LECTURES ON THE ELEMENTARY PSYCHOLOGY OF FEELING AND ATTENTION- BY EDWARD BRADFORD TITCHENER Netn gorfe THE MACMILLAN COMPANY 1908 All rights reserved 3787? Copyright, 1908, bt the macmillan company. Set up and electrotyped. Published July, 1908. Norfajoob i^rcgg J. 8. Cushing Co. — Berwick & Smith Co. Norwood, Mass., U.S.A. T53 EDMUND CLARK SANFORD PREFACE The eight lectures which make up this little book were read during my tenure of a non- resident lectureship in psychology at Columbia University, February, 1908. I have printed them as they were written for delivery, except that quotations from the French and German have, for accuracy's sake, been restored from English translation to their original form/ I have not been able, either in the lectures themselves or in the appended notes, to take account of all that is important in the current psychology of feeling and attention. Indeed, my sins of omission are obvious. I can only say that they weigh heavily upon my scientific conscience, and that, were it not for other and imperative claims upon my time, I should have delayed publication until I had done what I could to correct them. My thanks are due to my wife; to my col- league, Professor I. M. Bentley, who has read * Professor Pillsbury's English work on Attention reached me too late for reference in the text, though I have cited it in the notes. vii viii PREFACE the manuscript of the book and during its prep- aration gave me unsparingly of his time and counsel; to Professor J. McK. Cattell, of Co- lumbia University, whose invitation prompted the writing of the lectures ; and to many kindly critics among my hearers. I have dedicated the volume to Professor E. C. Sanford, of Clark University, my close friend and trusted mentor of the past sixteen years. Would that it were worthier of his acceptance ! Cornell Heights, Ithaca, N.Y., March, 1908. "> CONTENTS LECTtTEB I. Sensation and its Attributes . PAGE 3 II. Sensation and Affection : the Criteria of Affection 33 III. The Affections as Gefiihlsem'pfindungen 81 IV. The Tridimensional Theory of Feeling . 125 V. Attention as Sensory Clearness . . 171 VI. The Laws of Attention : I . 209 VII. The Laws of Attention : II 251 VIII. Affection and Attention 285 Notes to Lecture I 321 Notes to Lecture II 328 Notes to Lecture III 338 Notes to Lecture IV 345 Notes to Lecture V 352 Notes to Lecture VI 360 Notes to Lecture VII 374 Notes to Lecture VIII 385 Index of Names . 393 Index of Subjects 397 IX I SENSATION AND ITS ATTRIBUTES LECTURE I SENSATION AND ITS ATTRIBUTES THE system' of psychology rests upon a three- fold foundation: the doctrine of sensation and image/ the elementary doctrine of feeling, and the doctrine of attention. Our views of sen- sation, of feeling, and of attention determine, if we are logical, the whole further course of our psychological thought and exposition. Where systems differ by anything more than relative emphasis and fulness of treatment, their differ- ences invariably lead us back to the consideration of these fundamental doctrines. It is, therefore, more than important — it is necessary — that the student of psychology have a firm grasp of the issues involved and a comprehensive knowledge of the relevant facts. These requirements are, however, by no means easy of fulfilment. Look, first of all, at sen- sation. We know a great deal about sensation itself, as an elementary process ; we know a great deal about the simpler syntheses ; and we have working theories in most of the sense-depart- ments. On all these points we owe a debt, 3 4 SENSATION AND ITS ATTRIBUTES which we must gratefully acknowledge, to physi- ological interest and physiological equipment. Methods and results, together with apparatus that embodied methods and assured results, were at our disposal as soon as we had the skill to use and the funds to acquire. We have bor- rowed freely from physiology, and we have turned the loan to such good account that physiology is not ashamed, on occasion, to borrow again from us. Nevertheless, with all the advantage that comes of an experimental tradition, and with all the facilities for work afforded by the local dif- ferentiation of the sense-organs, we are still far removed, in this sphere of sensation, from finality or general agreement. A mental element can be defined only by the enumeration of its attri- butes. Turn, now, to the table of contents of the Physiologische Psychologies and you find but two attributes of sensation : intensity and quality. Turn to Ebbinghaus' Grundziige, and you find that sensations have both special and general attributes, and that the latter include such ap- parently heterogeneous things as extension and duration, movement and change, likeness and difference, unity and multiplicity.^ But if there is difference of opinion as re- gards sensation, what shall we say of feeling and attention ? The unsettled state of the psy- INTRODUCTION 5 chology of feeling is notorious. Here are prob- lems on which, as it would almost seem, the trained and the untrained, the professional and the amateur psychologist exchange ideas on equal terms ; here is a field in which one man's casual opinion is as valuable as another man's reasoned conclusion, in which a general impression is worth as much as an experimental result. And, what is worse, the path of inference is so precarious, and the experimental results are as yet so few, that psychologists von Fach are them- selves tempted to overhasty generalisation, and become dogmatic before criticism has done its work. Does not Wundt base the psychology of language on his theory of affective pluralism ? ^ Nor is attention in much better case : the first sentence of the preface to Pillsbury's recent book refers to the ' chaotic state of current theories of attention.' ^ It is, perhaps, true that the prob- lems of attention are less widely discussed, have attracted less general notice, than the problems of feeling. If, however, this is the fact, the prob- lems are none the less insistent ; and their neg- lect by the educated public means simply that popular psychology long ago worked out a theory of attention for its own use, and so far has not felt the need of reconsideration. It follows, plainly enough, that I cannot in 6 SENSATION AND ITS ATTRIBUTES these Lectures give you any complete or finished account of the psychology of affection and attention ; if time allowed, the nature of the case would forbid. It follows also that my account, such as it is, will of necessity take on an individ- ual colouring. It would be absurd to make the claim of impartiality when all one's efforts, whether of criticism or of construction, are de- termined by training and temperament. Be- sides, the attitude of impartiality is irrelevant, so long as every set of observations is coupled with the name of the observer, and every ob- server has his private interpretation. I shall, however, keep as closely as possible to docu- ments and to experimental results; and where I venture a personal opinion, I shall offer it as an opinion and as nothing more. So much may be said by way of general in- troduction. But now, before we come to close quarters with affection and attention, we must give a little time to sensation. This special introduction is necessary for the reason that, throughout the following discussions, sensa- tion will be our standard of reference. When we ask whether the affective processes show dis- tinctive features, w^e are in search of features that distinguish affection from sensation; when we speak of the laws of attention, w^e have always PSYCHOLOGY AND PS YCHOPH YSICS 7 in mind a distribution or redistribution of the sense-processes that make up the consciousness of the moment. Hence it is important that we understand clearly what sensation is : or at any rate, that we frame a working definition of sensation, adequate to our present purpose and free of ambiguity. It will help to clear the ground if we distin- guish, at the outset, between the sensation or sensory element of psychology and the sensory element of psychophysics. The sensation of psychology is any sense-process that cannot be further analysed by introspection : every one of the forty thousand lights and colours that we can see, every one of the eleven thousand tones that we can hear, is a psychological sen- sation. The sensations of psychophysics, on the other hand, are the sense-correlates of the elementary excitatory processes posited by a theory of vision or audition or what not. Thus the six Urfarben of Hering's theory of vision — black and white, blue and yellow, carmine and bluish green — are, if that theory be accepted, the psychophysical elements of vision ; they are the sources of the whole series of psycho- logical elements. These latter are, none the less, psychologically elementary : a light unsaturated 8 SENSATION AND ITS ATTRIBUTES yellowish green, while psychophysically com- pound, is introspectively simple; it cannot be factored into a white, a yellow, and a green, as a chord may be factored into a number of simple tones. Similarly, if we could accept Macli's notion of dull and bright components in tonal sensation,^ we should have only two psychophysical elements in audition; whereas the Helmholtz theory gives us parallel series of psychophysical and of psychological sensa- tions. Here, as elsewhere in experimental psy- chology, the failure to distinguish between psychophysics and psychology proper has led to much confused argument.® We are now concerned with the sensory element of psychology. And a mental element, as was said above, must be defined by an enu- meration of its attributes. What, then, are the attributes of sensation ? An attribute of sensation, as commonly de- fined, is any aspect or moment or dimension of sensation which fulfils the two conditions of inseparability and independent variability. The attributes of any sensation are always given when the sensation itself is given, and the anni- hilation of any attribute carries with it the annihilation, the disappearance, of the sensa- tion itself ; this is what is meant by the * insepara- THE DEFINITION OF ATTRIBUTE 9 bility' of the attribute. A sensation that has no quality, no intensity, no duration, etc., is not a sensation; it is nothing. Conversely, if a sensation is to exist, it must come into being with all of its attributes; we cannot have an intensive sensation that is dispossessed of qual- ity. These statements are evidently true, and so far the definition cannot be questioned. But we are told, in the second place, that the attri- butes of sensation are independently variable; quality may be changed while intensity remains constant, intensity changed while quality re- mains constant, and so on throughout the list. Is this statement true ? Relatively, yes : true for certain cases and under certain conditions. If it were not true, — true, within limits, for the attributes of intensity and quality, as originally recognised, — how could it have been made ? what could have suggested it ? It is matter of observation that the intensity of tone or noise may be varied while the quality is the same, that warm and cold may change in degree with- out change in kind. Absolutely true, however, the statement is not. In certain cases and be- yond certain limits the variation of one attribute implies the concomitant variation of another; and in extreme instances the separation of the two can be effected, if at all, only by a sort of 10 SENSATION AND ITS ATTRIBUTES analogical abstraction, — by neglect, we will say, of quality, and by direction of attention to what, in the light of previous experience, we con- ceive to be intensity. I come to the concrete in a moment. What I now wish to emphasise is the fact that there are bound attributes as w^ell as free, and that the test of independent •variability, useful enough for a preliminary survey, must be applied with caution when we demand accuracy of detail. Having thus amended the definition of 'attri- bute,' we might proceed at once to enumerate the distinguishable attributes of the various classes of sensation. The result would be a list, longer or shorter according to the sense- department, in which term followed term in conventional order, — an empirical list, in which every term stood apart from every other, and all terms were on the same level. I think that we shall do better to cast about for some prin- ciple of classification ; and I have seemed to find such a principle in Mliller's distinction of intensive and qualitative change. A sensation changes intensively, Mtiller says, when it moves along the shortest path to or from the zero- point ; i+ changes qualitatively when it moves in a diivction that neither carries it towards nor withdraws it from the zero-point.^ If we gen- QUALITATIVE ATTRIBUTES: VISION 11 eralise these statements, we may group all the attributes of sensation under the two headings, qualitative and intensive. I should, for instance, rank as intensive attributes, in the broad sense, de gree or intensity proper, duration, ex tension, and clearness. Duration varies between a limi- «■ . — — nal value and the maximum set by adaptation or fatigue; extension varies between a liminal value and the maximum set by the boundaries of the field of sense ; and clearness, too, varies between a liminal value and the maximum set by the limit of attention al concentration. On the other hand, I regard what is ordinarily termed the quality of sensation as, in several cases, a complex of distinguishable qualitative attributes. We will take the qualitative attributes first, and we will begin with vision. Visual sensa- tions fall, for psychology, into the two great classes of sensations of light and sensations of colour; the whole system finds representation in the double pyramid, which is itself a purely psychological construction. The sensations of light need not detain us. The sensations of colour, however, are interesting in that they have no less than three qualitative attribute- A given colour may be varied in hue or c/rved 12 SENSATION AND ITS ATTRIBUTES tone, in tint or brightness, and in chroma or saturation; and reference to the colour pyra- mid will show that, within limits, these three attributes, hue, tint, and chroma, are independent variables, — so that we may change hue while G tint and chroma remain the same, change tint with- out changing hue and chroma, and change chroma with constancy of hue and tint. The limits are set, of course, by the form of the double pyramid, which, as I have said, is an empirical, psycho- logical construction. But here are three dis- tinguishable attributes under the currently single heading of quality. When we turn to audition, we are on more debatable ground. I myself believe that tonal pensations show a qualitative duality, — that a ^ quality of tone is a resultant of the two withbutes known respectively as pitch and as BK Fig. 1. The Colour Pyramid. — H. Eb- binghaus, Grundziige der Psychologic, i., 1905, 199. QUALITATIVE ATTRIBUTES: AUDITION 13 voluminousness (Stumpf's Tongrosse). I dare say that, at first thought, it seems far-fetched, even a little ridiculous, to make volume a quali- tative attribute, especially in view of the uses to which it has been put in systematic psychology. But I would remind you, in the first place, that we are inveterately addicted to spatial meta- phors, and that the term * pitch' contains a spa- tial reference no less obvious, on consideration, than that of * volume.' Pitch means height, elevation; the German equivalent is Tonhohe, the French hauteur; and in characterising pitch, we speak, in English, of high, low, deep tones. Yet nobody nowadays would dream of making