CHAPTER XIV.
THE CONDITIONS OF KNOWLEDGE
EX PARTE SUBJECTI.

I. -- THE SENSES.

§ 256. We have seen that every being in the universe is for the realist ontologically true; it has a nature, a structure, a function, a purpose, all of which are knowable. We have also seen that every being is active, and is capable of producing effects in other things, which in so far as they depend upon it as upon their cause, must resemble it in some one of its characteristics. These are the conditions of knowledge ex parte objecti.

Ex parte subjecti we have a human being -- for it is with human knowledge alone that we are concerned -- and that being is endowed with all the characteristics of a sentient organism, and also with the higher functions of intellect and will. He is a substance, a unity of ground amid structural differences, and all his faculties and functions work together for a common end. He is passive, and in him other things may produce effects like to themselves. He is sentient, and therefore he can feel these effects he has an intellect and so can apprehend their meaning. He is also active. Consequently he can move about and so can bring himself into contact with an indefinite variety of things in an indefinite variety of circumstances. He has the power of action and hence, guided by his reason, can manipulate and experiment upon the objects which make up his environment, and in this way progressively increase both his knowledge of their nature and his control of their activities.

§ 257. What, then, is this knowledge that man possesses, and how does he come to get it? There are two questions here, and they must not be confused one with the other. I have said above that man is conscious of the effects produced in himself by physical objects; but, ordinarily, he is not conscious of them as effects. He does not reason within himself 'these are effects due to external objects, but effects must resemble their causes therefore these effects give me knowledge of objective reality.' I am going to argue in this way; but then I am treating the matter metaphysically, not psychologically, or from the point of view of ordinary common-sense. What are de facto sensations the form of which is determined by the causal action of external objects, man spontaneously and naturally apprehends, not as sensations, nor as 'effects,' but as the qualities of natural objects. The species sensibilis, therefore, is ordinarily the id quo percipitur, not the id quod percipitur; and similarly in regard to the intellect. For, as Aquinas says, "the species intelligibilis is to the intellect as the species sensibilis is to the sense. But the species sensibilis is not illud quod sentitur, but rather id quo sensus sentit. Hence the species intelligibilis is not id quod intelligitur, but id quo intellectus intelligit." Ideas, then, are the means by which we think of objects. De facto they correspond to the objects about which we think, for otherwise they could not give us knowledge of objects; but we do not first think of our ideas and then infer an object to correspond; we think directly of the object through and in the idea.

Knowledge, therefore, is one thing; but how we come to know, or what are the metaphysical conditions of knowledge is quite another matter. Yet to define knowledge, apart from the process by which it is acquired, is impossible. For knowledge is an ultimate notion. Everybody knows what it is, because everybody knows. But he cannot define it unless he also knows its metaphysical conditions. Knowledge implies a certain peculiar relation between thought and thing; or, again, it may be regarded as an action or a habit of mind which, being given, that relation arises; and it will be better, perhaps, to leave the matter in this vague form until we have discussed further the conditions of knowledge.

§ 258. We have, then, on the one side an intelligent knower and on the other side the knowable, things, material, organic, sentient, rational and, finally, God. Of all these we somehow have knowledge. Yet, if the knowable is distinct from the knower, just as one knower is distinct from another, how are they to be brought together unless we are to admit that abomination of Professor James, a salto mortale. The Aristotelian answer is by interaction.

Knowledge implies that the formal characteristics of the knowable are somehow immanently present to the knower. Consequently, as we have agreed to regard the knower and the known as existentially independent, and since in our world the only means by which one thing can influence another is by its actions, for knowledge to be acquired, directly or indirectly the knowable must act upon the knower in such a way as to reproduce in him that characteristic of itself which is known. Things, however, do not act directly upon the mind, but only by way of the body; and then consciousness is not affected unless they act upon that particular part of the body which we call the terminal organs of sense. Yet when things do act upon the organs of sense, they must -- according to our metaphysical principle that the effect resembles the cause -- produce therein a modification which is like to themselves in so far as it is an effect which depends upon their action as upon its cause. On the other hand, in receiving an effect, the recipient is not merely passive. The effect is received secundum modum recipientis; or, in other words, while the particular form which the effect takes is determined by the activity of the cause, its specific nature is determined by the potentia passiva, which is thus reduced to act. Hence, when effects are produced in a sentient organism, they must in general be felt; not because sensation is something superadded to the modification produced in that organism; but because the modification itself is a transitus de potentia ad actum, which takes place in a potentia passiva of a certain specific (scil., sentient) nature. Whenever sensation occurs, it is at once a species impressa and a species expressa. It is a species impressa because it is a modification, the form of which is determined by the action of an object external to the organ of sense and it is expressa because it is not merely a modification of the bodily organ, but a sensation having a particular quality.^{2}

