not essay to exhibit the actual genesis of thought but only an ideal for future thinking. Psychologically, however—that is to say, chronologically—the judgment is first. The growing mind, we may suppose, passes beyond simple perception when some striking peculiarity in what is at the moment perceived is a bar to its recognition. The stalking hunter is not instantly recognized as the destroying biped, because he crawls on all fours; or the scarecrow looks like him, and yet not like him, for, though it stands on two legs, it never moves. There is thus no immediate assimilation; recognition under such circumstances is in itself a judgment, involving an analysis more or less explicit. But of more account is the further judgment to which it leads, that which connects the new fact with the generic idea. Though actually complex, generic images are not explicitly known as complexes when they first enter into judgments; as the subjects of such judgments they are but starting-points for predication—It crawls; It does not move; and the like. Such impersonal judgments, according to most philologists, are in fact the earliest; and we may reasonably suppose that by means of them our generic images have been partially analysed, and have attained to something of the distinctness and constancy of logical concepts. But the analysis is rarely complete: a certain confused and fluctuating residuum remains behind. The psychological concept merges at sundry points into those cognate with it—in other words, the continuity of the underlying memory-train still operates; only the ideal concept of logic is in all respects totus, teres, atque rotundus. Evidence of this, if it seem to any to require proof, is obtainable on all sides, and, if we could recover the first vestiges of thinking, would doubtless be more abundant still.
But, if we agree that it is through acts of judgment which successively resolve composite presentations into elements that Logical Bias in Psychology. concepts first arise, it is still very necessary to inquire more carefully what these elements are. On the one side we have seen logicians comparing them to so many letters, and on the other psychologists enumerating the several sensible properties of gold or wax—their colour, weight, texture, &c.—as instances of such elements. In this way formal logic and sensationalist psychology have been but blind leaders of the blind. Language, which has enabled thought to advance to the level at which reflection about thought can begin, is now an obstacle in the way of a thorough analysis of it. A child or savage would speak only of “red” and “hot,” but we of “redness” and “heat.” They would probably say, “Swallows come when the days are lengthening and snipe when they are shortening”; we say, “Swallows are spring and snipe are winter migrants.” Instead of “The sun shines and plants grow,” we should say, “Sunlight is the cause of vegetation.” In short, there is a tendency to resolve all concepts into substantive concepts; and the reason of this is not far to seek. Whether the subject or starting-point of our discursive thinking be actually what we perceive as a thing, or whether it be a quality, an action, an effectuation (i.e. a transitive action), a concrete spatial or temporal relation, or finally, a resemblance or difference in these or in other respects, it becomes by the very fact of being the central object of thought pro tanto a unity, and all that can be affirmed concerning it may so far be regarded as its property or attribute. It is, as we have seen, the characteristic of every completed concept to be a fixed and independent whole, as it were, crystallized out of the still-fluent matrix of ideas. Moreover, the earliest objects of thought and the earliest concepts must naturally be those of the things that live and move about us; hence, then—to seek no deeper reason for the present—this natural tendency, which language by providing distinct names powerfully seconds, to reify or personify not only things but every element and relation of things which we can single out, or, in other words, to concrete our abstracts.[1] It is when things have reached this stage that logic begins. But ordinary, so-called formal, logic, which intends to concern itself not with thinking but only with the most general structure of thought, is debarred from recognizing any difference between concepts that does not affect their relations as terms in a proposition. As a consequence it drifts inevitably into that compartmental logic or logic of extension which knows nothing of categories or predicables, but only of the one relation of whole and part qualitatively considered. It thus pushes this reduction to a common denomination to the utmost: its terms, grammatically regarded, are always names and symbolize classes or compartments of things. From this point of view all disparity among concepts, save that of contradictory exclusion, and all connexion, save that of partial coincidence, are at an end.
Of a piece with this are the logical formula for a simple judgment, X is Y, and the corresponding definitions of judgment as the comparison of two concepts and the recognition of their agreement or
- ↑ See Wundt, Logik, i. 107 seq., where this process is happily styled “die kategoriale Verschiebung der Begriffe.”