pitch an intensive attribute. Now- adays, no! — but listen to Fechner. "Bei den Tonen," he says, *'hat die Hohe, obwohl als Qualitat des Tones fassbar, doch auch eine quantitative Seite, sofern wir eine grossere und geringere Hohe unterscheiden konnen." ^ There the spatial metaphor was at work. But if this spatial reference is to be ignored in the case of pitch, why should we pay regard to it in the case of volume ? May it not be the fact, simply, that the idea of tonal voluminousness is less familiar to us than that of tonal pitch, that we have observed the attribute of volume less fre- quently or less accurately than we have observed 14 SENSATION AND ITS ATTRIBUTES height and depth, — and so that we are misled by the name ? Secondly, I would remind you that, if we are to turn the attribute of volume to account for a theory of space-perception, we must be extremely careful to take it as what it intro- spectively is, an attribute of tonal sensation, and not to surround it with visual or tactual asso- ciates. I may illustrate by a quotation from James, though James is not dealing primarily with tones. **Loud sounds," he says, *'have a certain enormousness of feeling. It is im- possible to conceive of the explosion of a cannon as filling a small space. In general, sounds seem to occupy all the room between us and their source; and in the case of certain ones, the cricket's song, the whistling of the wind, the roaring of the surf, or of a distant railway train, to have no definite starting-point." ^ These statements are offered as evidence of the general principle that a spatial attribute, extensity, volu- minousness, vastness, is inherent in all sensa- tions without exception. But the sensation, as elementary process, knows and says nothing whatever of its stimulus or its organ or its object. An explosive noise, considered as sensation, is not the noise of a cannon or of anything else; a continuative noise, hiss or whistle or roar, considered as sensation, has nothing to do with QUALITATIVE ATTRIBUTES: AUDITION 15 a starting-point in objective space, definite or indefinite/^ The evidence must be sought else- where, — sought in sensation proper, under rigid introspective conditions, and sought, more es- pecially, under conditions that rule out the com- plicating attribute of intensity, — for *loud' sounds may be enormous in one way, and w^eak sounds in quite a different w^ay. Make, then, the experiment for yourselves. Take a series of tuning-forks, standing on their resonance boxes, — a series that extends from bass to treble, — and listen to their tones. There can be no manner of doubt that volume is an attribute of tonal sensation. There may, however, I think, be a very considerable doubt whether the volume is in any real sense spatial. Choose your adjectives. The deep tones are bigger, larger, not more massive, perhaps, but more diffuse ; the high tones are smaller, thin- ner, sharper. The spatial reference lies very near. Still, when you say more diffuse, thin- ner, sharper, you refer to more than space- form or space-extension ; there is a hint in the words of a difference of texture. Try now the terms milder, softer, for the deep tones, and shriller, harder, for the high; do they not fit the facts ? Surely, volume is not an in- tensive attribute, a mere bulkiness that ran2;es IG SENSATION AND ITS ATTRIBUTES between the extremes of pin-point concentra- tion and all-pervading vastness, but a qualita- tive attribute, moving between the extremes of mild and shrill. Volume and pitch are to the tonal sensation what hue and tint and chroma are to the sensation of colour; and the attri- butes are independently variable, in the sense that at the two ends of the scale volume changes more quickly than pitch, while over the middle region it changes more slowly/^ Little can be said, at present, of the sensations of noise. Both the explosive noise (the pop of a soap-bubble, the sharp drop of a wooden block upon a wooden table) and the continua- tive noise (the hiss of escaping steam), if heard singly, appear simple to introspection. If, how- ever, we make the observation serial, we can distinguish an attribute of pitch and a concomi- tant noisiness. The question then arises whether pitch is a constituent of noise quality, or whether it is due to the admixture of tone. Noisiness itself seems to remain constant over fairly wide regions of the scale of pitch ; but nothing more definite can be said about it. The qualitative attribute presents no difficulty in the spheres of taste, smell, and temperature. It is otherwise with cutaneous pressure, cutane- ous pain, and many of the organic sensations. QUALITATIVE ATTRIBUTES: PRESSURE 17 Suppose that a well-defined and responsive pressure spot is stimulated with increasing de- grees of intensity. We get at first, with the weakest stimulus, a sensation of tickle. At moderate stimulation, this passes over into pressure; either a quivering, wavery pressure, or a hard, 'cylindrical' pressure. If the in- tensity of stimulus is still further increased, but not carried to the point at which subcutaneous tissue becomes involved, we have the Gold- scheider sensation of 'granular' pressure. I am not now concerned with psychophysical ques- tions, but with psychological ; and the peculiar- ity of these observations, from the psychological side, is that the qualities just mentioned some- times overlap. I have not noticed, it is true, any overlapping of the granular by the cylin- drical pressure. But the ticklishness of weak stimulation is often sensed along with the differ- ent quality of quivering pressure, and this again may at times be sensed alongside of the Gold- scheider granular pressure. Suppose, again, that a pain spot is similarly stimulated. We get at first, with the weakest stimulus, a sensation of itch. At moderate stimulation, this passes over into prick or sting; and with further increase of the intensity of stimulus, into cuta- neous pain. And here, as before, there is over- 18 SENSATION AND ITS ATTRIBUTES lapping. A sting may be an itchy sting, and a pain may be a stinging pain. The same thing holds, apparently, of certain kinsesthetic sensations. The dragging, tired sen- sation which is probably attributable to the muscle-spindles passes through a sore, achy stage into dull pain : the three stages are intro- spectively distinguishable; but there is, never- theless, an overlapping. The strain sensation which seems to be due to stimulation of the Golgi spindles in tendon also passes into dull pain by gradual transition and overlapping of quality. Finally, I am inclined to think — though I say this with greater reserve — that the same thing holds in the case of alimentary sensations. Isolate hunger and nausea, at fairly low inten- sities, and you have a dull pressure. The same dull pressure ? It would be overhasty to assert a precise identity ; but, at any rate, the likeness revealed by analysis is surprising when we re- member the gross difference between the hungry and the nauseated consciousness. It looks as if, with increase of intensity of stimulus, a second qualitative factor — possibly a group of quali- tative factors — comes into play in the two cases, differentiating sensations which, at the beginning of the intensive scale, are so nearly alike as to run the risk of identification.^^ THE INTENSIVE ATTRIBUTES 10 At this point you may very well object that I am confusing two distinct things : the fusion of qualitatively different sensations, and the confluence of different qualitative attributes in one and the same sensation. Tiredness, you may say, does not pass over into pain, but is coloured by, fused with, a pain sensation ; the hungry and the nauseated consciousnesses are formations of great complexity, and imply the fusion of a large number of qualitatively differ- ent sensations. That may be true. On the other hand, I think that psychology has taken the simplicity of the qualitative attribute in too dogmatic a spirit. There can be no doubt that the sensation of colour is qualitatively compound ; there can be no doubt, I believe, about the ob- servations just described in the spheres of pres- sure and cutaneous pain. And there is no reason a "priori why the organic sensations should fol- low the type of taste and smell rather than that of touch. Even, then, if you do not accept the conclusions that I have suggested, you will perhaps be ready to admit that there is a great deal of work still to be done before we can make out a final list of the sense-qualities. We may now go on to consider the intensive attributes of sensation, and we may start out 20 SENSATION AND ITS ATTRIBUTES with intensity proper. Intensity has been so exhaustively discussed, in connection with the methods of quantitative psychology, that it is not necessary to enter into details. Let me re- mark, how^ever, that the attribute is, in practice, much less free than it is sometimes made in theory. I have already given instances from the cutaneous senses and possible instances from the kinsesthetic. But even in the case of sounds, those who have worked with the Fechner pen- dulum or with the gravity phonometer know that, beyond narrow limits, independent varia- tion of intensity is exceedingly difficult. The classical difficulty arises in the sphere of vision. Hering long ago denied the attribute of intensity "im iiblichen Sinne des Wortes" to the sensations of the black- white series. Hillebrand, in 1889, declares that intensive differences do not appear anywhere in the do- main of visual sensation, though there may be a constant intensity that is never noticed and therefore cannot be empirically demonstrated. Kiilpe, in 1893, writes that *' intensity cannot be ascribed to sensations of sight." Hering repeats, in 1907, that the "Begriff der Inten- sitat auf die Farbe nicht anwendbar ist." Per- sonally, I have never been able to subscribe to this doctrine. It is true, as Miiller says, that INTENSIVE ATTRIBUTES: VISION 21 "die Empfindung einer und derselben Grau- nuance kommt in der That in unserer Erfah- rung nicht mit merkbar verschiedenen Intensi- taten vor," and that "auch eine Farbenempfin- dung von ganz bestimmter Qualitat konnen wir . . . nicht in verschiedenen Intensitaten herstellen." That is matter of observable fact. But it is surely true, on the other hand, that we recognise degrees of intensity in visual sensa- tion, and that — by the process of analogical abstraction of which I spoke earlier in this Lecture — we are able in some measure to ig- nore the concomitant change of quality and to direct our attention to intensity alone. Mliller, in the paper just quoted from, has rescued the intensity of visual sensation, on the psycho- physical side, by his theory of central gray ; and Kulpe has now accepted, if not that theory itself, at any rate the attribute whose behaviour it is meant to explain. The theory, in brief sum- mary, is this : that w^e owe the intensive pecul- iarity of visual sensation to the dual character, peripheral and central, of the nervous processes involved. The retinal processes are antagonistic : two coincident stimuli — black and white, for instance — are effective for excitation only by their difference, by excess of the one over the other. The endogenous, central excitation is 22 SENSATION AXD ITS ATTRIBUTES constant. Hence a peripheral stimulation may result in the whitening, lightening, or in the blackening, darkening of the central gray, but there is no way of intensifying that gray without changing its quality, — no way of strengthening its black and white components at the same time and in the same degree/^ I have digressed thus briefly into psycho- physics, because it is precisely in such cases that that much-abused science shows to its best advantage. Introspection is at fault. Some psychologists will have it that the scale of tints, the black-white series, is a scale of intensities, and will hear nothing of quality; others affirm that it is a scale of qualities, and will hear noth- ing of intensity ; others, again, declare that it is at once qualitative and intensive. Psycho- physics not only resolves the difficulty, but shows why the difficulty was there. Here we may leave intensity, and pass to the consideration of the spatial and temporal attri- butes, extension and duration. This, as you know, is controversial ground. I cannot help thinking, however, that the psychology of these attributes is simpler than it is ordinarily repre- sented to be. We must, of course, distinguish : we must not identify physical with mental time, or physical with mental space; we must not EXTENSION AND DURATION 23 confuse processes that in some way mean time, or mean space, with attributes that in some way are time and are space; we must not run to- gether time-estimate and durational experience, or space-estimate and extensional experience. Granted ! But the attributes themselves are surely obvious enough. What is psychological extension ? It is the aspect of sensation that we attend to when we are called upon to answer the questions (perhaps with reference to an after- image, perhaps with reference to a cutaneous sensation) : How large is it ? What shape has it ? Is it regular or irregular ? large or small ? continuous or patchy ? uniform or broken ? And in the same w^ay, psychological duration is the attribute that we attend to w^hen we answer the questions : How long does it last ? When does it disappear ? Has it gone out yet ? Is it steady or interrupted.^ — That is all. The attributes of sensation are always simultaneously present, — evidently ! since the nullifying of any attxibute annihilates the sensation. But when we are thus attending to extension or duration we may have very hazy ideas indeed about in- tensity and quality; precisely as, when we are observing intensity, we may have very hazy ideas about quality and duration. The ques- tion what extension and duration are, in direct 24 SENSATION AND ITS ATTRIBUTES experience, is a nonsensical question ; we can only reply, tautologically, that duration is a going-on, and extension a spreading-out. But what, then, are quality and intensity 'in direct experience ' ? Does it help to say that quality is the individualising attribute ? That is only saying that quality is quality. Does it help to say that intensity is always a more or a less ? What, then, of clearness, or duration, or exten- sion ? Or that high intensities make a greater claim upon us, dominate consciousness more exclusively, than low ? What, then, of clearness or of the Eindringlichkeit that we are to discuss presently ? You cannot define the indefinable : at most you get a formal equivalent — ' simplest spatial determination ' or the like — that serves you as a paragraph heading. As to the difficulty that duration and extension must find expression in physical units, and that we have no right to equate the psychical and the physical, that is a difficulty which occurs also in the case of inten- sity, where it has been successfully met. To work over the whole ground again, with simple change of terms, is purely gratuitous labour.