§ 259. The correspondence here affirmed between sensation and its cause may seem to many crude, if not a sheer absurdity. It is, however, essential to Aristotle's theory of knowledge; and though physiology and physics have made great progress since the days of the famous Greek, it is only the details and not the general principles of his theory that have been affected. As a theory consisting of a complexus of hypotheses and principles, it cannot be established wholly a posteriori, nor can it be verified point by point by comparison with the facts; not, indeed, because these are relevant facts which it cannot account for, but because our 'facts,' in physiology especially, are comparatively scanty, and many things which are sometimes asserted as facts belong in truth to physiological theory. I would ask the reader, therefore, to hold whatever prejudices he may have in abeyance. To argue as some do that the 'correspondence notion of truth' is absurd, because copying is useless, is to beg the whole question. For if copying can give us knowledge and so enable us to increase the control we have of the things around us, and thus to make them more and more subservient to our needs, it certainly is not useless. While to affirm as many do, that sensation cannot be a copy of its object or cause, is again irrational; for it assumes that we know what the objective qualities are to which sensations are supposed to correspond, whereas, in fact, we know very little about them, and in some cases, smells and tastes for instance, nothing at all. Let us then suspend judgment until we have examined theories and facts, both physical and physiological, and I think that in the end we shall find there are none with which the Aristotelian theory of knowledge is incompatible.

Our theory affirms that sensation is like its object or cause; and two reasons may be given for this assertion, (1) that it follows logically from the general principle that the effect resembles its cause, and (2) that sensation being the id quo percipitur, though not itself either perception or knowledge, must correspond with its object, otherwise it could not be the empirical source whence all knowledge is derived. Were these principles generally recognised as axiomatic, we might leave the matter here; but as they are not, we must examine in detail the psychophysical process of sensation, in order to discover how far it resembles what we know -- or suppose that we know -- of the physical properties of objects.

§ 260. A sensation -- or, better, perhaps, a sense-impression -- has many characteristics; but the 'quality' of sensation is the characteristic in reference to which it is most difficult to establish correspondence. In regard to it I shall endeavour to establish two points (1) that what physicists and physiologists have to tell us about the matter tends to confirm rather than to upset our theory; and (2) that, so far as 'quality' is concerned, a precise correspondence is not required.

Let us first consider the species impressa, which I regard as identical with the nervous processes that are the immediate antecedents or physiological conditions of sensation and perception. Sensation is undoubtedly closely connected with the nervous impulse, and probably arises, so thinks Professor McDougall, as that impulse passes through the synapses that connect one neurone with another. Of the nature of a nervous impulse we know nothing for certain; but we do know that nervous impulses must differ, because sensations themselves are different. The question is, then, what is the ultimate cause of these differences? Are they determined objectively by the stimulus, or subjectively by the 'specific energy of the nerves'? The latter view originated with Professor Müller, and a short time ago was very widely accepted among physiologists. Professor Wundt, however, rejects it in favour of a more objective theory and it seems to me that he does so with good reason.

The 'Specific Energy' theory affirms that the quality of sensation must be due to the structure of the nervous tissues and not to the physical stimulus because (1) a specific sensation may be produced by other than 'adequate stimuli,' and also (2) by stimulating the cut end of a nerve; and (3) the Young-Helmholtz theory of colour-vision, and the Helmholtz theory of audition presuppose this hypothesis.^{3} These arguments, however, are by no means conclusive. For (1) the possibility of producing sensations by other than normal stimuli and in cases where the end organ has been lest or destroyed, may be accounted for in the same way as memory and imagination, viz., by functional differentiation, of the nervous tissue due to habitual activity; and (2) the Helmholtz theories presuppose not necessarily specific differentiation of the nerves, but of the nerve terminals in the eye and in the ear. Moreover, there is not the slightest anatomical evidence for original specific difference in the nature of the nerves or of the cells which belong to the various organs of sense but only in regard to their peripheral end-organs. Hence we conclude with Professor Wundt that both centrally and peripherally the general law holds that repeated stimulation of a particular kind changes the nervous substance and gives it a tendency to exhibit a certain specific process in whatever way it may be stimulated ; but that whereas peripherally specialisation of the nervous elements is inherited, centrally it is not, but is acquired during the life-time of the individual. "The central function is that of combining and inhibiting peripherally excited impulses and only later of reproducing such impressions spontaneously, owing to modifications of the substance of the neurones due to previous excitation." What we inherit cerebrally are "dispositions consisting in neural connections, and hence favouring certain kinds of reaction."^{4}

