surest proof that there are two concerned. Of such real identity, then, it would seem we must have direct experience; and we have it first of all in the continuous presentation of the bodily self; apart from this it could not be 'generated' by association among changing presentations. Afterwards, other bodies being in like manner personified, that then is regarded as one thing -from whatever point of view we look at it, whether as part of a larger thing or as itself compounded of such parts—which we take to have had one beginning in time. But what is it that is thus assumed to have had a beginning and to continue indefinitely? This leads to our last point.

e. So far we have been concerned only with the combination of sensory and motor presentations into groups and with the differentiation of group from group; the relations to each other of the constituents of such a group still for the most part remain. To these relations in the main must be referred the correlative concepts of substance and property, the distinction in substances of qualities and powers, of primary qualities and secondary, and the like1.

Of all the constituents of things only one is universally present, that above described as physical solidity, which presents itself according to circumstances as impenetrability, resistance or weight. Things differing in temperature, colour, taste and smell agree in resisting compression, in filling space. Because of this quality we regard the wind as a thing, though it has neither shape nor colour, while a shadow, though it has both but is nonresistant, is the very type of nothingness. This constituent is invariable, while other qualities are either absent or changeform altering, colour disappearing with light, sound and smells intermitting. Many of the other qualities-colour, temperature, sound, smell increase for us in intensity if we advance till we touch a certain body occupying a certain place; with the same movement too its visual or 'apparent' magnitude increases. At the moment of contact an unvarying tactual magnitude is ascertained, while the other qualities and the visual magnitude reach a fixed maximum; then first it becomes possible by effort to

1 The distinction between the thing and its 'properties' is one that must be more fully treated under the head of Real Categories (cf. ch. xiii, § 6). Still, inasmuch as the objective warrant for these concepts is contained more or less implicitly in our percepts, some consideration of them is in place here.

change or attempt to change the position and form of what we apprehend. This tangible plenum we thenceforth regard as the seat and source of all the qualities we project into it. In other words, that which occupies space is psychologically the substantial. It is strange that Locke did not lay more stress on this point; though, to be sure, in common with Descartes he recognised it as the one sense-datum that is a primary quality. But neither remarked that this 'sense-datum' is sui generis in being the only one that the subject gives to itself, or at any rate, gets for itself by its own activity, as we have already seen. The other real constituents are only the properties or attributes of this substance, the marks or manifestations which lead us to expect its presence.

Perception as partly re-presentative.

§7. But there is still an observation concerning percepts that we must not omit, though the full discussion which it opens up must be deferred1. Even the simplest percepts, we have seen, involve not only present experience but also experiences of the past in the language of Herbert Spencer they are 'partly presentative, partly representative.' On this account it has been usual to say that all perception implies both memory and imagination. But such a statement, we must here remark, can be allowed only so long as the terms memory and imagination are vaguely used. The dog's mouth normally waters only at the sight of food, but the gourmand's mouth will water even at the thought of it. We recognise the smell of violets as certainly as we recognise the colour when the spring brings them round again; but few persons, if any, can recall the scent when the flower has gone, so as to say with Shelley

Odours, when sweet violets sicken,

Live within the sense they quicken

though most can recall the colour with tolerable clearness. In like manner everybody can perform innumerable complex voluntary movements which only a few can mentally rehearse or describe without the prompting of actual execution. And

1 Ch. vii, § 2.

2 It can however be brought to water at the sight of any coloured object, a particular dish say, that has become associated with the food.