^^ It is more to the point to inquire into the em- pirical distribution of the two attributes. Dura- tion appears to attach to all sensations. Exten- sion attaches, without any doubt, to all visual EXTENSION AND DURATION 25 sensations. It is also ascribed, in common parlance, to the ' sense of touch.' Touch, how- ever, is an extremely ambiguous term; it may refer to cutaneous pressure, while it may cover all the cutaneous and many of the organic senses. If, now, you ask me which of these component senses has the extensional attribute, I must con- fess that decision is, in some instances, very difficult, and that my own opin- ion has differed at different times. Just now, I am inclined to -n^ ^ ^ u f ^r i a ' Fig. 2. Schema of a Visual Sensation. be liberal. I should 'The four vertical lines represent the - four intensive attributes : intensity, give the attribute to clearness, extension, duration. The allfoUroftheCUtane- |f'' horizontal Unes represent the three qualitative attributes : hue, tint, ous senses, — pres- chroma. sure, warmth, cold, and pain ; I should give it to the organic pains; and I should give it also to the organic sensations, kinsesthetic or other, whose quality suggests the term * pressure.' It seems to me that in all these sensations we get a true extension, different from the quasi-exten- sity of tones. Let me repeat, however, that de- cision is difficult ; I have no wish to be dogmatic. ^^ 26 SENSATION AND ITS ATTRIBUTES Clearness, our fourth intensive attribute, is no more definable than its fellows. It is the attribute which distinguishes the * focal' from the 'marginal' sensation; it is the attribute whose variation reflects the 'distribution of attention.' ^^ We may postpone its discussion until we come to deal with the subject of Attention. I must also touch, however briefly, upon the appearance of attributes of a higher order. The best illustration of what is meant by the phrase is afforded, perhaps, by tone-colour or tone- tint, — a certain colouring or timbre which at- taches to simple tones, and which may, but need not, be analysed. We owe the recognition of this compound attribute to Stumpf, w^ho de- rives it from pitch, intensity (high tones are in- trinsically louder than low), and volume. It finds expression in such antitheses as bright and dull, sharp and flat, full and hollow. Other instances are the penetratingness of certain scents, — camphor and naphthaline, e.g., as compared with vanilla and orris-root, — the urgency or importunity of certain pains or of the taste of bitter, the obtrusiveness or self- insistence of certain lights and colours and tones. All these latter attributes involve clearness, in THE CRITERION OF SENSATION 27 conjunction with quality, or with intensity, or with intensity and quality together. Their investigation in detail cannot but prove fruitful, whether for psychology or for psychophysics/^ I spoke, earlier in this Lecture, of the forty thousand lights and colours that we can see, and the eleven thousand tones that we can hear. The 'forty thousand' was a rough guess at the number of discriminable qualities included in the colour pyramid ; a modest guess, too, when you compare it with Ebbinghaus' **many hundred thousand," or Aubert's "many million"! But I prefer underestimation to overestimation ; and I think that Ebbinghaus would find it difficult to bring convincing evidence even of a single hundred thousand visual qualities. On the other hand, the eleven thousand tones are dis- tinguished on the basis of pitch alone ; and that number must be increased if investigation proves — what is a ^priori extremely probable — that pitch may remain the same while the qualita- tive attribute of volume undergoes noticeable change.^ ^ Let us, however, raise in conclusion a more general question. Why do we identify the num- ber of sensations furnished by a particular sense-department with the number of distin- 28 SENSATION AND ITS ATTRIBUTES guishable sensory qualities? Why is quality the 'individualising' attribute? Why are not the different intensities of a given pitch ' different sensations ' ? I suppose that, in strict logic, any noticeable change in an attribute of sensation gives us a 'different' sensation. As soon as ever intro- spection turned to the attribute of intensity, it found differences, not of simple more and less, but of 'kind.' Lotze, for instance, declares that a strong sour does not taste the same as a weak; there are "qualitative Veranderungen des Empfindungsinhalts, die von jenen [inten- siven] Differenzen des Reizes abhangen." Now, to call intensive change a change of 'quality' is to introduce unnecessary confusion of terms. We need not do that; but we need not either, it seems to me, quarrel with those who hold that sensations of the same quality but of differ- ent intensity are, psychologically regarded, dif- ferent sensations. The innovation would not lengthen our list of visual sensations ; it would, very considerably, lengthen the list of auditory sensations. And what of clearness, duration, extension ? Are we, in their case, dealing again with differ- ences of 'kind,' or merely with differences of degree ? It is really impossible to say ; the intro- THE CRITERION OF SENSATION 29 spective judgments are lacking. From general impression, I incline to the view that differ- ences of clearness are, like intensive differences, ultimate and distinctive. On the score of dura- tion and extension I do not like even to hazard a conjecture; though, if I were compelled to take sides, I should fall back on the analogy of intensity. If, therefore, there is anything to be gained by substituting ' attributive difference ' for ' quali- tative difference' as a criterion of sensation, I shall be willing to make the change. As things are, I do not see the gain; and I do not see, either, the necessity of logical strictness. Our classification of sensations is a matter of utility, of expediency; the question involved is general, but it is not scientifically important. In science, as in ordinary life, we call things different when their difference is striking and outweighs their likeness, and we call things like when their like- ness is striking and outweighs their difference. Red and blue, sour and sweet, are in this sense * different'; loud and soft, light and heavy, are *like.' Until it is shown that the new and more elaborate classification brings positive advan- tage to descriptive psychology, I shall accord- ingly rest content with the traditional list of sensible qualities. ^^ 30 SENSATION AND ITS ATTRIBUTES So this hasty review comes to an end. I do not apologise for its imperfection, its sketchi- ness; for sensation is not our primary subject, and at the best one cannot say very much about sensation in a single hour. I have tried only to raise such points and to discuss such issues as will put you in tune with me, so to say, for our later study of affection and attention. When I speak of sensation, in the following Lectures, I shall mean by it the kind of process that we have been considering to-day ; the fringe of asso- ciation with which the word is surrounded will be drawn from the circle of ideas within which we have now been moving. i II SENSATION AND AFFECTION: THE CRITERIA OF AFFECTION LECTURE II SENSATION AND AFFECTION: THE CRITERIA OF AFFECTION THE psychology of feeling, as I said in the introduction to the preceding Lecture, is in a notoriously unsettled state. We have psycholo- gists of the first rank who posit an elementary affective process alongside of sensation ; we have psychologists of the first rank who deny the dis- tinction. Wundt and Lipps stand over against Brentano and Stumpf .^ I propose, now% in the present hour, to examine the principal arguments that have been urged in favour of an inde- pendent feeling element, and the arguments that have been brought forward in reply. I shall use the term 'afi^ection' for the elementary process in question, and for the sake of clear- ness I shall speak only of the qualities of pleas- antness and unpleasantness. These, of course, are recognised by all psychologists alike, — by those who hold a plural as well as by those who hold a dual theory of affective processes at large. What we may call the gross reason, the obvious reason, for assuming an affective element is, i> 33 34 SEXSATIOX AND AFFECTION I suppose, the gross and obvious difference be- tween the intellectual processes of the adult mind, on the one hand, and the emotive processes, on the other. As thought differs from emotion, so must the element of thought, the sensation, differ from the element of emotion, the affection. Personally, I attach more weight to this argument than its formal expression might seem to warrant. I believe that the simple feelings — our expe- riences when we 'feel hungry,' 'feel dizzy,' *feel tired,' 'feel comfortable,' 'feel poorly,' 'feel first- rate ' — represent a stage or level from which we ascend to the emotions ; and that the emo- tions, again, represent a stage or level from which we descend to secondary feelings : our anger weakens and simplifies to a feeling of irritation, our resentment to a feeling of chagrin or annoy- ance, our joy to a feeling of pleased content- ment, our grief to a feeling of depression. I believe that we are here in presence of a general law^ or uniformity of mental occurrence; that all conscious formations show like phenomena of rise and fall, increase and decrease in complexity, expansion and reduction. Nevertheless, as sys- tematic psychology stands to-day, the argument has no objective validity, no power to carry con- viction. It may be traversed, flatly and finally, in two different ways: by the James-Lange FEELING AND EMOTION 35 theory of emotion, and by the theory of Stumpf. If we accept a strict version of the James-Lange theory, and identify the specifically emotive or affective processes in emotion with organic sensations, then evidently we dispense, at this middle level, with the independent affective element, and the argument from continuity falls to the ground. And if we divorce the sense- feeling from the emotion, in Stumpf's way, and assert that the 'psychological nucleus' of the emotion, the central and characteristic process that makes it what it is, is altogether different from sense-feeling, — that "die Sinnesgeflihle den Gemiitsbewegungen heterogen sind," — then, again, we have a sharp severance of continuity, and the argument lapses. Neither of these alternative views can be lightly brushed aside : the James-Lange theory has aroused prolonged discussion, and has gained many adherents; and the Stumpf theory, in essential points, com- mands the assent, e.g., of Stout and Irons.^ It would be interesting to take representative statements of the three views of emotion — say, the statements of Wundt, James, and Stumpf — and to estimate each one in the light of the other two. I doubt, however, whether the com- parison would be profitable. Surely, if we are to reach anything like a conclusion, we must 36 SENSATION AND AFFECTION begin lower down ; we must go, not to emotion, but to sense-feeling. Is there any one who, when weighing James' theory in the balance, has not heartily wished that he had given us a chapter on the feelings ? Is there any student of the Tonpsychologie and of Stumpf's later w^ork who has not felt the want of that 'Abschnitt tiber die durch Sinneseindrlicke erweckten Gefiihle' which was promised in 1883 and has delayed until 1906 ? ^ Can any one doubt that the issue raised by Wundt's tridimensional theory of affections is, systematically, a more fundamen- tal issue than is involved in the most radical doctrine of emotion ? I may be seeing things crookedly ; but as I see them, the heart of the problem lies in feeling. Let us, then, attack the problem at this point; let us consider, as criti- cally as we may, the alleged criteria of affection. (1) We may take up, first, the statement that sensations are the objective and affections the subjective elements of consciousness; and we will try to give these terms, 'objective' and * subjective,' a tangible psychological meaning. Let us be clear that the meaning must be psychological ; the difference, if it exist, must be a difference that is open to introspective verifica- tion. Anything in the way of epistemological AFFECTION AS SUBJECTIVE 37 argument is wholly out of place. It is out of place for two reasons. On the one hand, psy- chology is an independent discipline, and can no more take dictation from epistemology than it can from metaphysics or ethics. And, on the other, epistemology is concerned with the principles of knowledge — whether with the material and formal principles together, or with the material principles alone, is matter of defini- tion ; while the psychological element has no part or lot in knowledge, has no reference or meaning or object or cognitive contents of any sort. Let us be clear, also, that the meaning which we give to the terms ' objective ' and ' subjective ' must cover a difference in the elementary pro- cesses regarded as elementary. It has been urged, for instance, that the sensory elements in perception are looked upon, in ordinary thought, as properties of external things, whereas feeling is always personal, reflects always a state of the mind itself. Heat seems to reside in the burning coals; but the pleasantness, the grate- fulness, of the warmth is in me. I will not now dwell on the epistemological implications of this argument, but will accept it at its face value, as an argument from the psychology of percep- tion and feeling. And I reply, first, that the 38 SENSATION AND AFFECTION statement which it makes is not true, the dis- tinction which it draws cannot be drawn. For the pleasant or grateful feeling which is subjec- tive, in me, is a feeling and not an affection ; it comprises certain organic sensations; and nobody confuses organic sensations with prop- erties of external things. I reply, secondly, that the argument, even if it were true, would be irrelevant. For it is an argument based, not on introspection of the elementary processes as such, but on the character or behaviour of these processes in combination. We, however, are dealing with the mental elements in their status as elements. There are, I think, three interpretations of the terms 'objective' and 'subjective' that have claims upon our attention, (a) The first is that of Wundt. In a recent study of Wundt's doctrine of psychical analysis, Hollands has made the subjectivity of affective process, in Wundt's system, the topic of detailed study. I need not here attempt any summary of the discussion, since Hollands' articles are easily accessible in The American Journal of Psychology. The upshot of the investigation is that Wundt con- trasts, under the two rubrics, tendency to fusion and persistent discreteness. "Feeling ... is always falling into unitary masses, it forms a AFFECTION AS SUBJECTIVE 39 single continuum. This ... we may take as Wundt's final meaning in psychology for sub- What are we to say in criticism ? This, evi- dently : that while Wundt has, as Hollands main- tains, given the distinction an "introspective definition," he has not derived it from a com- parison of isolated sensation with isolated affec- tion. A 'tendency to fusion' is not an attribute that shows, like intensity or quality, in the single element. Besides, there is also a tendency to fusion in the organic sensations; they, too, are * always falling into unitary masses.' Indeed, if we reject Wundt's theory of the plurality of affective qualities, the criterion becomes meaning- less : the ' unitary masses ' and the ' single con- tinuum' formed with pleasantness-unpleasant- ness by excitement-depression and strain-relaxa- tion take us out of the affective sphere and into that of organic sensation ; the subjectivity that should characterise affection now characterises a group of sensations. Finally, it must be remarked that the doctrine of the Totalgefilhl is not universally accepted. "Es giebt," says Saxinger, "einen grossen Kreis von Thatsachen, welcher Zeugniss flir das Vorkommen coexis- tirender Geflihle ablegt." Here is no fusion, no continuum, but separation and discontinuity. 