§ 261. Physiologically, then, there is no evidence whatever, so far as the nervous impulse and the quality of sensation are concerned, which can be urged against our metaphysical theory. On the contrary, if, as Professor Wundt says, "functional differentiation of nervous matter is due to special conditions arising from external circumstances and external stimuli," it seems probable that the purpose of peripheral adaptation is that the effect of the stimulus may be the more easily produced. And we are confirmed in this opinion when we take into consideration physical theories in regard to the constitution of the various objects perceived by the senses. If light is a periodic motion in the ether, and colour a property of the molecules in a body in virtue of which certain rays of light are absorbed and others reflected, clearly that a periodic motion of (say) 500 billion vibrations per second should be able to affect the organ of sense requires some special adaptation; and the same may be said of heat-waves which are also periodic, but of considerably greater length, and therefore require a different adaptation in the end organ. Similarly for pressure if it be due to a bombardment of molecules. I do not wish to suggest, of course, that the rapid vibrations of a hypothetical ether are reproduced in the organ of sight or of temperature precisely in the same manner that they exist in the ether (or in what Aristotle would have called the Diaphanum); for, colour differences on this hypothesis being quantitative, it would be sufficient if the relative values of the wave-lengths were preserved; and there are grounds for thinking that the propagation of nervous impulses is periodic.

It is, however, hardly worth while seeking for further analogies between physiological processes and the interpretations put upon the sensibile proprium of the various senses by physical theory. The uncertainty of both our physical and our physiological theories makes all speculation of this kind decidedly premature. Nevertheless, what knowledge we have seems to confirm rather than to disprove our contention that an organ of sense is a potentia passiva, the actuation of which is due to a stimulus, and ultimately to an object, to which the species impressa corresponds.

§ 262. We come now to the species expressa, to sensation itself, and here, if our physical theories are correct, there is no correspondence in the literal and punctual sense between the quality of the sensation and the objective quality of the thing. Exact correspondence in every detail, however, is not required; indeed, if there were a correspondence of this kind sensation would practically be useless for knowledge. Knowledge begins with sense-experience, and what we want first are not minute analyses, but broad distinctions. We do not want to know how many billion times a second there is a vibration in the ether, but what in general are the qualities of an extended thing which we wish to distinguish from some other thing. And for purposes of distinction the 'qualities' of sensation are extremely efficient. They do not tell us anything about wavelengths or about the chemical constitution of a molecule, for their function is not to give us a knowledge of details, but to represent them synthetically like the mathematician who plots out a curve. And, as a matter of fact, the stimulus itself comes to us in synthesis. Hence, if we did get a distinct sensation corresponding to each element in the stimulus as analysed by the visual substances of the retina or by the basilar membrane in the cochlea, the representative value of sensation, as well as its utility for knowledge, would be far less than what it is. The first condition of all knowledge is the power of discrimination, and the accuracy of the senses in this respect could scarcely be greater than it is. The normal eye never mistakes a wave-length of .000656 mm. (red) for one of .000527 mm. (green); still less do our senses confuse these shorter wave-lengths of light with the longer ones of heat. Stimuli, whether differing quantitatively or qualitatively, are most accurately distinguished by means of the 'qualities' of sensations, and are automatically arranged in classes under the general headings of colour, heat, sound, pressure, flavour, odour, etc. Thus, though the quality of sensations tells us nothing of the nature of objects, it tells us a great deal about their differences, and, consequently, is of immense value for knowledge, since thereby we are enabled to distinguish one thing from another and so to make them the objects of further research. Nor can it be said that the senses deceive us by leading us to believe that to be objective which is really subjective. If you ask the plain man what colour is, he will reply that it is a quality in the object, and this is true; for, according to the scientists, it consists in the power of the object to analyse light by absorbing some rays and reflecting others. That the objective quality, colour, is of such a nature, the plain man does not know, for his senses tell him nothing of the nature of colour. But he does know that colour is a quality of objects, and that colours are objectively different; and his knowledge is valid scientifically as well as for common-sense. Even in regard, then, to the quality of sensation, there is a certain correspondence between it and the qualities of objects; and though the correspondence is not precise or detailed, it is sufficient, and were it greater than what it is, the value of sensation as the id quo percipitur would be lessened instead of increased.