not only does such reproduction as suffices for perception fall short of that involved in reminiscence or memory in the narrower sense, but the manner in which the constituent elements in a percept are combined differs materially from what is strictly to be called 'the association of ideas.' To realise this difference we need only to observe first, how the sight of a suit of polished armour, for example, instantly reinstates and steadily maintains all that we retain of former sensations of its hardness and smoothness and coldness; and then to observe next how this same sight gradually calls up ideas now of tournaments, now of crusades, and so through all the changing imagery of romance. Though the percept is complex, it is but a single whole, and the act of perception is single too. But, where, as is the case in memory and imagination, attention passes—whether voluntarily or non-voluntarily-from one representation to another, it is obvious that these several objects of attention are still distinct and that it is directed in turn to each. The term 'association' seems only appropriate to the latter. To the connexion of the partial presentations in a complex, whether perception or idea, it will be better to apply the term 'complication,' which was used in this sense by Herbart, and has been so used by many psychologists since. When we actually perceive an orange by sight we may say that its taste or feel is represented, when we perceive it by touch or taste we may in like manner say that its colour is represented. The whole complex may be symbolized sufficiently for our present purpose, in the first case as Ctf, in the second as Fct. We might also symbolize the idea of an orange as seen by ctf and the idea of an orange as felt by fct, using the accented letter to signify that different constituents are dominant in the two cases. What we have yet to observe is briefly (1) that the processes by which the whole complex tf or f'ct is brought into consciousness differ importantly from the process by which C or F reinstates and maintains the parts, tf or ct, and (2) that c, t, and ƒ seem never to have that distinct existence as representations which they had as presentations or impressions'.

1 Cf. next chapter, §§ 2 and 3.

CHAPTER VII

IMAGINATION OR IDEATION1

Impressions and Ideas distinguished.

SI. Before the intuition of things has reached a stage so complete and definite as that just described, imagination or ideation as distinct from perception has well begun. In passing to the consideration of this higher level of mental life we must endeavour first of all analytically to distinguish the two as precisely as may be, and then to trace the gradual development of the higher.

At the outset we have to note the distinction between impressions and ideas, which Locke with his epistemological bias too much overlooked, but which Hume placed in the forefront of his Treatise. "All the perceptions of the human mind," he begins, "resolve themselves into two distinct kinds, which I shall call Impressions and Ideas." Both alike may be either 'simple or complex,' he tells us: the difference between them "consists in the degrees of force and liveliness with which they strike upon the mind, and make their way into our thought or consciousness." In all this Herbert Spencer blindly followed Hume. But it is very questionable whether Hume was right in applying Locke's distinction of simple and complex to ideas in the narrower sense as well as to impressions. Regardless of his first statement that they are distinct in kind he goes on to say: "That idea of red, which we form in the dark and that impression which strikes our eyes in the sunshine differ only in degree, not in nature." What he seems to overlook is that, whereas we may once have received the bare

impression called 'red,' we now usually have an image or idea

1 Ideation-"a word of my own coining" said James Mill.

2 Treatise of Human Nature, vol. I. pt. i. § I.

only of a red form or a red thing, i.e. of red as it was present in a percept, in some way ideationally projected or intuited. An incomparable observer in this department, in the course of summarising his results, remarks: "I have succeeded a few times in seeing bare colours without object: they then filled out the entire field of sight." In other words, we seem to have no 'ideas' or images-though we have concepts-answering to simple or isolated impressions. The complication which has taken place during the evolution of the percept can only partially fail in the image or idea, can never fail so far as to leave us with a chaotic 'manifold' of mere sensational remnants. On the contrary, we find that in 'constructive imagination' a new kind of effort is often requisite in order partially to resolve these representational complexes as a preliminary to new combinations. But it is doubtful whether the results of such a process are ever the ultimate elements of the percept, that is, are merely isolated impressions in a fainter form.

As to the one difference, which Hume finally recognised'the force or liveliness' of primary presentations or impressions as compared with secondary presentations or 'ideas,' what exactly are we to understand by this somewhat figurative language? A simple difference of intensity can hardly be all that is meant; for, though we may be momentarily confused, we can usually perfectly well distinguish the faintest impression from an image: moreover, we can imagine such minimal faintness as easily as the maximal". Between moonlight and sunlight or again between midday and dawn we can discriminate many grades of intensity; but it does not appear that there is any corresponding variation of intensity between these extremes when they are not seen, but imagined. Many persons suppose they can imagine a waxing or a waning sound or the gradual abatement of an intense pain; but what really happens in such cases is probably not a rise and fall in the intensity of a single representation, but a change in the complex represented. In the primary presentations there was, if not a change of quality along

1 G. H. Meyer, Untersuchungen über die Physiologie der Nervenfaser, 1843, p. 241. I have repeatedly tried to repeat this among other of Meyer's experiments and, as it seemed, with occasional success; but the colour was far more like a sensation than an image, as was undoubtedly Meyer's experience.

2 The whole subject of the intensity of representations, however, awaits experimental investigation.