40 SENSATION AND AFFECTION Let us try another interpretation. We might argue (b) that sensations are objective because they are experienced in the same way by every one, and that affections are subjective because they are experienced differently, individually, by different persons or by the same person at different times. Here, again, however, it would \ be enough to point out in answer that the single elements carry no such distinction upon them. Stumpf has also brought up a factual objection : he reminds us that what is supposed to hold of the affective processes holds very definitely of sensations of temperature. A room that seems overwarm when you come in from the outside air may seem chilly to those who have been sit- ting in it for some time. Stumpf might have generalised this objection, and referred simply to the phenomenon of adaptation. Wherever we have adaptation, there we have the possibility that like stimuli will arouse different sensations in different minds. And if you rejoin that the sensations are, nevertheless, always the same under the same conditions, then I ask : How do you know that this rule does not also apply to affections ? The variability of affective expe- rience may be due, precisely, to difference in affective adaptation. There is still the third possibility. We might AFFECTION AS SUBJECTIVE 41 express, in the terms 'objective' and 'subjec- tive,' the fact (c) that sensations can stand alone in consciousness, independently oi affection, while affection never appears alone, but always and of necessity as the concomitant of some sen- sation. Many psychologists, as we know, have looked upon affection not as an elementary pro- cess, coordinate with sensation, but as an attri- bute of sensation; they speak of Gefilhlston, affective tone, feeling tone, algedonic quality. The hypothesis that underlies these phrases I shall discuss in the next Lecture ; I am here con- cerned simply with the alleged fact that sensa- tions occur in isolation, affections only in con- nection with sensations. If the difference exists, it is an admissible ground of distinction ; for although it is not a difference of attribute, it is nevertheless a difference that shows in the com- parison of element with element : the attempt to isolate an affection will result, always, in the iso- lation of paired sensation and affection. But does the difference exist ? Listen to Ktilpe. "We find sensations present," he says, '* where feeling is absent; that is, we have sensations which are neither agreeable nor dis- agreeable ; and we further find (such at least is the author's experience) feelings present w^liere sensation is absent ; that is, we have feelings 42 SENSATION AND AFFECTION which are not accompanied by or attached to definite sensations, or which arise where the ner- vous conditions of sensation are debarred from the exercise of their ordinary influence on con- sciousness." I do not think that many of the psychologists who recognise the independence of the affective element would subscribe, without qualification, to this opinion. But it is by no means uncommon — e.g. in experimental work upon the association of ideas — to find cases re- corded in which a feeling precedes or lags behind or outlasts its idea. And if Kiilpe is too extreme, Ladd can probably claim a widespread accept- ance of his view that "in the flow of the one stream of conscious life the feelings may assume either one of the three possible time-relations * towards the sensations and ideas by which we classify them ; they may fuse with them in the 'now' of the same conscious state, or they may | lead or follow them." Our final possibility is thus suflBciently disposed of.^ We have considered three meanings of the term 'subjective.' We have taken it to imply a tendency towards fusion ; individual variability of experience; and what we may call a second remove or a higher power of conscious existence. In every instance argument has been met by counter-argument, authority for by authority AFFECTION AS NON-LOCAL 43 against. We must, I think, conclude that, if there really is a difference between sensation and affection, the words 'objective' and 'subjec- tive' are ill chosen to express it.* (2) The distinction that we have next to con- sider is the distinction of local and not-local. Sensations, it is said, may be localised; affec- tions are not localisable. The distinction is ambiguous, since the 'locality' may be a position in perceptual space or a place in consciousness. We will take the question of 'outer' localisation first. Are all sensations localisable at some point of space .'^ "Allen Sinnesempfindungen," says von Frey, "ist die Beifiigung eines Lokal- oder Merkzeichens eigentiimlich." And he adds, "fiir den Unbefangenen wird gerade das Lokal- zeichen ein Beweis sein, dass der Schmerz ein den ubrigen Sinnesempfindungen gleichwertiges Element des Bewusstseins darstellt." That is definite enough. As usual, however, there are statements on the opposite side. "Eine Lokal- isation der Geruchsempfindungen als soldier," * In my own mind, the difTerence of subjective and objective appears always as a difference of texture: affection is softer, flimsier, more yielding than sensation, — however organic the sensation may be. This textural difference is what I 'feel* when I read, e.g., that "feeling as such is matter of being rather than of direct knowledge." 44 SENSATION AND AFFECTION writes Nagel, "gibt es genau genommen nicht. Ich fiir meine Person wenigstens vermag meine sehwachen Geruchsempfindungen gar nicht zu lokalisieren." When odours are localised, they are localised because their stimuli affect more than one set of end-organs: *'bei dem Geruchs- sinn ist das lokalisierende Vermogen gleich Null." Definite again ! Angell and Fite tell us, simi- larly, that "genuinely pure tones are essentially unlocalisable in monaural hearing"; *'it seems quite safe to say that in monaural hearing really pure tones are unlocalisable." And even with binaural hearing, it is not difficult so to arrange the conditions of observation that localisation is impossible. If you work with sounds of very low intensity, or if you work with tuning-fork tones in the open, your observer surrounded with a curtain, you will find cases in which there is sheer inability to localise. *' There are sounds," says Pierce, "that prior to all accessory expe- rience are sharply and definitely located. . . . But over against these sharply located sounds are others that can be assigned no position what- ever." Finally, Orth insists that there are or- ganic complexes, vague resultants of diffuse, weak stimulation, which cannot be localised. Not all sensations, then, are capable of localisation. But, on the other side, is affection unlocalis- AFFECTION AS NON-LOCAL 45 able ? Stumpf reports that the agreeableness and disagreeableness which accompany sensa- tions of the higher senses seem to him to have a certain spatial moment ; they are not localised, it is true, in the colours and tones themselves, but are felt "als im Kopf ausgebreitet." " Auch diese etwas unbestimmte Lokalisation ist aber Lokalisation." I have known observers to in- sist, similarly, that the pleasantness of the taste of chocolate cream is localised in the mouth, the pleasantness of tones and chords in the head or chest. Lagerborg bears witness to the same effect: "an einem Katermorgen nehmen wir Unlust im Kopf, im Rachen, im Magen wahr." And Storring distinguishes a Stimmungslust, "an der . . . die gesammten jeweilig vorhan- denen Bewusstseinsinhalte teilhaben," from a localised Emj)fi7idu7igslust, "die an die . . . [betreftenden] Empfindungen allein gebunden erscheint." There remains the question of * inner' locali- sation. Sensations, it is said, run their course side by side in consciousness ; affection is always coextensive with consciousness. The argument hinges, therefore, on the possibility of what are called 'mixed feelings.' Can pleasantness and unpleasantness exist simultaneously in conscious- ness ? 46 SENSATION AND AFFECTION If we appeal to the text -books, we find the expected divergence of opinion. "It is hardly possible in the present state of our knowledge to decide positively for or against the reality of these mixed feelings. ... In our own view, mixed feelings are certainly less well authenti- cated than cancellation of feeling." This is Kiilpe's statement in the Outlines. "Just as we may sense cold in the feet and warmth in the hands at the same time, so may we experience the pleasantness of a savoury dish along with the unpleasantness of a severe headache. . . . The affective accompaniment of complex mental formations may be extremely complicated." This is Ebbinghaus' statement in the Grimdzilge, "All the affective elements present in conscious- ness at a given moment connect to form an uni- tary affective resultant": that is Wundt. "A full sense of conflict between pleasure and pain arises when the two feelings are both present in a distinct and strong form, and are not so unequal in point of strength as to allow of one over- powering the other": that is Sully. And so we might continue. So far as I know, the coexistence of pleasant- ness and unpleasantness has only in three cases been inade the subject of experimental inquiry. In two, the result has been negative. Ortli, MIXED FEELINGS 47 in 1903, gave seven tests to four observers with a view to the analysis of the emotion of doubt. He records unpleasantness ten times and pleas- antness five times (in nine and four of the twenty- eight tests, respectively) ; there is no instance of simultaneity, but one very striking instance of succession, in which the order is pleasantness, unpleasantess, pleasantness again, and terminal unpleasantness. Alechsieff, in 1907, attacked the problem directly. He made twenty-nine experiments with pairs of stimuli (tastes and odours, tones and colours) so chosen that the one, taken alone, would be pleasant and the other unpleasant. "Aus diesen Versuchen kam- en wir zu dem Schlusse, dass Lust und Unlust nicht gleichzeitig in unserem Bewusstsein ex- istieren konnen, sie konnen nicht nebeneinander, sondern immer nur nacheinander von uns erlebt werden." I may add that in 1906 experiments of the same type were begun by Hayes in the Cornell laboratory, and — so far as they went — yielded a like result ; they were, however, too few in number to warrant separate publication. It is, perhaps, unfortunate that Orth was a pupil of Kiilpe's, Alechsieff a pupil of Wundt's, and Hayes a pupil of my own; for all three of us may be suspected of j)arti pris, and all three of the experimenters may therefore have 48 SENSATION AND AFFECTION been influenced — despite our efforts at im- partiality — by what is called ' laboratory atmos- phere.' It is still more unfortunate, I think, that the experiments themselves are so scanty. All the more welcome, then, is Johnston's paper of 1906. The investigation covers a period of two years ; the observers are twelve graduate students in Harvard University or Radcliffe College; the paired stimuli include colours, tactual surfaces, tuning-fork tones, noises, forms filled with different colours, and odours, as well as more complicated material ; and the outcome is definitely positive. We are informed, e.g., that, after training, eleven of the twelve observers "were all convinced that both feeling-tones, for tactual and visual impressions, could be present at once." I should be very sorry, now, to criticise for the sake of criticising. On the contrary, I would give a good deal, as the saying is, to have this question of mixed feelings settled in the one way or the other ; it is a question that has been with me, more or less insistently, for the past dozen years ; and I should have attacked it experimentally long ago, had I found an ade- quate method. Theories, believe me ! sit more lightly on their owners than is commonly sup- posed ; I would cheerfully exchange all my MIXED FEELINGS 49 * views' of feeling for a handful of solid facts. And if Johnston had proved his conclusion, I should accept it. So far, however, is he from proof that it is even difficult to say, in precise terms, what his conclusion is meant to be. Consider ! There were twelve graduate ob- servers, seven of whom "had had from one to five or more years' training in laboratory in- vestigations." Here is no levy of tiros, but a band of veterans. Had they never heard of feeling, never run across theories of feeling, never thought out for themselves what feeling might mean, never discussed the various defini- tions of feeling ? Moreover, several members of the group were available for the whole period of two years. Did they not work out a definition of their own, adopt some particular criterion or criteria of feeling in the course of the period ? Not a word is said upon these two points ; we do not know what the observers meant by feel- ing either at their down-sitting or at their up- rising. At the most we can guess from the intro- spective reports. And the very first report cited — the description of feeling for a particular shade of red — reads thus: "It feels as if it would be soft." No doubt it does ! But in what sense is ' soft ' a feeling ? The instruction given to the observers was of 50 SENSATION AND AFFECTION a very general kind. They **were requested to give themselves up to the situation and to report as accurately as they could the kind of affective state experienced." In preliminary series, with single stimuli, the same observers had, how- ever, been given a far more complex instruction. They had been told, in effect, to describe the feeling ; to report always all concomitant organic sensations; and to distinguish the significant organic concomitants from the accidental. This is a large order ! The rule of work in Aus- frageexperimeiite is, surely, to make instruction narrow and definite, for any given series, and thus to fractionate the introspections. There is no other way to secure unequivocal results. Wliile, now, the instruction in the experiments with paired stimuli was simpler than that in the experiments with single stimuli, there can, I think, be little doubt that the habit of observa- tion formed in the first series was carried over to the second ; only thus can I account for cer- tain of the introspections recorded. At any rate, the complexity of the original instruction was a mistake ; and the general instruction of the second series should, in my judgment, have been narrowed by specific regulations concerning, e.g., the direction and distribution of attention. Or if it seemed advisable to take series with MIXED FEELINGS 51 general instruction, then these should have been paralleled by other series in which the instruc- tion was variously narrowed. It is odd that Johnston says nothing, gives not a single refer- ence, on the score of the AusfragemetJwde. I have sometimes been charged with pre- ferring method to result. I do not know that that would be a crime ; I do not know why the search for truth should not be the sole end of a man's endeavours. If he sinned, he would sin in good company. But on lower ground the point is, of course, that your result is, after all, a function of your method; method is the road to result ; given a method, — and in fairly com- petent hands results will follow of themselves. Let us see, then, to what kind of result the method of which I have just spoken has led. It is essential that the results of this form of the method of impression be stated in the ob- servers' own words. Orth gives his complete records. Alechsieff gives complete samples. Johnston does not. While he writes out a temperamental analysis of his twelve observ- ers, later verified by themselves, he has edited and arranged the introspections, and only occa- sionally mentions an initial or puts a phrase into inverted commas. The temperamental analysis does not help us ; we want to know 52 SENSATION AND AFFECTION who said what, and how often, and in what context. Moreover, Johnston's own account is both meagre and confused. I can find no men- tion of the time during which the paired stimuli were exposed, — though this fact is of cardinal importance when it is a question of the coexist- ence or succession of affective processes. I find no mention of the number of experiments made with each observer, or of their arrange- ment, or of the time-interval between them. As for the outcome, I hope, but I cannot be sure, that the following summary is correct. Johnston notes (a) phenomena of complete fusion. This appears to be identical. with what he terms "a total mood with similar or harmoni- ous constituents." To be distinguished from fusion is (6) summation, where, e.g., two un- pleasant elements "exist throughout, each in turn intensifvinc: the whole undertone of feelinc:, but also remaining a feeling-tone of a particular kind." In (c) partial reenforcement, *'both feel- ing-tones contribute to a feeling of the same kind, yet do retain some individual characteristics which stand out for themselves." I do not see how this differs from (b) ; at most there is a slight difference of degree. What is differen- tiated by Johnston as (d) partial inhibition seems to be only a name given by certain observers to MIXED FEELINGS ^ 53 partial reenforcement. At any rate, these four are all cases of fusion or summation, and do not directly concern us. Next comes {e) total inhi- bition, which does interest us here. *' Cases of total inhibition . . . are by far the most frequent, as would naturally be expected [.