in order to decide whether or not they all alike possess the attribute of volume, the sensations are clear; if we are absorbed in work of a different kind, and someone is ex- perimenting with tones in the next room, we still have the tonal sensations, but they are obscure. Lastly, duration is, as its name implies, a temporal attribute; it is the attribute which makes the course of a sensation in time — its rise, poise and fall as process in consciousness — characteris- tically different from the course of another sensation.
54 Sensation All sensations, without exception, possess the attributes of quality, intensity, clearness and duration. The list may be lengthened in two ways: by the splitting up of an attri- bute which has hitherto been regarded as simple, and by the discovery of attributes altogether different from the essential four.
On the first point, we have already remarked that what appears to ordinary observation as a simple attribute of quality may turn out, on closer examination, to be the unanalysed resultant of two or three distinct attributes. This is the case with tones, and also, as we shall see pres- ently, with colours. The second point is best illustrated by reference to sight and touch. Sensations of colour are spread out areally into length and breadth; they appear as spatial extents. And this attribute of extent is part of their very constitution. Reduce the colour to a pin point, and it still occupies space; think away the spatial attribute, and the sensation has disappeared with it. So with press- ures: set the point of a stiff horse-hair lightly down upon the skin, and the sensation is extended, diffused over a mental area. Certain sensations, then, have this attribute of extent; others, as odours and tones, show no trace of it. We come back to these facts, in detail, when we discuss the various classes of sensations.
The longest list of attributes is possessed by colour sensations. What we ordinarily term the quality of colour is the resultant of three qualitative attributes: colour-tone or hue, light-tone or tint, and colour-depth or chroma. To these three must be added the intensive attributes of intensity proper, clearness, duration and extent.
It is to be noted that, in certain cases, the concurrence of two or more attributes gives rise to what we may call an § 13- The Classification of Sensations 55 attribute of the second order. Thus, certain sensations have been credited with an attribute of insistence. They are self-assertive and aggressive; they monopolise con- sciousness, as a forward and pushing guest will monopolise conversation at a social gathering. We speak of the pene- tratingness of odours like camphor and naphthaline; of the urgency or importunity of certain pains or of the taste of bitter; of the obtrusiveness or glaringness of certain lights and colours and tones. This character of insistence is, however, not a new primary attribute of the sensations. It is the resultant of clearness conjoined with quality, or with intensity, or with quality and intensity together.
§ 13. The Classification of Sensations. — Introspection leaves no doubt that the sensations, regarded as qualitative processes, fall into a number of separate groups. All colour sensations, for instance, go together; all tonal sensa- tions go together. Colours, again, are more akin to greys, and tones to noises, than are colours to tones. This kin- ship between certain sense qualities means, in general, that the sensations fall into continuous series, so that one can pass from quality to quality as if along a straight line, without leaving one's path or meeting with a gap or inter- ruption. It is possible to travel in this way from bass to treble, through the middle tones of the scale, or from red to white, through a series of pinks: it is not possible to travel from a colour to a tone.
We may, then, classify sensations in terms of their in- trospective resemblances. We may also, with change of standpoint, classify them by reference to the body, since observation has shown that every group of sensations comes to us by way of a definite, specially developed bodily < 56 Sensation < 56 Sensation organ. The sensations of a particular group will then be those whose parallel bodily processes, in spite of difference of detail, have the same general effect within the organism. We may accordingly speak of eye sensations, muscle sen- sations, etc. Such a list, if completed, would be perfectly accurate.
Finally, we may classify sensations by reference to the stimuli which arouse them. Sensations at large fall into two principal groups, according as their stimulus is external, originating outside the body, or internal, originating within the body. Light, the stimulus to vision, is an external stimulus; muscular contraction, the stimulus to muscular sensation, is an internal stimulus. We therefore distinguish between sensations of the special senses, which are stimu- lated from without, and organic sensations, whose stimulus consists in a changed state of the internal bodily organs from which they come.
Not all the sense qualities that are ordinarily grouped together fall into continuous series, like the series of colours and tones. We naturally think, for example, of the sensations of pressure and temperature as forming a group of qualities, although no transition is possible from one quality of the group to another. We naturally think, again, of the sensation of warmth as very closely related to the sensation of cold, although there are no qualities which join these two sensations, and although their sense organs are distinct. It might seem, then, that the sensations of pressure and tempera- ture are bracketed together simply by reference to the skin as their common organ, and the sensations of warmth and cold simply by reference to their common stimulus. Nevertheless, there is a real introspective resemblance between them. Pressure is more like warmth and cold than it is like tone or colour; and we do not feel the jar, in passing from cold to warmth, that we should feel if the disappearing cold were followed by a low tone or a References for Further Reading 57 faint odour. The kinship which introspection finds among these sensations is, in the last resort, a matter of conscious context: the sensations from part of consciousnesses of the same pattern, make the same sort of connections in consciousness, are more or less in- terchangeable in consciousness.
