82 PSYCHOLOGY by the very insight into things which these analogies have themselves made possible. That in their first form they were mythical, and that they could never have been at all unless originated in this way, are considerations that make no difference to their validity, assuming, that is, that they admit, now or hereafter, of a logical transformation which renders them objectively valid. This legitimation is of course the business of philosophy; we are concerned only with the psychological analysis and origin of the concep- tions themselves. Causal- As it must here suffice to examine one of these categories, let us itv take that which is the most important and, central of the three, viz., causality or the relation of cause and eftect, as that will necessarily throw some light upon the constitution of the others. To begin, we must distinguish three things, which, though very different, are very liable to be confused. (1) Perceiving in a de- finite case, e.g., that on the sun shining a stone becomes warm, we may say the sun makes the stone warm. This is a concrete in- stance of predicating the causal relation. In this there is, explicitly at all events, no statement of a general law or axiom, such as we have when we say (2) "Every event must have a cause, a statement commonly known as the principle of causality. This again is dis- tinct from what is on all hands allowed to be an empirical general- ization, viz., (3) that such and such particular causes have invariably such and such particular effects. With these last psychology is not directly concerned at all : it has only to analyse and trace to its origin the bare conception of causation as expressed in (I) and involved in both these generalizations. Whether only some things have causes, as the notion of chance implies, whether all causes are uniform in their action or some capricious and arbitrary, as the unreflecting suppose, all this is beside the question for us. One point in the analysis of the causal relation Hume may be said to have settled once for all: it does not rest upon or contain any immediate intuition of a causal nexus. The two relations that Hume allowed to be perceived (or " presumed to exist ), viz., con- ti<niity in space of the objects causally related and priority in time of the cause before the effect, are the only relations directly dis- cernible. We say indeed "The sun warms the stone as readily as we say "The sun rises and sets," as if both were matters of direct observation then and there. But that this is not so is. evident from the fact that only in some cases when one change follows upon another do we regard it as following from the other: casual coin- cidence is at least as common as causal connexion. Whence the difference, then, if not from perception? Hume's answer, 'repeated in the main by English psychologists since, is, as all the worJ knows, that the difference is the result of association, that when a chan^ B in an object B has been frequently observed to succeed a change a in another object A, the frequent repetition determines the mind to a transition from the one to the other. It is this determination, which could not be present at first, that constitutes "the third relation betwixt these objects." This "internal i lilt - LI1UU. ICiai/lUlA W^UTT *Al> Wi*v> v^jww ^ f pression " generated by association is then projected; tor tis a common observation that^the mind has a great propensity to spread itself on external objects." The subjective origin and the after-projection we must admit, but all else in Hume? s famous doctrine seems glaringly at variance with facts. In one respect it proves too much, for all constant sequences are not regarded as causal, as according to his analysis they ought to be; again, in another respect it proves too little, for causal connexion is continually predicated on a first occurrence. The natural man has always distinguished between causes and signs or portents; but there is nothing to show that he produced an effect many times before regarding himself as the cause of it. J. fc>. Mill has indeed obviated the first objection epistemplogically by addin^ to constant conjunction the further characteristic of "uncon- ditionality." But this is a conception that cannot be psychologi- cally explained from Hume's premisses, unless perhaps by resolving it into the qualification that the invariability must be complete and not partial, whereupon the second objection applies. ' Uncondi- tional " is a word for which we can find no meaning as long as we confine our attention to temporal succession. It will not do to say both that an invariable succession generates the idea, and that such invariable succession must be not only invariable but also uncondi- tional in order to generate it. We may here turn the master against the disciple : "the same principle," says Hume, "cannot be both the cause and the effect of another, and this is perhaps the only proposition concerning that relation which is either intui- tively or demonstratively certain " (op. cit., p. 391). Uncondition- ally is then part of the causal relation and yet not the product of invariable repetition. Perhaps the source of this element in the relation will become clear if we examine more closely the internal impression of the mind which according to Hume constitutes the whole of our idea of power or efficacy. To illustrate the nature of this impression Hume cites the instant passage of the imagination to a particular idea on hearing the word