^]. When sandpaper is being applied, and no re- pose is felt in the body, a colour, suddenly pre- sented, for a moment pleases the eye, but quickly loses all feeling-character, and can only be * in- tellectually perceived.'" "In cases of feelings of opposite nature occurring together, the stronger generally prevails, finally in most cases effacing all specific tone for the weaker element. An odour, for example, even when always un- pleasant, becomes less so when one looks at a pleasant colour, w^hen a feeling-tone can, or often when it cannot, be detected for the colour at the time." I understand from these sentences that when two opposite feeling-tones are aroused by two stimuli, operating at the same time, the regular or usual result is cancellation ; what we feel, if we feel at all, is the excess of the one over the other. But Johnston has a sixth category, of ij) merely simultaneous, independent coex- istence. *'When a very unpleasant form . . . is being felt, a slightly unpleasant colour tends to arouse often in this situation, as if by con- 54 SEXSATIOX AXD AFFECTIOX trast, a simultaneously pleasant element in the total experience." "When there is a clear strife between the two [feeling-tones], they both can exist as equal partial tones with an undertone of un- pleasantness in the failure to coordinate them." Now these experiences must, in the light of what has just been said, be rare. Why, then, — see- ing how critical the observations are, — were not the complete introspections given ? It looks to me as if, in both of the instances quoted, we w^ere in presence of fairly complex emotive pro- cesses ; the pleasure that arises 'by contrast,' and the displeasure that comes from 'failure to coordinate,' are not the feeling-tones of the stimuli. Are they feeling-tones at all ? Or are they organic complexes, the organic sensations characteristic of relief and of disappointment ? And again : if they are feeling-tones, were they strictly coexistent ? Is it possible to experience three affective processes at once — a pleasant- ness, an unpleasantness, and another unpleas- antness — and to hold them distinct at a given moment of time ? Very little v. eight, I am afraid, can be attached to this imperfect report of what are, admittedly, exceptional cases. — Let us now glance back over this whole dis- cussion. We found that the distinction of local and not-local, as referred to sensations and AFFECTION AS NON-LOCAL 55 affections, might mean two different things. It might mean, first, that sensations are, and affections are not, localisable in perceptual space. We found, however, statements to the effect that some sensations cannot be localised, while we found also alleged instances of the localisa- tion of affection. We may therefore reject this criterion, without going into the further question whether locality, some form of Merkzeichen, is an attribute that shows in the single sensation. The distinction might mean, secondly, that sensations run their course side by side in con- sciousness, while affections are always coex- tensive with consciousness. The experimental evidence, so far as it goes, appears to bear out this contention. Orth, Alechsieff, and Hayes find no mixed feelings ; Johnston finds that mixed feelings are the exception and not the rule ; and we have seen that the exceptional in- stances are themselves not above suspicion. On the other hand, the experimental evidence is scanty and incomplete; and psychological opin- ion at large is sharply divided. Moreover, it might be urged that there are occasions when consciousness reduces to a single sensation : pain, or a deafening noise, or a blinding glare. So we seem to be as uncertain at the end as we were at the beginning.^ 56 SENSATION AND AFFECTION (3) A third distinction, which we owe to Wundt, is that of difference and antagonism. Sensations range between maximal differences; feelings, between maximal opposites. "Allge- mein werden die Empfindungsqualitaten durch grosste Unterschiede, die Gef iihlsqualitaten durch grosste Gegensatze begrenzt." I think that there has been a tendency, in the discussions of feeling, towards too cavalier a treatment of this distinction. It is really not quite easy to see what the difference means, and not quite easy to bring valid argument for or against it. Before, how^ever, we come to details, let us notice that Rehmke refuses to connect pleasantness and unpleasantness in any way whatever. "Lust und Unlust sind 'incommen- surable Grossen,' wie Ton und Farbe es sind." *'Die Thatsachen des Seelenlebens . . . geben nicht den geringsten Anlass zu der Behauptung : 'Lust und Unlust sind gegensatzliche Zustande, welche durch einen Lidifferenzpunkt in einander iibergehen.'" "Lust und Unlust als that- sachlich besondere Bestimmtheiten der Seele haben nichts mit einander gemein." This view is, no doubt, exceptional; but it deserves con- sideration in the present context. The objection usually brought against Wundt 's formula is that there are sensations, too, which AFFECTIONS AS OPPOSTTES 57 range between maximal opposites. This is probably what Stumpf has in mind when he says that the statement is **so offenbar mit den Tatsachen in Widerspruch, dass wir nicht darauf einzugehen brauchen." What, then, are the 'Tatsachen'? Orth refers to warmth and cold: "die Empfindungen des Temperatur- sinnes bewegen sich in derselben Gegensatz- lichkeit." He cites also the organic complexes of hunger and satiety, and the bodily states that we term 'fresh' and 'tired.' Klilpe, m his sec- tion on sensations of temperature, argues in the other direction. "A simple increase or diminu- tion of temperature can change either sensation into its opposite, the path of change lying through a point of indifference or zero-point. There is no analogy to this fact in the sphere of sensation, though there is a very complete one in that of feeling." Ebbinghaus passes very lightly over the 'gegensatzliche Gliederung' of the affective processes: "sie stehen hiermit iibrigens nicht allein," he says, and quotes the three cases given by Orth. All this seems to me rather super- ficial. What we mean by maximal differences of sen- sation is clear enough. The attributes of sen- sation are, as we saw in the last Lecture, either qualitative or intensive. If they are intensive, 58 SENSATION AND AFFECTION they range — so to say, vertically — between zero and infinity, or rather between a lower and an upper limiting value. If they are qualitative, they range — so to say, horizontally — between extremes that are equally remote from both of these limiting values. How can there be an * opposition ' of sensory qualities ? We are re- ferred to the sense of temperature : but there is no sense of temperature. There are a cold sense and a warmth sense, — different senses. The thermometric scale is continuous ; but that has nothing to do with the case. Years ago I w^as troubled by this antithetical account of warmth and cold, and made a series of experiments, from warmth to cold and from cold to warmth, in order to trace the passage of the one to the other through the point of indifference. I never found that point. Ktilpe, who entirely believes in its existence, confesses that he, too, has been unable to verify its occurrence. The whole construction is artificial ; and the appeal to tem- perature is an appeal to physics, not to psy- chology. As for hunger and satiety, — try them ! Introspect your organic sensations in moderate hunger and after a hearty dinner. So far from finding opposition, antagonism, you will find a very general resemblance. Lastly, the sensa- tions of freshness and tiredness, in so far as they AFFECTIONS AS OPPOSITES 59 are muscular in the strict meaning of that term, — in so far, that is, as they belong to a single sense, — range from bright to dull, from light to heavy, from lively to dead : but these are qualitative differences, akin to the differences of black and white in vision and of high and low in audition; there is no opposition or antago- nism between them. Now look at the other side of the shield. Our ordinary speech is very apt to couple words which, in a loose w\ay, may be considered as *opposites.' We speak of hard and soft, rough and smooth, sharp and blunt, wet and dry, strong and weak, keen and dull, light and heavy, warm and cold ; we speak of dark and fair, hungry and thirsty,* wide-awake and drowsy, fresh and tired, good-looking and ugly, clever and stupid, good and bad. The list might go on indefinitely. It is perfectly clear that, in most instances, there is no real opposition be- tween the paired terms ; they stand simply for extremes of possible difference, whether in an attribute of sensation or in formations as com- plex as character and intelligence. Why, then, are two or three of them singled out, as express- * When Alice tried to 'quench her thirst' with the Red Queen's biscuit, — "and it was very dry," — the antagonism between hunger and thirst was at least as real as the alleged antagonism of hunger and satiety ! 60 SENSATION AND AFFECTION ing opposition ? Why, indeed, — unless be- cause an affective opposition is implied ? Pleas- ant warmth and unpleasant cold; pleasant satiety and unpleasant hunger; pleasant fresh- ness and unpleasant tiredness : is not that the opposition ? I think that wherever the opposi- tion is conscious, it is affective. Notice, too, that it is never absolute; cold may be pleasant in summer, unpleasant in winter. Hunger may be pleasant, a 'jolly' hunger. Tiredness may be a * comfortable ' tiredness. We thus oppose degrees of cold, degrees of hunger, degrees of tiredness, as well as cold and warmth, etc. On the whole, warmth and satiety and freshness are pleasant, and cold and hunger and fatigue are unpleasant ; here is the general opposition to which our authorities appeal : but there are special oppositions that, if sensory at all, must be intensive and not qualitative. Notice, lastly, that other paired terms may be brought into conscious opposition if only we grant them an affective colouring; a carving-knife may be beautifully sharp or horribly blunt, a bed may be comfortably soft or dreadfully hard. And here as before there are oppositions of degree; comfortably soft may contrast with too soft, dreadfully hard with just hard enough. This interpretation of the facts of 'sensory INTENSITY OF IDEAL FEELINGS 61 opposition' squares very well with the system- atic doctrine that all * psychological ' contrast — • I use Lipps' phrase — is a matter of feeling : that the ordinary man looks small by the side of a giant because you are disappointed, and looks large by the side of a dwarf because you are surprised. To that doctrine I subscribe. But neither it nor the considerations which I have just been urging tell us what affective opposition is. Is it mutual incompatibility in consciousness ? Those who — like Lipps, in his earlier writings — deny the possibility of mixed feelings might agree to such a definition, and we have seen that the evidence against mixed feelings is fairly strong. Only, we saw also that it is not conclusive.^ (4) A fourth criterion of affection is suggested by Kiilpe. You will remember that Ktilpe classi- fies sensations as peripherally excited and cen- trally excited ; the distinction corresponds to that between sensation and image, and we shall do well, perhaps, to employ the more familiar terms. He classifies feelings in the same way, as peripherally and centrally excited. Since, however, very few psychologists agree with him that affective processes can stand alone in con- sciousness, w^e shall do w^ell, again, to phrase 62 SENSATION AND AFFECTION the difference as that between the affection which accompanies a sensation and the affection which accompanies an image. Now the greatest dis- parity between sensation and image, Kiilpe says, shows on the side of intensity, and the intensive difference between the two processes is "nor- mally recognised in every case by introspection." On the other hand, the affection which accom- panies the image is '* usually as vivid" — that means here, as intensive — as that which goes with the sensation. "Only the very highest degrees of sense-pleasure and sense-pain are now able to overpower the centrally excited, 'higher' feelings." Here, then, is our criterion. Image is weaker than sensation, but the image- affection is intensively equivalent to the sense- affection. I wish to avoid all reference to the systematic question of affective reproduction, and I shall therefore let the phrase 'centrally excited affec- tion' pass without comment. We know what Kiilpe means. But shall w^e accept the state- ment of fact upon which his criterion rests ? Ladd very definitely does not. "In general," he writes, "ideal pleasures and pains, when measured by a strict standard of quantity, are much inferior to those occasioned by strong sensations." And more strongly still: "Ideal INTENSITY OF IDEAL FEELINGS 63 pains and pleasures are not comparable in mere intensity with sensuous pains and pleasures." Contradiction could hardly be flatter. And con- tradiction is what we shall have, here and else- where in the psychology of feeling, until we can work out an experimental control of introspec- tion. As Wundt said long ago, *'Selbstbeo- bachtung ist ausfiihrbar, sie ist es aber nur unter der Bedingung der experimentellen Beobach- tung." We cannot, however, leave the matter at this point, since Stumpf has taken Klilpe's sugges- tion seriously, and has brought two arguments against it. First, of course, he rules out the appeal to emotion ; he denies the continuity of sense-feeling and emotion. Then he says : sup- pose that Klilpe's statement were literally and universally true; still, the difference that he signalises would not be very important. '*Denn wir finden unter den verschiedenen Sinnen doch auch bei allem Gemeinsamen genug charakter- istische Verschiedenheiten : der eine zeigt Simul- tankontrast, der andere nicht, der eine zeigt messbare Ausdehnungsunterschiede, der andere nicht, u.s.w." He points out, also, that if you regard sensation and image as the same in kind and different only in degree, then Klilpe's dis- tinction loses its theoretical significance. 