In making out a list of the various departments of sensation, we are at times compelled to speak in terms of sense-organ or of stimulus, for the simple reason that the sensations themselves have not received a name. There is, for instance, no name to designate the peculiar quality of the sensation aroused by con- traction of striped muscle. Language has developed at the com- mand, not of theoretical interest, but of practical convenience; and there has been no pressing reason for the naming of all the separate sensations. Even in the case of colours, we have terms like violet and orange, the names of flower and fruit, alongside of the much older terms blue, red, etc.; and to denote a particular tone we have to use such clumsy expressions as ' the c of the thrice accented octave.'
References for Further Reading §§ 10, ii. E. B. Talbot, The Doctrine of Conscious Elements, Philo- sophical Review, iv., 1895, 154. M. F. Washburn, Some Examples of the Use of Psychological Analysis in System-Making, Philosophical Review, xi., 1902, E. H. Hollands, Wundfs Doctrine of Psychical Analysis and the Psychical Elements, and Some Recent Criticism: 1. The Cri- teria of Elements and Attributes, American Journal of Psy- O. Kuelpe, Outlines of Psychology, 1909, § 40 (elementary quality of will). § 12. M. W. Calkins, Attributes of Sensation, Psychological Review, M. F. Washburn, Notes on Duration as an Attribide of Sensa- M. Meyer, On the Attributes of the Sensations, ibid., xi., 1904, 83.
E. B. Titchener, Lectures on the Elementary Psychology of Feel- ing and Attention, 1908, Lecture i.
58 Sensation It should be said that current works on psychology differ, not only as regards the nature and number of the elementary mental processes and their attributes, but also as regards the principles and aims of psy- chological analysis in general. The fact is not surprising, when we remember that the fundamental questions treated in the Introduction — the questions of subject-matter, method and problem of psychology — are still in debate. The reader may compare with the discussion of the preceding paragraphs G. F. Stout, The Groundwork of Psychology, [1903] ch. iii.; C. H. Judd, Psychology: General Introduction, 1907, ch. iv. He should, nevertheless, look for underlying agreements rather than for superficial differences. Judd's preface, for instance, opens with the sentence: " There is very general agreement as to the main topics which must be treated in a text-book on psychology."
THE QUALITY OF SENSATION: VISION § 14. The Visual Qualities. — It needs but a casual glance at our surroundings, indoors or out, to assure us that the world of vision comprises a very large number of sense qualities. Besides all the wealth of colour, there is the whole scale of light, from the most brilliant white to the deepest black. Both alike are qualitative systems: black, white and grey, red, yellow, green and blue, — one and all are qualities of sensation, individual and elementary mental processes. To a certain extent, the sensations of light and of colour are independent of each other: a land- scape or a coloured painting may be translated, by pho- tography, into an arrangement of blacks, whites and greys. They are also, however, closely related. We speak of cer- tain colours, without hesitation, as being darker or lighter, that is, nearer black or nearer white, than other colours; and we meet with colours of all grades or degrees, from the full quality, the deep red or rich green, to the merest tinge which is but a step from grey.
If we look, first, at the sensations of light, we find that they form a single linear series, extending from white through the lighter, neutral and darker greys to black. Language has very few words to denote the qualities of this series. We speak of black, for instance, as if it were a single quality. But glance, in succession, at black card- board, black cloth, black velvet, and the black of a com- paratively lightless space, — say, the blackened interior of 59 60 The Quality of Sensation: Vision a long pasteboard tube. You realise at once, not only that these four blacks are qualitatively different, but also that their differences are quite considerable, so that there must be several intermediate blacks between the successive terms of the series. The same thing holds of white. Lay upon the window-sill a sheet of white paper, and on this place a cover-glass, silvered on the under side, in such a position that the glass reflects a patch of uniformly bright sky. The reflected light is astonishingly white, and the white of the paper seems, by comparison, greyish. There are in all, if we count up the distinguishable whites,- greys and blacks, between six and seven hundred qualities of light sensation.