§ 263. Positive knowledge, however, is derived not so much from the 'quality' of sensation as from other characteristics which correspond to the 'primary' qualities of objects or to sensibilia communia, i.e., to qualities of objects perceptible by more senses than one. All sensations seem to have extension or extensity, and in what is presented by sight or by touch, there is a spatial arrangement of qualitative differences which is the basis of our knowledge of figure, shape, distance, magnitude and number; while changes within a presentation and succession among presentations themselves are data presupposed by our knowledge of motion and rest, time, change, duration, etc. Now, between the extensity and configurations which characterise sense-presentations, and objective extension and figure, as we understand it, there is clearly a correspondence; and this correspondence seems to apply also to the intermediate link, the species impressa, or the physiological modification produced in the nervous system. At any rate, it is true in regard to the peripheral terminals of the organs of sight and touch. The impression on the retina corresponds -- approximately of course -- to the figures and spatial arrangements of the things which we see. Nor is it of any consequence that the image on the retina should be reversed, for spatial positions are essentially relative. There is no such thing as an absolute upside-down. Moreover, positions observed by sight are co-ordinated with those observed by touch -- a fact which may be proved by experiment. For if spectacles which again reverse the retinal image are used, this coordination is disturbed, and things at first appear upside-down relative to positions determined by touch. But after a time a new co-ordination of the two sets of spatial impressions takes place, and we see things the right way up. For touch, too, the configuration of the surface of a body with which we are in contact corresponds approximately with the configuration of the end-organs of touch in the part affected; though the correspondence here is far less accurate than it is in sight, and depends largely upon the number of nerve-terminals which are present in the part of our body which is affected. So much so, in fact, that experiment is able to show that an illusion occurs where, for instance, a straight edge (say, the edge of a piece of paper), is applied to a line of terminals in which a distinct bend occurs, for in that case the edge of the paper itself appears to be bent. Nevertheless, even in 'touch,' there is a rough correspondence between the configuration of the object touched and of the nerve-terminals affected.

§ 264. The perception of shape, and especially of distance, is due, however, not merely to touch and sight proper, but also to what is called 'the muscular sense.' A movement of the eye involves sensations arising from the muscles which bring about that movement, from the muscles concerned in the processes of accommodation and convergence, and more especially from the close-fitting socket of the eyeball. A movement of the hand or foot involves sensations arising from the skin, the joints, the muscles, the sinews and the periosteum. In each of these two sets of sensations, not only coordination, but also 'fusion' occurs, and the product is a perception of three dimensions by sight or by movement or by touch. In what a 'fusion' of sensations consists, or how they fuse, it is impossible to say; hence it is impossible to show precisely how there is 'correspondence' in regard to the sensations involved in the perception of three-dimensional space. But it is clear that such a correspondence is possible, for not only does each element in the complex imply an intensively graduated series of sensations, but many of these elements have spatial values of their own. Thus the sensations arising from each of the joints in the finger, the wrist, the arm, and the shoulder are capable of giving us a perception of motion in two dimensions. Indeed, looking at the matter in the abstract and disregarding altogether the fact that our limbs do move in a space of three dimensions, I do not see why, if there had been a fourth or a fifth dimension, we should not have been able to perceive it. For if our sensations are arranged in a number of series such that each can give us a perception of two dimensions (a, b), (a', b'), (a'', ''), (a''', b'''), why should not these series combine in such a way that (v.g.), while the a's fused to form a connecting link between the series, the b's should remain distinct? In the abstract such a combination seems possible, and had it actually happened, we should have perceived, it would seem, not three, but (in the above case) five dimensions. The fusion, however, does not take place in this way, for its product de facto is the perception of space in three dimensions. And since we cannot account for this either physiologically or psychologically, it would appear that its cause must be objective. Whence we may infer that our space is three-dimensional, and that the combination of sensations to form a complex in which distance is perceived is itself objectively determined.

§ 265. So far as concerns motion and rest, duration and number, it can hardly be disputed that there is some sort of correspondence between psycho-physical processes and objective reality. But in regard to the psycho-physical processes which condition the perception of space, exception may be taken to my proof on the ground that it rests upon false assumption. I have endeavoured to show that there is a correspondence between the spatial arrangement of the things we perceive and the impression made by them upon our senses; but this correspondence, it will be said, applies only to the impression made on the peripheral end-organs, and I have no right to assume that these are psychophysical processes at all.