64 SENSATION AND AFFECTION The second objection offers an alternative to Klilpe's view. May it not be, Stumpf asks, that the image-affection is normally weaker than the sense-affection, — just as the image is normally weaker than the sensation, — but that the image-affection is very easily transformed into a sense-affection ? In other words, may we not be liable '*in ganz gewohnlichen Fallen" to affective hallucinations, just as '*unter beson- deren Umstanden" we are liable to hallucinatory images ? No doubt, cause must be shown ; but cause can be shown, Stumpf thinks, in terms of his own theory of Gefuhlseinpfiiidungen, — the theory w^liich we are to discuss in the next Lecture. I do not know how^ to meet the first objection. If the attributes available for definition are merely quality, intensity, extent, and duration, and if extent is not an universal attribute of sensation, then we might, certainly, classify the mental elements at large as spatial and non- spatial. The classification would, indeed, be superior to Klilpe's distinction of sensation and affection, in the sense that it is based upon a difference observable in the single element, whereas Klilpe's intensive criterion requires the presence, along with affection, of sensation or image. We should grow accustomed, after a AFFECTIVE HABITUATION 65 while, to placing sight and touch in a class by themselves, and bracketing pleasantness-un- pleasantness with tones and odours and the rest. At the same time, I hardly suppose that Stumpf meant his argument to be worked out in detail ; the gist of it is, simply, that Kiilpe's difference is unimportant. And to that we can only reply that Kiilpe thinks it important, and that Ladd denies its existence. The second objection stands or falls with Stumpf 's personal views : its consideration must, therefore, be postponed. "^ (5) It has been said that the fact of habitu- ation, the loss or change of quality with lapse of time, marks off affection from sensation. The habitual sensation is indifferent, has ceased to affect us, w^hile its sense-quality remains unchanged. The obvious reply is that affective adaptation has its direct analogue in sensory adaptation. As we become adapted to colours, tastes, odours, pressures, so do we become habituated to pleas- antness and unpleasantness. This is Stumpf 's position. Ebbinghaus, on the contrary, finds only *'eine verhaltnismassig schwache Analogic" between the two sets of phenomena. We must therefore inquire further. 66 SENSATION AND AFFECTION The statements in the text-books are conflict- ing. Kiilpe says that habituation means in- difference. '* There is no evidence that unpleas- antness passes into pleasantness. Observations that seem to point towards any such process are referable to other causes. At least, the reverse passage, of pleasantness into unpleasantness, will be found to be of hardly less frequent occur- rence ; and no one would attempt to explain it by habituation." Ebbinghaus declares that "diese blosse Abschwachung der Gefuhlswerte erst eine Seite der Sache ist," and seeks to show how, in terms of adaptation, pleasantness may pass through indifference to unpleasantness and unpleasantness through indifference to pleasant- ness, — the very occurrences that Kiilpe ex- cludes. Both authors, I suppose, have Leh- mann in mind ; but they put a different estimate upon Lehmann's conclusions. Lehmann distinguishes between affective ha- bituation to continuous and to intermittent stimuli. Where the stimulus is continuous, affective blunting is "ein rein scheinbares Phii- nomen," pure illusion. You begin, we will say, with a pleasantness. As time goes on, the sense- organ becomes adapted; you have an "Ab- stumpfung der Empfindung" which naturally means also an "Abstumpfung des Gefiihls." AFFECTIVE HABITUATION 67 If the stimulus persists, unpleasurable. sensations from foreign stimuli make incursion into con- sciousness, and the indifference becomes un- pleasantness. The two factors, of sensory adap- tation and foreign interference, may operate singly or in various combinations. Or you begin with an unpleasantness. This continues, with increasing intensity, until the onset of sensory adaptation in the form of nervous ex- haustion. If indifference occurs, it occurs only when and because the sensory side of your ex- perience drops out of consciousness; as soon as the sensation reappears, the unpleasantness re- appears with it. You may, then, reach a stage of indifference, of forgetfulness, but the original unpleasantness never changes to the opposite quality. Now turn to intermittent stimuli. You have, according to Lehmann, precisely the same phe- nomena as before : either the sensory side of the experience becomes obscure, through diver- sion of attention, or unpleasurable sensations from foreign sources invade consciousness, or both factors cooperate to change the original affective quality. But he goes on to point out that the intermittently repeated stimulus does not wholly lose its affective significance ; there is a law of the * indispensableness of the habitual.' 68 SENSATION AND AFFECTION General sensory adaptation, the "Gewohnung des Organismus," leaves a need, a '*Bedurfnis." As the satisfaction of a need is pleasurable, you may have, in terms of this law, a shift of affective quality from unpleasant to pleasant ; only, the shift is indirect, from unpleasantness of stimulus to indifference, from that to the unpleasantness of need, and from that again to the pleasantness of satisfaction. The converse change, from pleasant to unpleasant, is not mentioned by Lehmann. Kulpe, then, repeats Lehmann exactly. There is no evidence of the change, under habituation, from unpleasantness to pleasantness ; Lehmann shows that the passage is indirect. No one would think of ascribing to habituation the change from pleasantness to unpleasantness ; Lehmann says nothing of such change. Eb- binghaus, on the contrary, reinterprets Lehmann. He accepts the law of custom, of the indis- pensableness of the habitual, but makes the unpleasant stimulus pass directly, through in- difference, to pleasantness. And he parallels this law by a law of tedium or ennui, which is realised when an originally pleasant stimulus passes directly, through indifference, to un- pleasantness. In my own opinion, affective habituation is AFFECTION AND CLEARNESS 69 a phenomenon of the same order as sensory adaptation, and results always and only m in- difference. Even if Ebbinghaus is correct, and quality passes into opposite quality, we have a sensory analogy in the case of vision : adapta- tion to yellow means blue-sightedness, local adaptation to green means a purple after-image. However, our present concern is with the dif- ference between sensation and affection; and we have gone far enough with the phenomena of habituation to see that, in the present state of psychology, appeal to them is hopeless.^ (6) I have postponed to the last the discussion of a criterion which, to my mind, is the most obvious and the most important of all. It is this : that affections lack, what all sensations possess, the attribute of clearness. Attention to a sensation means always that the sensation becomes clear; attention to an affection is im- possible. If it is attempted, the pleasantness or unpleasantness at once eludes us and dis- appears, and we find ourselves attending to some obtrusive sensation or idea that we had not the slightest desire to observe. Kulpe emphasises this difference between the elementary processes, and at the same time forestalls a misunderstanding. " A weakly pleas- 70 SENSATION AND AFFECTION urable feeling," he writes, "is intensified by the direction of the attention upon its concomitant sensations, and an impression which stands on the border line between pleasantness and un- pleasantness may be made unpleasant by an in- tense concentration of the attention upon it. In a certain sense, then, attention is a favourable condition for the feelings as it is for sensation." That is the removal of the possible misunder- standing. "But," Kiilpe goes on, "curiously enough, the result is quite different if attention is turned upon the feeling itself. It is a familia r fact that contemplatio i i of the feelincrs, the de vo- tion of special attentio n t o them, lessens the ir intensity and prevent s thei^jiatural expressi on . This diminution of intens ity . . . [is shown by] a tendency of the affective contents to disappe ar altogether, to make way for the state of indiffer - ence. . . . Attention, then, is advers e to t he feelings, when concentrated directly upon them ." He then quotes the introspective report which accompanied certain experiments made by the method of expression. "The subject often in- sisted that the feeling had altogether disappeared under attention, and that it was very difficult, in any case, to attend to pleasantness or unpleas- antness. Feeling has too little objectivity and substantiality for the attention to be directed AFFECTION AND CLEARNESS 71 and held upon it. It is focussed for a moment, and then other processes, especially organic sensations, interpose and take possession of the conscious fixation-point." And later on, when describing the effects of attention, he says : "While pleasure and pain {Lust und Leid) are brought far more vividly to consciousness by the concentration of attention upon their con- comitant sensations, they disappear entirely when we succeed (and we can succeed only for a moment) in making the feeling as such the object of attentive observation." I myself, in 1894, published a brief account of experiments which had led to a like result. Further evidence is furnished by Zoneff and Meumann. These investigators made experi- ments in which the observers were instructed to attend now to the stimulus, now to the feeling aroused by the stimulus. The instruction proved to be ambiguous. In certain cases, "man con- cent rirt sich, d. h. man behalt das Gefiihl will- kiirlich eine relativ langere Zeit im Blickpunkte des Bewusstseins und analysirt dasselbe. Es wird etwa dartiber nachgedacht, ob das Gefiihl mehr oder weniger angenehm bezw. unangenehm ist. Hier findet eine wirkliche Analyse des Ge- fiihls statt, die von einer gewissen korperlichen Spannung begleitet ist." In other cases, "der 72 SENSATION AND AFFECTION Reagent suelit sicli das Geflihl moglielist zum Bewusstsein zu bringen, ohne aber dasselhe zu analy siren, oder mit anderen Worten, das Gefiihl tritt in den Blickpunkt des Bewusstseins, dabei hleiht es aber, es geschieht mit ihm nichts weiter. Das Gefiihl wird so zu sagen mit Hingebung gefiihlt." As for the result: "eine blosse Rich- tung der Aufmerksamkeit auf das Gefiihl [the second case] verstarkt dasselbe, wird dagegen das Gefiihl zum Gegenstand einer psycholo- gischen Analyse gemacht und in diesem Sinne Geojenstand der Aufmerksamkeit, so wird es be- deutend geswacht, ja sogar ganz aufgehoben." The general sense of these passages is clear. The unpractised observers, when instructed to 'attend' to the feeling, thought that they were to do the best they could for it, to assume the mental attitude most favourable to it; to resign themselves passively to the feeling, to let it have its way with them. Under these conditions, the feeling naturally attained its fullest intensity. The more practised observers — this distinction is drawn by the authors, not by me — sought to abstract from the sensible concomitants, and to attend strictly to the affective contents as such. Under these conditions, the feeling was notably weakened, if not entirely destroyed. I say that the general sense of the passages is AFFECTION AND ATTENTION 73 clear. The wording, however, is obscure and can be justified, if at all, only by the considera- tion that Zoneff and Meumann are purposely taking a non-committal position as regards sys- tematic questions. Thus, the unpractised ob- servers certainly did not 'direct their attention to the feeling' in the same sense that one directs one's attention to a sensation. So far from at- tending, reaching out to the feeling, they sat back and let the feeling come. The very fact that they did this — that they instinctively refrained from attention to the feeling itself, in order to give it first place in consciousness — shows how unnatural, not to say impossible, the literal in- struction was. Again, the practised observers are said to 'analyse the feeling.' But how could they 'analyse' what a few pages back has been called an 'elementary' feeling.? What they did was to ideate the feeling, to reflect upon it, to ask themselves questions about it : all this, in the vain effort to hold it as a sensation is held. Once more : what is meant by the pres- ence of a feeling "im Blickpunkte des Bewusst- seins".? The associations that come with the phrase are drawn from the sphere of sensory attention, so that here Zoneff and Meumann seem to have departed from their non-committal attitude. You cannot, after all, free yourself 74 SENSATION AND AFFECTION from * Voreingenommenheit ' by a fiat of will; you have approached the problem by way of certain concepts, and though you deny them your speech will betray you. In his recently published Experimentelle Pdda- gogik, Meumann writes as follows: *'[Es] muss noch von emotionaler und voluntionaler [ !] Auf- merksamkeit gesprochen werden, denn unsre Aufmerksamkeit kann sich ebensogut auf Will- enshandlungen und Gefiihle richten und es ist ein blosses theoretisches Vorurteil, wenn das von manchen Psychologen geleugnet wird. Sie konnen sich jederzeit selbst davon uberzeugen, dass wir unsre Gefiihle einer analysierenden Beobachtung zu unterziehen vermogen, wie jeden anderen Bewusstseinszustand, dann richtet sich die analysierende Aufmerksamkeit auf das Gefuhl." I must confess that this passage stag- gers me. Anybody at any time may convince himself that he can attend to his feelings ? And yet, four years earlier, Meumann had attached a "quite especial importance" to his and Zoneif's experimental work on that point, and had stressed the fact that *'*die Richtung der Aufmerksam- keit' auf ein psychisches Erlebniss ein sehr vieldeutiger Ausdruck ist, mit dem ganz hetero- gene Vorgange bezeichnet werden." But, you will say, he speaks now of 'analytical observa- AFFECTION AND ATTENTION 75 tion,' * analytical attention.' So he does, — without defining the adjective. And then this analytical attention to the feelings is of the same order as analytical attention to *'any other state of consciousness." Four years earlier we were told that the feelings were "bedeutend ge- schwacht, ja sogar ganz aufgehoben"! — I began this discussion with a reference to Klilpe. I did so, because I know of no experi- ments earlier than his. The doctrine which I am defending — "der Schlendrian der alten Aufmerksamkeitstheorie " which, according to Meumann, is a mere theoretical prepossession — is, of course, very much older. It is aptly phrased, e.g., by Hamilton. "Reflection," he says, "is properly attention directed to the phe- nomena of mind." "We are, indeed, able to constitute our states of pain and pleasure into objects of reflection, but in so far as they are objects of reflection, they are not feelings, but only reflex cognitions of feelings." It is fully worked out by Ward. It is implicit in Wundt's theory of feeling as the "Reaction der Appercep- tion auf das einzelne Bewusstseinserlebniss," the " Reactionsweise der Apperception auf den Bewusstseinsinhalt." Ktilpe's acceptance of the Wundtian theory is grounded upon the impli- cation ; and Wundt's own language is exceed- 76 SENSATION AND AFFECTION ingly careful. Hollands writes : " Concerning the method of feeling-analysis, we find the statement that feeling . . . cannot be isolated as an object of attention. . . . This is the invariable teach- ing of Wundt." Even when Wundt slips, he furnishes his own corrective. While, e.g., he speaks incautiously, in one passage, of the * ap- perception' of the affective tone of an obscurely perceived sensation, he explains in another that well-marked feelings may accompany a * Vorstel- lungsinhalt' which "wegen der vorwaltenden Richtung der Auf merksamkeit " is not apper- ceived; and the two accounts refer to the same experience. I need not multiply quotations. No doubt, there are dissentient voices. Saxinger, for in- stance, cites Lehmann to the effect that we are able, by voluntary direction of attention, to bring a feeling to the forefront of consciousness. True, this is a misquotation : Lehmann speaks, in fact, of the direction of attention upon a "betonte Vorstellung," not upon a feeling; but Saxinger's words show that he, at any rate, has no difficulty in conceiving of attention to a feel- ing. Indeed, he speaks, a little later on, of the ' Beleuchtung ' of feelings by the attention. And the authors who identify affection with sensation must, of necessity, take Saxinger's view, — un- THE CRITERIA OF AFFECTION 77 less they avoid the present issue altogether. On the whole, however, I think that this sixth criterion stands its ground more firmly than any of the others that we have considered.