The system of colour sensations is less simple; the colour qualities cannot be arranged upon a single straight line. Let us take, as the arrangement of colours with which we are most familiar, a chart or a projection of the solar spectrum, and let us work through it, from the left or long-wave to the right or short-wave end. On the ex- treme left we have the quality of red. As we travel to the right, the red takes on more and more of a yellowish tinge, until it passes through orange to a pure yellow. Here, then, we have a linear series of qualities, precisely similar to the series of light sensations. Now, at yellow, we change our direction. The yellow gradually becomes tinged with a new quality, that of green; it passes through yellow-green to a pure green. Here is a second line of qualities. Again we change our direction; the green becomes more and more bluish, until it passes through blue-green to a pure blue. Here is a third line of qualities. Once more we change our direction. This time, however, the tinge that our initial quality takes on is not new; the blue becomes § 14- The Visual Qualities 61 increasingly reddish, as we travel to the right-hand end of the spectrum. Here is a fourth line of qualities, but a line which in the spectral series is left incomplete at violet. If we continue it, by adding the purples and carmines, we are finally brought back to our starting-point, — the red of the extreme left. We notice, however, that this red is not, in reality, the starting-point of a psychological colour-line; it is not a pure red, but an orange-red; the red that stands at the beginning of the red-yellow line lies outside the spectrum, toward the carmines.
All the colours that can be distinguished upon these four lines are ultimate qualities of visual sensation. We speak, it is true, of pure red and of orange-red; but these terms merely indicate the position of the qualities upon a colour- line: pure red comes at the beginning, orange-red towards the middle of the line. No orange-red can be analysed, by introspection, into a red and an orange. The lines them- selves, and with them the system of colour qualities, are most simply arranged in the form of a square, with R, Y, G and B at the four corners.
So far, then, we have a single straight line to represent the sensations of light, and four straight lines, forming the closed figure of a square, to represent the sensations of colour. But so far, also, we have dealt with sensations of colour only under one aspect, — that of colour-tone or hue. Colours, as was said above, differ further from one another as darker and lighter. Thus, in the spectrum, yellow is undoubtedly the lightest, and violet the darkest colour. Here, then, is a second attribute of colour, the attribute.of light-tone or tint, in virtue of which a sensation of colour may be matched or equated with a sensation of light. Let us assume that all the hues upon the lines of our colourf>2 The Quality of Sensation: Vision square are of the same tint, and that this tint is that of a grey situated midway between white and black. Then, if the line of light sensations is erected as a vertical, the four colour-lines will lie about it in the horizontal plane, and at ; the level of this middle grey. All the lighter reds, or pinks, will stand in order above the point R, opposite the particular light greys to which their tint corresponds; all the darker reds will stand in order below R, opposite the dark greys to which their tint corresponds; and so with the lighter and darker Y, G and B. The vertical line, which represents the whites, greys and blacks, is thus sur- rounded by a square tube, whose walls are made up of all the hues in every possible variety of tint.
Suppose, now, that a line is drawn from some point on the wall of the tube to the corresponding point upon the vertical line: say, from the point R to the middle grey. The sensations that lie upon this new line form a series of the same hue and the same tint; but they differ in respect of a third attribute, that of colour-depth, or degree of saturation, or chroma. The red at R, which is farthest out from the corresponding grey, is a rich, full, deep red; it contains all the red that a visual sensation can contain, just as a saturated solution in chemistry contains all of the dissolved substance that it can contain; it shows the attri- bute of chroma at a maximum. As we proceed inwards, towards grey, the reds become less saturated, more washed- out; their chroma decreases and finally, when the grey is I reached, entirely disappears.
As these chromatic lines may be drawn from any point upon the square shell to the corresponding point upon the vertical, our representation of visual sensations becomes a solid figure, a square prism. In actual observation, how- §14- The Visual Qualities 63 ever, the lines are not all of the same length; they are longest for hues of a middle tint, shortest for hues of very- dark and very light tints. There are, obviously, but few intermediate steps between a very dark blue and the corre- sponding greyish black, or be- tween a very light yellow and the corresponding greyish white. The square prism thus becomes a double pyramid. At the two poles stand the extremes of white and black; upon the ver- tical axis, between the poles, are arranged the remaining sensa- tions of light. Round the base of the figure lie all hues of a middle tint and of maximal chroma. Between base and poles lie the same hues in all their further variety of tint; all are still of maximal chroma, though the chromatic maximum decreases steadily, above and below. If we cut into the pyramid, from any point on the outside to a corresponding point upon the axis, we lay bare a series of sensations of the same hue and tint, but of varying chroma.
