Analogical thought is now understood to be imaginative only, and is confined to the metaphors and similes of poetry or rhetoric; though it is not very long ago that it was seriously trusted in argument, as in defending absolute monarchy in the State by the examples of patriarchy in the family, and even by the supposed “regiment” of bees and quails in their societies, of the lion over beasts and of the eagle over birds.
In this spirit, Malays, having identified the life of the rice plant with human life, regard the flowering rice as in its infancy, and proceed to feed it with pap: and carry out the analogy at further stages of its development. Similarly, to facilitate childbirth, or to liberate the struggling soul of the dying, it is a respected recipe to untie all knots, unfasten all buttons, unlock all doors, open all windows; for opening or loosing, no matter what, is always the same process-relation.
These seem to be the chief modes of fallacious thinking—(1) false parallels and (2) analogies—which mislead the untutored mind and give to imagination-beliefs such coherence as they ever attain. Two accounts of superstitious reasoning have been given by those who admit that savages reason at all; one is that they reason correctly from absurd premises; the other that they reason absurdly from correct premises. If the foregoing analysis is sound, there is some truth and some error in both these doctrines. So far as primitive ratiocination is purely analogical, it is quite futile, whether its premises be true or false; for it cannot be cast in any admissible logical form. So far as in superstition it imitates parallel reasoning, according to cause and effect or substance and attribute, the major premise is, for the most part, empirically true; the minor premise is false; and the conclusion is a vain imagination. There are three types of ratiocination: (1) equations, as in mathematics; and here primitive man for a long time got no further than the counting of things by his fingers and toes. (2) Parallels of premise and inference, according to causation or substance and attribute, as in the physical and natural sciences; and here the savage collects by experience much common sense, and by inevitable fallacies much superstition. (3) Analogies of imagination. The natural progress of reason consists in relegating analogies to poetry and rhetoric; in introducing greater and greater accuracy into the judgments that serve as major premises, and greater caution in assuming minor premises; at last, in counting and measuring the facts reasoned about, and so preparing the beginnings of mathematical method. Such progress is promoted by the high biological value of greater definiteness of thought. Immature man in the necessary practical life—which may be called the biological life—has many definite perceptions and judgments and well-adjusted actions; outside that life, in the region of superstitious observance, he is not a rational, but an imaginative animal.
§ 7. General Ideas at the Savage Level
The language of savages is often wanting in names of classes of things for which names are with us a matter of course, and it has been supposed that those who use the language must be without the corresponding general ideas. Thus it is reported that a tribe had a name for each kind of tree but none for tree in general; another had a name for coco-nuts at various stages of growth (when they serve different uses) but none for coco-nut at all times: therefore, it is inferred, they had no general idea of tree or of coco-nut. A Siberian example is still more remarkable. The Tunguses depend entirely upon reindeer for food, clothes, tents and locomotion, and keep herds of them; yet they have no name for the animal. But they have a name for wild and another for tame reindeer; a name for domestic reindeer that have been broken in, and another for the unbroken; a name for the female fawn, for the doe with young, for a doe with one fawn, a doe in the third year with two fawns; a name for each age-class of buck, and so on.[95] Are we to infer that the Tunguses have no general idea of reindeer?
It was, no doubt, natural to assume that we first perceive individuals, which now stand clearly before us, and then, having compared them, arrive at general ideas. But if knowledge grows by the assimilation and differentiation of experiences, the class on the one hand and the individual on the other, must be joint products of this process: classes becoming clearer as more and more individuals are discriminated. Classes, or class-ideas never stand before us as individuals do; but the greater part of the meaning of every perception of an individual is the kind of thing it is. And as to priority, to perceive the kind of thing is (biologically) far more important than its individuality. If the individual did not mean the class, to perceive it would be useless as a guide to action. The general idea derived from the assimilation of experiences is the apperceptive mass that converts sense-stimulation into cognition: when unconscious Romanes called it a “recept.”
The primitiveness of general ideas is shown by gesture-language, which probably precedes speech, and which (except in direct indication of what is thought of) depends wholly upon general ideas suggested by imitative or significant actions. Primitive language must have described things by general characters, so far as it consisted in onomatopœia; to growl like a lion could only suggest the kind of animal. Primitive drawing (whether by children or savages) is nearly always generic: dog, horse, frigate-bird, hammer-headed shark, but not any individual.
There can be no doubt that savages are capable of general and abstract ideas; and no one now supposes that language is an adequate measure of thought. A language contains names only for things, groups, and aspects or actions of things which the people who use it need to discuss: if they do not need to speak of abstractions, there are no words for them. But we cannot assume of the contents of the mind, any more than of the outside world, that things do not exist unless we have noticed and named them. Professor Franz Boas has shown[96] that languages of the northern Amerinds, that do not idiomatically express abstract ideas, may be made to do so without violence, and that the abstract expression is intelligible to men native to the languages. “Every one who knows people of low culture,” says Dr. Rivers, “must recognise the difficulty which besets the study of any abstract question, not so much because the savage does not possess abstract ideas as that he has no words of his own to express them.”[97]
Amongst the ideas attained by savages, and having an important part in their lives, though often taken for granted and unexpressed, are some of the highest generality: for example “force.” The notion of force is derived from the experience of effort in our own muscular exertions; but with the development of our perception of physical objects by the integration of sense-data—sight, touch, movement and resisted movement (kinæsthesis), smell, hearing—this sense of effort, being transferred to objects as equivalent to our own exertions about them, becomes the all-important core of every object (or meaning of every perception of an object), without which the thing would be a mere show, neither useful nor injurious—sheep and tigers, rocks and pumpkins alike indifferent. Probably the perceptions of all the higher animals (down to reptiles) have this meaning. Force is reality, and by primitive man is thought of as the essence of whatever he conceives of as real, such as spells, talismans and ghosts. Having a subjective ground, from which it never becomes free, the notion of it is indefinite, varies for each of us with our constitution, age, health, and not only lends itself to the wildest whims of superstition, but has misled scientific investigations.