^ What, now, is to be our general conclusion ? This, I think : that two of the proposed criteria of affection must probably be given up; that two others are extremely instable ; and that two deserve very serious consideration. The two that we must apparently discard are those fur- nished by habituation and by central intensity. The two that we pronounce doubtful are those of subjectivity and of non-localisableness. The two that, at any rate, give us pause are those of qualitative antagonism and lack of clearness. This statement is as near as I can get to an impartial verdict. The whole discussion illus- trates the difficulty of discriminating between elementary processes in any other way than by appeal to experience itself. And I would ask you to remember that that appeal still remains. All of the distinctions between sensation and affection profess to be drawn from experience ; the wording may be clumsy, or suggestive, or individually coloured; but the difference itself is either there or not there, in your own intro- spection. 78 SENSATION AND AFFECTION I would ask you, also, to remember one other thing. A psychologist who definitely accepts any single criterion, and so makes affection an independent mental element, may very well revise our conclusion, and ascribe value to cri- teria that we have disputed or rejected. We have proceeded serially, taking each distinction by itself. That was necessary, in the interests of clearness; but I do not know that it was quite fair. If, for instance, we were to consider sub- jectivity and coextension with consciousness along with antagonism and lack of clearness, making the four characters interdependent, as, to a large extent, they really are, we could, I believe, con- siderably strengthen the case for an elementary affection. I can do no more than mention the point here ; I shall recur to it, briefly, in my con- cluding Lecture/^ Ill THE AFFECTIONS AS GEFUHLSEMPFINDUNGEN LECTURE III THE AFFECTIONS AS GEFUHLSEMP- FIN DUN GEN MANY attempts have been made, and for various reasons, to identify affection with sensation, and thus to reduce all the mental ele- ments to a single kind. We must rule out of con- sideration here, as we did also in the previous Lecture, anything that savours of epistemology. For that matter, there is more than enough to occupy us on the purely psychological plane ; the introspective resemblance between pleasantness- unpleasantness, on the one hand, and certain sensations, on the other, has been urged again and again in the history of psychology. I take two modern instances. Bourdon, in 1893, iden- tifies pleasure with the sensation of tickling. " Le plaisir est une sensation speciale et non pas une sensation commune ni une propriete de toutes les sensations ; et il est de meme nature que la sensation speciale de chatouillement." "Le plaisir serait un chatouillement diffus, de faible intensite, tandis qu'au contraire le chatouillement serait en quelque sort un plaisir G 81 82 GEFUHLSEMPFINDUNGEN bien localise et de grande intensite." ^ In 1894 von Frey, working from the opposite direction, identifies unpleasantness with pain, while he makes pleasantness a negative matter, absence or cessation of pain. "Der Schmerz," he says, '* [ist] die einfachste Form des Unlustgefuhls " ; and "die tagliche Erfahrmig lehrt, dass es die Aufhebung des Schmerzes ist, welche uns Lust bereitet." ^ Bourdon is able to quote, on behalf of his theory, authorities as high as Descartes and Bain ; ^ and von Frey, though he does not say so, is championing a doctrine of pleasure which is as old as Plato/ I do not, however, intend to enter upon the history of our subject; that would take me too far afield. I intend only to expound and criti- cise one notable attempt, made very recently, to bring the sense-feelings, all simple affective experience, under the rubric of sensation. I refer to Stumpf's paper Ueber Gefilhlsempfin- dungen, which was read before the Society for Experimental Psychology at Wlirzburg in April, 1906, and was published in December of the same year, with slight additions and modifica- tions, in the Zeitschrifl Jur Psychologie.^ The range of Stumpf's inquiry is limited to what are ordinarilv called the 'sense-feelings,' SENSATION AND AFFECTION 83 die sinnlichen GefiXlile. These include, "first, the purely bodily pains (that is, those which appear without any essential concernment of intellectual functions *), whether they are set up from without or from within the organism; secondly, the feeling of bodily well-being in its more general and its more special forms, the latter including the pleasure-component in tickling, the feeling produced by itching, and the sexual feelings ; and lastly the agreeableness and dis- agreeableness that may be connected, in the most various degrees or gradations, with the sensations of all or nearly all the ' special ' senses, with temperatures, odours, tastes, tones, colours." The point of view which the inquiry adopts is primarily descriptive.^ We begin with a discussion of the three pos- sible views of elementary affective process."^ The affection may be an attribute of sensation, — an 'affective tone' of sensation. Or the affec- tion may be a mental element, distinct from and coordinate w^th sensation. Or, lastly, the affec- tion may be itself a sensation, a sensation of a special kind, like the visual or the kinsesthetic. The first of these three views Stumpf disposes of as it deserves. It is a view which received its * I take this to be the meaning of the phrase "ohne integrier- ende Beteiligung intellektueller Funktionen." 84 GEFUHLSEMPFINDUNGEN coup de grace at the hands of Kiilpe in 1893,^ — the year of publication, be it remembered, of the fourth edition of the Physiologische Psy- chologies in which affection still figures as a 'third attribute' of sensation.^ Kiilpe points out that affection cannot be an attribute of sensation of the same sort as the recognised attributes, be- cause it has attributes of its own. Sensations show differences of intensity, quality, time, and (in some instances) space; affection shows dif- ferences of intensity, quality, and time. Ziehen, in the 1906 edition of his Leitfadeii, seeks to meet this argument. "Dem gegeniiber verweise ich Sie auf das Beispiel eines chemischen Prozesses (z. B. einer Oxydation), welclie selbst eine be- stimmte Intensitat und Qualitat hat und oft zugleich noch von einem Liclit von bestimmter Intensitat und Qualitat begleitet ist." ^^ Stumpf replies, rightly, that the attribute of an object, as these terms are employed in everyday life, is one thing; and that the attribute of a sensa- tion, as these terms are employed in psychology, is another and a very different thing. An 'object' is an empirical collocation of attributes which are themselves sensations or sense-deriva- tives ; we can think away the scent of a flower, and leave the flower a concrete object as it was before. But the attributes of sensation are THE DOCTRINE OF AFFECTIVE TONE 85 known only by abstraction ; they are the modes of variation of a wholly simple contents ; we cannot think any one of them away without at the same time thinking aw^ay the sensation/^ Stumpf's rejoinder thus leads us to Kiilpe's second argument : that the annihilation of an attribute of sensation carries with it the disap- pearance of the sensation, whereas a sensation may be non-affective, indifferent, and still be far removed from disappearance. I do not see how these arguments can possibly be answered, and I agree with Stumpf that "man sich wundern muss, wie [die betreffende Anschauung] immer noch von manchen festgehalten werden kann." ^^ Yet we find Marshall, in January of the present year, positing an ' algedonic quality ' of sensation : "each elementary presentation must display either agreeableness or disagreeableness, or indifference which is a mode of transition be- tween the other two." ^^ What is this *mode of transition ' ? If it is really indifference, neither pleasantness nor unpleasantness, it is nothing at all ; and how can nothing at all be a * qual- ity ' ? If it is indifference in the affective sense, the indifference of satiety, of ' having had enough of a thing,' of 'being tired of it,' then — as Ziegler says — "hat das Gleichgiiltige stets etwas vom Unangenehmen an sich";^^ indifference 86 GEFUHLSEMPFINDUNGEN is a stage on the road to aversion, nausea, dis- gust; it is already unpleasant, not a *mode of transition' from pleasantness to unpleasantness. I have been told that, in philosophy, errors never die; and it may be that they die hard, in psy- chology, because that earlier habit of immortality is still strong upon them. There remain, then, the alternatives : that affection is a second mental element, and that it is a kind of sensation. Stumpf here throws the burden of proof upon the advocates of affec- tive independence. Unless the differences be- tween sensation and affection are primary and universal, the separation of the processes runs counter to the scientific principle of economy. ^^ He accordingly reviews the three principal argu- ments of his opponents : the relation of sense- feeling to emotion, the subjectivity of feeling, and its lack of spatial localisation and extension. ^^ What he has to say on these topics we already know; his objections were mentioned and al- lowed their due weight in the preceding Lecture. You will remember that he makes a sharp divi- sion, in his own system, between sense-feeling and the higher, intellectual feelings, the Affekte oder Gemutshewegungen, and that he definitely rejects the two remaining criteria. I will only add now that he has, in my opinion, passed too CUTANEOUS AND ORGANIC PAIN 87 lightly over certain other proposed criteria ; and that the appeal to a * principle of economy' is worth very little, because the appeal in science lies always to the facts of observation. How- ever, it is precisely to the facts that Stumpf next takes us. He opens with a section on 'sensations of pain, and the sensations of pleasure (Lustemp- findungen) that take their origin from cutaneous stimulations or from vegetative states . ' ^ ^ And he makes every effort, at the outset, to prove that pain is a department of sense, and the pain-qual- ity a quality of sensation. " Es ist also die Isolie- rung dieser Empfindungsqualitat, sozusagen die Reinzlichtung des Gefiihlssinnes, gelungen." ^^ **Die principielle Frage, auf die es fur uns gegenwartig ankommt, ist . . . ob es Schmerz- empfindungen in der gleichen Bedeutung wie Farbenempfindungen , Geruchsempfindungen gehe, als echte und eigentliche Silmesqualita- ten."^^ Let me say at once that nobody, who knows anything at first hand of the psychology of cutaneous sensation, would be tempted nowadays to traverse this position. That there is a sense of pain is a fact as well established as that there is a sense of pressure. So far, I agree entirely with Stumpf. Nevertheless, there are two points in his exposition, two very closely related points, 88 GEFUIILSEMPFINDUNGEN that seem to invite criticism. The first concerns the nature of the pain-quality, and the second the nature of the unpleasantness of pain. Stumpf is fully convinced of the painfulness of the pain-quality. ''Der Schmerz ist eben schmerzhaft, dass ist seine berechtigte Eigen- tumlichkeit, daran kann, glaube ich, selbst die feinste Psychologic nichts andern." "^ Believers in a separate affective process regard the under- lying sensation in the experience of pain as pain- less, schmerzlos; "wahrend wir uns zu der Ansicht gefiihrt sehen, dass das sogenannte Schmerzgeflihl die sinnliche Qualitat selbst ist, und dass der Schmerz in jener angeblichen nur zugrundeliegenden Sinnesempfindung schon durchaus komplett gegeben ist." -^ A difficulty arises, of course, in connection with the sensation of prick or sting. Goldscheider expressly re- marks that his * secondary sensation' need not be painful ; under the most favourable conditions it is a '' iem-stecheiides Geflihl von nicht schmerz- liaftem Charakter." -^ iVnd the ' Stichempfind- ungen ' aroused directly by cutaneous stimulation are not painful. Stumpf meets the difficulty by suggesting that the sensations in question may belong to the sense of pressure. But he does not, in any case, regard it as momentous ; whether the 'Stichempfindung' is mediated by the nerves PAIN AND UNPLEASANTNESS 89 of pressure or by the nerves of pain, the important thing is the appearance of pain as a sensible quality. That is the first point. In the second place, Stumpf looks upon the unpleasantness of pain, not as concomitant affective process, but as the qualitative character of pain itself. To add an affective tone of unpleasantness, he says, "scheint mir zwecklos." ^^ Why should you call upon a second genus of mental elements to make a thing what it is in its own right, — to make a pain painful ? ^^ We hear sometimes of pleas- urable pains ; but the phrase is misleading. You may, perhaps, have pain-sensations and agreeable sensations at one and the same time from different regions of the skin, and the two qualities may possibly enhance each other,** by a kind of contrast" ; but that is not the same thing as feeling an agreeable pain. The pleasures of asceticism and of martyrdom are matters of emo- tion ; the sensible pain either persists, but is held in check by intellectual rapture, or disap- pears in the analgesia of the ecstatic state. Pain, then, falls into line with the other sensa- tions just by reason of the fact, paradoxical as it may seem, that it has no affective tone.-^ — I pointed out in my first Lecture that stimu- lation of a pain-spot gives qualitatively different 90 GEFUHLSEMPFINDUNGEN sensations, according to the intensity of the stimulus. At a very low intensity we have itch; then prick or sting; and lastly, at higher inten- sities, pain. The observation demands a certain amount of technical skill, and calls for a respon- sive pain-spot ; but I cannot doubt its accuracy. It seems, then, doubly unfortunate that Stumpf has claimed the sensation of itch as a sensation of "pleasure?