The double pyramid, then, as drawn in Fig. 2, embodies the two systems of visual sensation, sensations of light and sensations of colour, and shows these systems both in their mutual independence and in their mutual relations. There are at least a hundred and fifty distinguishable hues round the base. In counting up the whole number of visual sensations we must, however, take account also of FlG. 2. The Colour Pyramid.
64 The Quality of Sensation: Vision differences in tint and in chroma. These are ultimate differences: a pink is no more analysable by introspection into a red and a white than an orange-red is an- alysable into a red and an orange. All in all, the full tale of the visual elements cannot fall far short of thirty-five thousand.
§ 15. Visual Stimulus and Visual Sensation. — The trains of ether waves which constitute light, in the physical sense, differ in three ways: in wave- length, in wave-ampli- tude or energy, and in wave-form or composi- tion. In general, these three aspects of the ether waves correspond to the three attributes of colour sensation, — hue, tint and chroma. At the red end of the spectrum, the light-waves have a length, roughly, of ^00 \xjl\ at the violet end a length, roughly, of 400 fifx; and the intervening hues are corre- lated with intermediate wave-lengths. If we increase or decrease the energy of these waves, within certain limits, the colours as a whole become lighter or darker, — there is a change of tint; but the distribution of hues, and the relative degrees of tint and chroma, remain unchanged. Lastly, the fact that the spectral hues possess a high FlG. 3. Demonstrational Colour Pyra- mid.— American Journal of Psychology, § 15. Visual Stimulus and Visual Sensation 65 chroma, while the colours that we see in nature and art are less saturated, is due to the relative simplicity of wave- form in the one case and its relative complexity in the other. These are, however, only general correspondences. In detail, the relation between visual stimulus and visual sensation is very far from simple.
There is, indeed, one case — the case of the sensation black — in which there seems, at first thought, to be no relation at all. It is often said, paradoxically, that the stimulus to black is the absence of physical light. If this were true, black would stand apart from all the other visual qualities. But, on the one hand, what we see in the absence of light is not black. ' In the night all cats are grey '; and the absence of light means, in point of fact, that our surroundings appear in a neutral grey. On the other hand, we can see black only in a good light; so that the sensation of black, like the after-images and contrast-sensations of which we speak in § 18, is indi- rectly, although not directly, dependent upon the presence of physical light. We discuss the question of its origin Black is, of course, a positive sensation; as it is not due to the absence of light, so it does not represent the absence of visual sensation. To suppose that the blind live in darkness is altogether wrong, because it is to suppose that they are able to see. Blind persons see things, as Helmholtz puts it,1 in the same way in which we ourselves see what is behind our backs: that is to say, they do not see them at all.
§ 16. The Dependence of Visual Sensation upon Wave- length and Energy of Light. — In general, as we have 66 The Quality of Sensation: Vision said, change in the wave-length of physical light means change in the hue of colour sensation. This statement must now be qualified in three ways.
First, we find by reference to the spectrum that the correspondence of wave-length and hue is not uniform throughout the colour series. At the two ends, in R and V, we may travel for a considerable distance without noticing any marked change of hue. If, on the other hand, we look at the regions on either side of Y^ or at the BG region, we see at once that a very large number of hues is there packed into a very small space. Secondly, we find that difference of wave-length brings with it a differ- ence of chroma. The most saturated colours of the spectrum are R and B; the least saturated are Y and BG. Chroma, therefore, depends upon wave-length as well as upon wave-form. Thirdly, we find that change of wave- length brings with it a change of tint. We have already remarked that Fis the lightest and V the darkest colour of the spectrum. Now the energy of the light-waves de- creases continuously from the long-wave to the short-wave end; 1 so that, if tint depended solely on energy or wave- amplitude, R and not Y should be the lightest colour, and B, as compared with R, should be much darker than is actually the case.
