complex, if representation is to 'imply the presence of a free or independent idea. To call this “ something ” a tied or nascent idea on the ground of its possible later development into an independent representation seems, then, nearest the truth. The same meaning is sometimes expressed in a wholly different and designedly paradoxical way, by saying that all cognition (perception) is recognition. This statement has been met by elaborate expositions of the difference between knowing and knowing again, the irrelevance of which any lexicon would show; and, further, by the demand: How on such a view is a first cognition possible, or how is an indefinite regress of assimilation to be avoided? We may confidently reply that it cannot be avoided: an absolute beginning of experience, whether phylogenetic ally or onto genetically, is beyond us. Assimilation means further assimilation; in this sense all cognition is further cognition, and a bare sensation is, as said, an abstraction representing a limit to which we can never regress.
We find evidence, again, of ideas in the making in what Lewes called preperception. Of this instances in plenty are furnished by everyday illusions, as when a scarecrow is hailed by the traveller who mistakes it for a husbandman, or when what is taken for an orange proves to be but an imitation in wax. In reality all complex percepts involve preperception; and, so far, it must be allowed that such percepts are directly analysable into preventative-representative complexes. Nevertheless, the representative element is not yet, and may never become, an idea proper. The sight of ice yields a forefeel of its coldness, the smell of baked meats a foretaste of their savour. Such prepercepts differ from free ideas just as after-percepts do: they are still sense-bound and sense-sustained. Nor can this complication be with any propriety identified either with the association pertaining to memory or with that specially pertaining to ideation; though, no doubt, the two processes-complication and association-are genetically continuous, as are their respective constituents, nascent and free ideas* The whole course of perceptual integration being determined and sustained by subjective interest, involves from the outset, as we have seen, concurrent conative impulses; and thus the same assimilation that results in familiarity and preperception on the subjective side results in facility and purpose on the conative. Knowing immediately what to do is here the best evidence of knowing 'what there is to do with; the moth that flies into the candle has assuredly no preperception of it, and does not act with purpose. Bearing this in mind, we may now see one way, and probably the earliest, in which tied ideas become free.
The contrast between the actual and the possible constitutes, as we have seen, the main difference beween experience at the perceptual and experience at the ideational stage. A subject confined to the former level knows not yet this difference. Such knowledge is attained, not through any quasi-mechanical interaction of presentations, but usually through bitter experience. The chapter of accidents is the Bible of fools, it has been said; but we are all novices at first, and get wisdom chiefly by the method of trial and failure. Things are not always different in what to us are their essential properties, but they so differ from time to time. Resemblances are frequent enough to give us familiarity and confidence; yet uniformity is fiecked by diversity, and thwarted intentions disclose possibilities for which we were not prepared. What was taken for sugar turns out to be salt; what was seized as booty proves to be bait. We catch many Tatars, and so learn wariness in a rough school. In such wise preperceptions displaced by the actual fact yield the “ what ” severed from the “ that, ” the “ ideal ” freed at length from the exclusive hold of the real. In a new situation after such adventures the attitude assumed-if, for brevity, we describe it in terms of our own still more advanced experience-is of this sort: “ It may be a weasel, if so, I back; it may be a rabbit, if it is, I spring.” Instead of unquestioned preperception that “makes Hence the earlier process has been named “ impressional association ” (Stout, Analytic Psychology, 1896, ii. pp. 27-29), and again “animal association " (Thorndike, Animal Intelligence, an Experimental Study of the Associative Processes in Animals, 1898, pp. 71, 87, and passim). But it seems preferable to confine the term “ association " to the later process, in which alone the component presentations have that amount of distinctness and individuality which the term properly connotes.
the mouth water, ” we have the alternative possibilities presenx as “ free ideas, ” and action is in suspense, the alternative courses, that is to say, again present only in idea. It is easy to see how in such situations one free idea, a “ what ” sundered from its “ that, ” will tend to loosen the sensory ties of alternative, still implicit ideas. On the cognitive side, from immediate assimilation an advance is made towards mediate cognition, towards comparison; on the active side there is advance from impulsive action towards deliberate action?
We conclude, then, that implicit ideas-the products of assimilation, and integrated as such in complex percepts and the motor co-ordinations to which they lead-are more likely to emerge as free ideas the more this perceptual complexity increases. Perception in the lower animals, who give no signs of either memory or ideation, has apparently no such complexity. A fish, for example, can feel, smell, taste, see, and even hear, but we cannot assume solely on that account that it has any percepts to which its five senses contribute, as they do to our percept, say, of an orange or a peppermint. Taking voluntary movements as the index of psychical life, it would seem that the f1sh's movements are instigated and guided by its senses, not collectively but separately. Thus a dog-fish, according to Steiner, seeks its food exclusively by scent; so that when its olfactory bulbs are severed, or the fore-brain, in which they end, is destroyed, it ceases to feed spontaneously. The carp, on the other hand, appears to search for its food wholly under the guidance of sight, and continues to do so just as well when the fore-brain is removed, the mid-brain, whence the optic nerves spring, seeming to be the chief seat of what intelligence it has? Again, Bateson observes: “ There can be no doubt that soles also perceive objects approaching them, for they bury themselves if a stroke at them is made with a landing-net; yet they have no recognition of a worm hanging by a thread immediately over their heads, and will not take it even if it touch them, but continue to feel for it aimlessly on the bottom of the tank, being aware of its presence by the sense of smell.”4 To this inability to combine simple percepts into one complex percept of a single object or situation we may reasonably attribute the fish's lack of true ideas, and consequent lack of sagacity. The sagacity even of the higher animals does not amount to “general intelligence, ” such as enables a child “ to put two and two together, ” as we say, whatever “ two and two ” may stand for. So far as life consists of a. series of definite situations and definite acts, so far the things done or dealt with together, the contents of the several foci or concentrations of attention, form so many integrated and comparatively isolated wholes. Round the more complicated of these, and closely connected with them, free ideas arise as sporadic groups, making possible those “lucid intervals, ” those fitful gleams of intelligence in the very heat of action, which occasionally interrupt the prevailing irrationality of the brutes. And as we cannot credit even the higher animals with general trains of ideas, just as little can we credit them with a continuous memory: indeed, it is questionable how far memory of the past, as past, belongs to them at all. For they live entirely in an up-stream, expectant attitude, and it is in this aspect that “ free ideas ” arise when they arise at all. We cannot imagine a dog regretting, like one of Punch's heroes, that he “ did not have another slice of that mutton.”5
The free idea (a) then at its first emergence has neither an assignable position in a continuous memory-record, as of or ag, nor has it a definite relation as a “ generic idea ” to possible specializations such as a' or a". These further developments bring us to the general consideration of mental association. 2 Some light is perhaps here thrown on the reciprocal relation of “ association by contrast " and “ association by similarity " as severally the differentiation of partial similars and the integration of partial dissimilars.
3']. Steiner, Die Fnnctionen des Centralneroensyterns n.:.w., 2te Abth. Die Fische (1888), pp. 50, 126, 19 seq., 1or. W. Bateson, “The Sense-Organs and Perceptions of Fishes, " Journ. Marine Biol. Assoc. (1890), p. 239.-5
Cf. Stout, Manual of Psychology (1899), vol. ii. ch. i.; also F. H. Bradley, “ Memory and Inference, " Mind (1899), pp. 145 sqq.; and especially Thorndike, Animal Intelligence, cited above.