Two replies may be made to this objection. First of all, the older view that the stimulation of the peripheral end-organs is itself accompanied by consciousness does not seem to have been disproved. In fact, it seems to me probable that the whole afferent process is psycho-physical. The possibility of arousing specific sensations by stimulating the cut-end of a nerve can be explained, as I have said, by the differentiation of cerebral paths arising from their habitual use. The occurrence of hallucinations in a person who has been deprived of some organ of sense may be similarly explained; for the vividness of the hallucination is sufficiently accounted for by the abnormal condition of the brain in such a case, a condition which implies excessive activity and irritability in certain parts. Again, the lag of sensation may easily be accounted for by inertia in the end-organs themselves. While, on the other hand, if the modification produced in the end-organ is not a psycho-physical process, it is difficult to account for the marked difference between a sense-impression and an image.

Secondly, even if sensation is in the brain and not in the sense-organ itself, it cannot be proved that the spatial arrangement to be found in the sense-organ does not hold also in the cerebral cortex. No physiologist, so far as I am aware, has ever succeeded in tracing an individual chain of neurones from its terminal organ to its corresponding centre (if it has one) in the cortex. And even if this could be done, and it should turn out that the chain of neurones do get mixed up, so to speak, 'local signs' would establish all the correspondence that is necessary. For if each nerve-terminal has a characteristic 'local sign,' it does not much matter where sensation first arises, since that 'local sign' will be quite sufficient to determine the 'place' of the stimulus.

§ 266. The scholastic dictum, then, that cognitum est in cognoscenti secundum modum cognoscentis, does not seem so far to have invalidated the general principle that omnis cognitio fit secundum similitudinem cogniti et cognoscentis. Sensation is an effect produced in a sentient organism by an objective cause which it resembles; and that resemblance is not destroyed by the co-operation of the organism in the production of the effect. Before leaving sensation, however, and passing on to intellectual cognition, I should like to point out one characteristic of sense-impressions which seems to offer a striking confirmation of our causal theory. I refer to what is known as 'external projection.' Our sensations, in external perception, are projected outwards as it were. Apart altogether from the fact that man perceives through them, i.e., intellectually apprehends in them the qualities of objects, the latter appear at a distance. Distance, in other words, is felt as well as perceived. Hence localisation as a mere datum of experience does not imply thought. Yet appearances are localised not in the brain or the organ of sense, nor yet in the objective medium, but in their determining cause. Colour is projected back into the object; sounds are referred to an external source, and in touch when an instrument is used, such as a walking stick, we feel not with the hand, but with the end of the stick. Feelings again are localised in the hair, the beard, and in the teeth, no matter whether the latter be false or real, dead or alive. In short, independently of thought-reference, feelings are localised or projected back into their causes, regardless of medium, of instrument, and of the organ alike. This projection, both normal and 'eccentric,' is, doubtless, a function of the aforesaid 'muscular sense;' but how is it to be accounted for and what is its significance, if not that the form or 'species' of the sense-impression is determined by the objective cause?

From the arguments here adduced it should, I think, be clear that sensation has a representative as well as an affective value. The question, therefore, which now concerns us is how that representative value is, so to speak, converted into knowledge. Sensation, and even sense-perception -- if we abstract from thought, which in man usually accompanies it -- is not knowledge. On the other hand, according to the scholastic, "nihil est in intellectu quod non fuit prius in sensu." How, then, does this transition take place? How does sensation generate ideas and so acquire a meaning? This is a question which I shall endeavour to answer aristotelico modo in the following chapter.

{1} Summa Theologica, p. I, q. 85 @ 2. Note: All scholastics grant that in the idea or concept we know the object immediately, though some prefer to say that the verbum mentale or concept is id in quo res intelligitur rather than id quo intelligitur. And there is some foundation for this view, since in the thought of absent objects we are aware that we are thinking by means of ideas, and the idea, as well as the object, is, in a certain sense, that about which we think, or, in other words, we are conscious that we are thinking of an object which is not really present, but present only in idea. Cf. Aquinas, De differentia verbi divini et humani, and Urraburu, Psychologia, Lib. II, disp. 2, cap. 2, art. 3.

{2} The account of the act of perception given in this chapter is based upon the theories of Aristotle and Aquinas; but, the more adequate knowledge which we now possess of physiological psychology has made it necessary to introduce certain modifications. The species impressa, for instance, is not identified by Aquinas with the physiological process that underlies perception but is merely the species sensibilis quâ an effect. Indeed, the term species impressa is not used by him at all; though he admits an immutatio naturalis in the organs of sense which resembles the external cause, and which is to sensation as matter to its form and similarly, it would appear, in regard to cerebral processes that accompany and condition the phantasm, or what we should now call the percept or image.

{3} cf. McDougall, Physiological Psychology, pp. 58 et seq.

{4} Wundt, Physiological Psychology, vol. I., p. 320 (Eng. trans.).