In the same way, change in the energy of physical light may change, not only the tint, but also the hue and the chroma of colour sensations. If the energy of the light- waves which form the spectrum is greatlyhicreased, the 1 This statement holds of the dispersion spectrum of an artificial light, such as is usually observed in psychological laboratories. In the diffraction spectrum of sunlight — what physicists term the 'normal' spectrum — the maximum of energy is in the blue.
d § 1 6. Effect of Wave-length and Energy of Light 6y hues are reduced to two, Fand B, both of light tint and low chroma. The R, O and YG of the original series be- come lighter and yellower; the BG and V become lighter and bluer; G becomes simply lighter, and finally passes into a light grey. If the energy is greatly decreased, the hues are reduced to three, R, G and B V, all of dark tint and low chroma. At the same time there is a marked change in the relative distribution of tint throughout the spectrum: all the colours are darker, but the position of the relatively lightest tint moves from Y to G, — so that the long-wave end is darkened, and the short-wave end lightened, in com- parison with the ordinary spectrum. This shift of tint with decrease of energy is called, after the Austrian physiologist who first observed it, the Purkinje phenomenon.1 The Purkinje phenomenon appears most clearly when the energy of the spectrum has been so far diminished that the colours fade out altogether, and only a series of sensations of light is left. It appears, in any case, only if the energy of light is decreased over the whole field of vision; it is not enough to reduce the energy of the spectrum alone. You may get an idea of it by laying pieces of red and blue paper side by side, and observing them first in ordinary diffuse daylight and then through a pin-hole in a card. Or if you watch the reds and blues of a carpet, as twilight comes on, you will notice that the reds quickly become dark grey or black, while the blues change into a silvery blue-grey. This aspect of the phenomenon, in which it depends upon the general illumination of our surroundings, is dealt with in § 17. The Dependence of Visual Sensation upon Compo- sition of Light. — In tracing out the dependence of visual sensation upon the form or composition of light-waves, we 1 J. Purkinje, Beobachtungen und Versuche ztir Physiologie der Sinne, ii., 68 The Quality of Sensation: Vision may most conveniently employ the colour-mixer, with com- pound discs of the sort mentioned in § 6. Everyone knows that a glowing match, whirled round in the air, is seen as a fiery circle: the effect produced upon the eye by the moving stimulus persists for a time, in what is termed the positive after-image, until the stimulus comes back again to the same point. This is the principle which underlies the use of the rotating discs. It is, further, a law of physiological optics that if such particoloured discs are rotated at a rate of speed sufficient to prevent flicker, the resulting impression upon the eye is the same as it would be if the physical light, reflected from the several sectors, were spread uniformly, layer on layer, over the whole disc-surface.1 From this point of view, then, the use of the discs is as safe as it is convenient. Whether we may argue from the results obtained with the discs to the results obtainable with other forms of stimulus, — whether, more especially, we may argue from the disc- colours to the pure lights of the spectrum, — is a question which experiment must decide. We return to it presently. The facts to be considered in this section may be grouped under three laws, which are known as the laws of colour mixture. The first law states (i) that for every colour there may be found another, antagonistic or com- plementary colour, which if mixed with it in the right pro- portions produces a sensation of light, and if mixed in any other proportions a colour sensation, of low degree of chroma and of the hue of the stronger component. An- 1 This law, called Talbot's law, was first propounded in 1834 by the physi- cist \V II. F. Talbot, one of the inventors of photography ( The London and Edinburgh Philosophical Magazine and Journal of Science, series 3, v.
§ ij. The Laws of Colour Mixture 69 tagonistic colours, in this sense, are carmine, and bluish green, red and verdigris, orange and greenish blue, yellow and blue, yellowish green and violet, green and purple.
The second law states (2) that the mixture of any two colours which are not antagonistic produces a colour sen- sation of intermediate hue; this hue varies with the rela- tive amounts of the two component colours, and the chroma varies with their nearness or remoteness in the colour series. Thus, a mixture of R and B will give violet, purple or carmine, according to the amounts taken; and the K' Fig. 4. Demonstrational Colour Mixer, for six sets of discs.
mixture of R and Y will give an O of relatively low, that oiRO and OYan O of relatively high chroma.
These two laws are, evidently, laws of the same order; they sum up observations of the same general kind. The third law takes us a step farther and shows the effect of mixing, not single colours, but colour mixtures. It is sometimes termed Newton's law of colour mixture, since it is implicit in his theory of the composition of light.1 We may formulate it as follows: (3) if two colour mix- tures arouse the same sensation of light or colour, then a mixture of these mixtures will also arouse that sensation. If, for instance, the grey produced by a mixture of car- 1 1. Newton, Opticks: or, A Treatise of the Reflections, Refractions, Inflec- tions and Colours of Li%ht, [1704] bk. i., pt. ii., props. 4-6.
70 The Quality of Sensation: Vision