commonly annexed to it, when "'twill scarce be possible for the mind by its utmost efforts to prevent that transition" (op. cit., p. 393). It is this determination, then, which is felt internally, not perceived externally, that we mis- takenly transfer to objects and regard as an intelligible connexion between them. But, if Hume admits this, must he not admit more? Can it be pretended that it is through the workings of association among our ideas that we first feel a determination which our utmost efforts can scarce resist, or that we feel such determination under no other circumstances? If it be allowed that the natural man is irresistibly determined to imagine an apple when he hears its name or to expect thunder when he sees lightning, must it not also be allowed that he is irresistibly determined much earlier and in a much more impressive way when overmastered by the elements or by his enemies? But further, such instances bring to light what Hume's "determination" also implies, viz., its necessary correla- tive, effort or action. Even irresistible association can' only be known as such by efforts to resist it. Hume allows this when he says that his principles of association "are not infallible causes; for one may fix his attention during some time on any one object without looking farther" (op. cit., p. 393). But the fact is, we know both what it is to act and what it is to suffer, to go where we would and to be carried where we would not, quite apart from the work- ings of association. And, had Hume not confused the two differ- ent inquiries, that concerning the origin of the idea of causation and that concerning the ground of causal inference or law of causation, it could never have occurred to him to offer such an analysis of the former as he does. Keeping to the former and simpler question, it would seem that when in ordinary thinking we say A causes this or that in B we project or analogically attribute to A what we experience in acting, and to B what we experience in being acted on; and the structure of language shows that such projection was made long before it was suspected that what A once did and B once suffered must happen in like manner again. The occasions suitable for this projection are determined by the temporal and spatial relations of the objects concerned, which relations are matter of intuition. These are of no very special interest from a psychological point of view, but the subjective elements we shall do well to consider further. First of all, we must note the distinction of immanent action and transitive action; the former is what we call action simply, and implies only a single thing, the agent; the latter, which we might_with adyan- tae call effectuation, implies two things, i.e., a patient distinct from the agent. In scientific language the agent in an intransitive act is called a causa immanens and so distinguished from the agent in effectuation or causa traiisieiis. Common thought, however, does not regard mere action as caused at all; and we shall find it, in fact, impossible to resolve action into effectuation. But, since the things with which we ordinarily deal are complex, have many parts, pro- perties, members, phases, and in consequence of the analytic pro- cedure of thought, there ensues, indeed, a continual shifting of the point of view from which we regard any given thing, so that what is in one aspect one thing, is in another many (comp. p. 56 c). So it conies about that, when regarding himself as one, the natural man speaks of himself as walking, shouting, &c.; but, when dis- tinguishing between himself and his members, he speaks of raising his voice, moving his legs, and so forth. Thus no sooner do we re- solve any given action into an effectuation, by analytically d tinguishing within the original agent an agent and a patient, than a new action appears. Action is thus a simpler notion than causation and inexplicable by means of it. It is certainly no easy problem in philosophy to determine where the resolution of the complex is to cease, at what point we must stop, because in the presence of an individual thing and a simple activity. At any rate, we reach such a point psychologically in the conscious subject, and that energy in consciousness we call attention. If this be allowed, Hume's critique of the notion of efficacy is really wide of the mark. "Some," 2 he says, "have asserted that we feel an energy or power in our own mind; and that, having in this manner acquir d the idea of power, we transfer that quality to matter, where we are not able immediately to discover it. ... But to convince us how fallacious this reasoning is, we need only consider that the will, being here consider'd as a cause, has no more a discoverabl nexion with its effects than any material cause has with its proper effect . . The effect is there [too] distinguishable and separable from the cause, and cou'd not be foreseen without the experience of their constant conjunction " (op. cit., p. 4.15). This is logical analysis, not psychological; the point is that the will is not considered as a cause and distinguished from its effects, nor in fact consider Treatise of Human Nature, pt. iii., xiv., "Of the idea of necessary connexion. " 2 Hume here has Locke and Berkeley specially in view. Locke as a patient and acute inquirer was incomparably better as a psychologist than a man addicted to literary foppery like Hume, for all his genius, could possibly be. On the particular question, see Locke, u. 21, 6-