^ Psychophysically, it is a weak sensation of pain; itch passes into sting, and sting into pain, within the same peripheral organ. Psychologically, itch is not pleasure, but itch; its quality is not pleasurable, but itchy. The psychophysical question was raised by Stumpf only in order to be dismissed, and I do not wish on my side to intrude psy- chophysical considerations into a piece of descriptive psychology. I merely note that "identische Nervenf asern " ^^ presumably medi- ate ' Lustempfindungen ' as well as pain and painless sting. But the psychological argu- ment is important. In a passage already quoted, Stumpf speaks of "das durch Jucken entstehende Gefiihl" and of the "Lustkom- ponente des Kitzels." In both expressions, the pleasure seems to be something additional to, superadded upon, the sense-quality proper. The feeling produced hy itching, or arising PAIN AND UNPLEASANTNESS 91 through itch, is not — if language means any- thing — the itch-quality itself ; and if itch is a special form of pleasurable sensation, there must be some mark or sign upon it to inform us of the fact. In pain, the quality tells its own story; pain is painful. How does itch — which, by the same reasoning, is just itchy — tell us that it is pleasant ? Let us see how the psychologist of affection would read the facts. Itch, he would say, is a sensible quality which is ordinarily attended by pleasantness. Itch passes into sting, which may be weakly pleasant, or indifferent, or slightly unpleasant. Sting passes into pain, which is ordinarily attended by unpleasantness. Under certain circumstances, itch may be unpleasant ; an itch that is widely diffused over the skin, or that persists for a long time, and more especially an itch that is both widely diffused and of long duration, may be distinctly unpleasant. Here, as elsewhere in sensation, space and time may produce the effect of intensity. ^^ Pain is, of course, always painful, in the sense that it always shows the pain-quality; it is painful, that is, just as sting is stinging and itch is itchy. I cannot understand how Stumpf reaches the con- clusion that, for the advocates of an independent affective process, the pain-quality ceases to be 92 GEFUHLSEMPFINDUNGEN pain, becomes schmerzlos. But the pain-quality need not always be unpleasant, — that is the point upon which issue must be taken with Stumpf. It is not easy to find instances, though I think that they are not uncommon in daily life. Ebbinghaus cites the scratching of an irritated area of the skin : ^^ and it is true, in my obser- vation, that if you rub the nails of the right-hand fingers briskly up and down over the back of the left hand you get, particularly when the hand is dry and the skin a little rough, lines of pain that are undeniably pleasant ; though the consequent after-image is sore, and undeniably unpleasant. Kelchner says, apropos of one of her experi- ments : "hier macht Vp. die Angabe, dass das Abklingen des physischen Schmerzes von schwa- cher Lust begleitet sei, — ein Ausspruch, der wieder den Empfindungscharakter des Schmerzes zu bezeugen scheint." ^^ The sensory character of pain is above the need of witnesses ; but the testimony to the possible pleasantness of weakly intensive pain is valuable. Stumpf 's illustration of pain-sensations {i.e. disagreeable sensations) and agreeable sensations, set up simultaneously at different parts of the skin, and enhancing each other by a sort of contrast, is, of course, hypotheti- cal only ; it is not intended to describe an expe- rience of his own. For that reason, and also CUTANEOUS AND ORGANIC PLEASURE 93 because it raises further questions of the dis- tribution of attention and of the nature of the contrast-effect, we need not seriously consider it. Which, now, has made out the better case ? Stumpf, who terms itch a Lustempfindung and pain an Unlustempflndung, or the affective psy- chologist who declares that both itch and pain may be, according to circumstances, pleasant, indifferent, or unpleasant ? — I may pass over Stumpf 's mention of the inter- nal, organic pains. ^^ When we turn to his dis- cussion of the sensations of pleasure, we find the same general difficulty that I have just re- marked in the special instance of itch. *'An- nehmlichkeit, Wohlsein," he says, "[ist] die zweite Hauptqualitat des Gefuhlssinnes." ^^ And he distinguishes, among cutaneous pleasures, tickling, itch, and lust ; among vegetative pleas- ures, satiety, repose, and general comfortable- ness {allgemeines Wohlhehagen). Yet he says that the question of the " Gleichartigkeit aller Lustempfindungen " may be left open ! ^^ Now, if Annehmlichkeit is a sensible quality, there must obviously be different kinds of it ; how do we distinguish itch from tickling, tickling from lust, lust from satiety, and so on, save by their qualitative differences .^ If, on the other hand, all the sense-pleasures are of the same kind, then 94 GEFUHLSEMPFINDUNGEN certainly such diverse things as lust and satiety cannot be exhaustively described as * sense- pleasures ' ; they are that, and they are something that is qualitatively differentiated as well. Is it not clear that, with the best will in the world, Stumpf cannot wholly rid himself of the doctrine that he is combating, — that this doctrine creeps into the argument with all the seeming inevitable- ness of fact ? ^^ Stumpf 's second section deals with the 'affec- tive tone' of the remaining senses, the agreeable- ness and disagreeableness of temperatures, press- ures, odours, tastes, colours, tones. ^ He begins by drawing a valid distinction between very intensive and moderate or weak stimulation. Where the stimulus is very strong, or abnormally strong, he says, it is likely to involve pain-organs, over and above the organ to which it is specif- ically addressed. Temperature-pains and press- ure-pains are pains proper, due to stimulation of the cutaneous pain-spots. The pains of blinding light and of deafening noise are, intro- spectively, of essentially similar character, and may be referred, w ith a high degree of probability, to contraction of the iris and of the muscles of the middle ear.^^ All this we may cheerfully grant ; and its result is that certain experiences, THE SPECIAL SENSES 95 which at first thought would seem to fall within the sphere of pressure or temperature, sight or hearing, are in reality phenomena of cutaneous or organic pain, and thus have already received implicit consideration in the foregoing section. We stand, theoretically, where we stood at the completion of that section. *'Aehnliche Be- trachtungen," Stumpf goes on, *'lassen sich auch iiber die peripherisch durch starke Reizungen erregten Lustempfindungen anstellen." ^^ Here, I think, he falls into that schematism as regards pleasure which is one of the besetting sins of the sensationalists. Are they not all apt to give full details concerning pain, and then to say, offhand, 'Just the same thing holds, mutatis mutandis, of pleasure ' ? The passage must mean that pleasures of touch or temperature, sight or sound, aroused by intensive peripheral stimulation, de- pend for their pleasurableness upon the coexci- tation of the organs of tickling, itch, lust, etc. What is Stumpf thinking of.^ The 'feel' of silk next the skin is exceedingly pleasant, — but largely so, most of us would say, because it does not tickle. And silk is not an intensive stimulus. It is pleasant to come into a warm room when one is chilled with the cold. But while, under such circumstances, we may get circulatory sen- sations of tingling, it is not necessary, in my expe- 96 GEFUHLSEMPFINDUNGEN rience, that we have either tickle or itch. It is pleasant to turn the eyes from the white glare of snow to the dull green of fir, and it is a relief to the ear when a factory whistle ceases to t/ sound. But the eyergreens and the silence are not intensive peripheral stimuli. Perhaps Stumpf has in mind such things as the pleasure of violent bodily exercise and the supreme comfort of a Turkish bath, — organic rather than peripheral pleasures. But where is the analogue of these pleasures in the domain of sight or hearing ? — We come now^ to the crux of the w^hole argu- ment : to the explanation of the agreeableness and disagreeableness of moderate or weak stimulation in the departments of vision and audition, taste and smell. This 'affective tone' Stumpf regards as a 'concomitant sensation' ("das Wort Mitempfindung im weitesten Sinne genommen").^^ Psychologically, a concomitant sensation is, of course, a sensation like any other, an elementary mental process with a cer- tain status in consciousness and a certain set of attributes. We should therefore expect, since on Stumpf's own admission there is no such thing, in strictness, as indissoluble association,^^ that he would at once cast about for instances of dissociation, and would seek to show us the THE SPECIAL SENSES 97 'affective sensation' in isolation. Instead of doing this, lie offers us reasons for shifting the scene of our debate from sensation to idea. Why is it, he asks, that we can not sense the agreeableness of a colour or a scent alone, without being obliged at the same time to sense the colour as visual and the scent as olfactory quality ? ^' Well ! Stumpf replies, there might be anatomi- cal reasons. It might be a matter of physio- logical fact that the excitation underlying agree- ableness cannot be set up independently of the excitation underlying colour or odour. And after all, it was only the other day that physiologists succeeded in separately stimulating the cutane- ous pain-organs ; here too, then, the future may bring results that we cannot now foresee. Or, again, we might think that colour and the agreeableness of colour are intimately fused, as are, e.g., taste and smell, or the two tones of the octave. Or, lastly, it is possible, even prob- able, that the concomitant sensations of agree- ableness and disagreeableness are sensations of central origin. In that case, we can separate them from their companions, if at all, only by change of central conditions, not by modifica- tion of peripheral stimulation. I would call your attention to the fact that two 98 GEFUHLSEMPFINDUNGEN of these arguments are psychophysical, and that the third, if it is not psychophysical, is irrelevant. To talk of anatomical reasons for the conjunc- tion of colour with agreeableness of colour, and to talk of that agreeableness as a centrally excited concomitant sensation, — to talk in this way is to leave the field of descriptive psychology for the field of psychophysics. And it seems to me, in general, that Stumpf is inclined to bar out psychophysical reference where it does not sup- port his views, and to bring it in, without apology, wherever it can furnish him with even a specu- lative confirmation/^ I know very well that sweeping criticism of this kind is likely to be both unfair and ineffective. I have, however, too much respect for Stumpf to be consciously un- fair, and too serious a concern for my own posi- tion to be consciously ineffective. My general impression is as I have stated ; and I believe that, if you read Stumpf's paper for yourselves, you will come to the same conclusion. At all events, the appeal in these two cases lies, frankly, to psychophysics. So it does also in the third argument, if that is relevant. Psychologically, the fusion of the octave, under the most favour- able conditions, is analysable into its two con- stituent tones : here, then, is no analogy. Psy- chologically, the blends of taste and smell are THE SPECIAL SENSES 99 not analysable; the most experienced psycholo- gist cannot tell, by introspection, that the 'taste' of his coffee is partly taste and partly smell. Only by psychophysical procedure — by hold- ing the nose, or what not — can the components in the blend be separated. In other words, it is only as subject-matter of psychophysics that the taste-smell blend may be termed a fusion. And the analogy that it affords to the fusion of colour and agreeableness of colour is, therefore, a psychophysical analogy. There remains the possibility that we may some day isolate the pleasure-organs of vision as we have already iso- lated the pain-organs of the skin. How serious that is, I leave you to judge for yourselves. My objection is not by any means to psycho- physics as such. I do object, however, to the basing of a psychological argument upon a speculative psychophysics. And we have a peculiar right to object, in the present instance, because Stumpf promised us a descriptive psy- chology. " Wir wollen nicht," he says, ** Behaupt- ungen tiber die anatomischen Gebilde oder die physiologischen Prozesse aufstellen, die den sinnlichen Gefuhlen zugrunde liegen."^^ What else, then, is he now doing ? His text stands as I have given it ; psychological considerations are relegated to a footnote. In the text, Stumpf has 100 GEFUHLSEMPFINDUNGEN accepted the inseparability of colour and agree- ableness of colour as a fact ; in the note, he cites four cases which appear to tell against the fact, — though he himself practically reduces them to two. Scripture infers, from his experiments on association of ideas, that feeling may stand alone in consciousness ; but the results, we are told, admit of very various interpretations. Vogt reaches the same conclusion by way of his method of hypnotic suggestion; '*uber die Brauchbar- keit dieser Methode," says Stumpf, '*habe ich kein Urteil." There remain Kiesow's investi- gation of the taste-feelings, in which sensible was paralleled by affective discrimination, and Stumpf's own work with beating and dissonant tones. Kiesow, however, was simply attempt- ing a quantitative form of the method of im- pression ; and Stumpf's experiences simply illus- trate the occurrence of affective habituation or adaptation. ^^ We are still only on the threshold of the dis- cussion. The essential thing in Stumpf's view, you remember, is not that colour and agreeable- ness of colour should be separable in sensation, but that they should be separable in idea. He clearly sees — no doubt, he saw very early in the course of this inquiry — that the agreeable- ness and disagreeableness of tone and colour, IS THERE AN AFFECTIVE IMAGE? 101 taste and smell, cannot possibly be constituted a class of sensations in the ordinary meaning of that term, — cannot possibly be put on a level with pain and tickle and itch. The psycho- physical hypotheses which I have just been criticising are therefore introduced to explain how two mental processes, possibly separate in idea, may be altogether inseparable in sensa- tion. If the explanation is accepted, if we waive the objection that something which is termed a sensation cannot be separately sensed, then we are free to enter upon the argument which will lead us through separateness in idea to the theory of central concomitance. You see, I hope, how pivotal those psychophysical hypotheses were, although Stumpf brings them in as it were paren- thetically, by way of excursus. A little specula- tive physiology — and we are prepared to revise our definition of sensation, and to look for proof of sensory character in the realm of ideas ! 3^his whole question of ideation, or (to put it in more elementary terms) of the existence of an affective image, is very thorny. More than a decade ago I argued, as against Eibot, that there is no such thing as affective reproduction, but only affective renewal or revival.^^ I should argue to the same effect to-day, though with greater caution in statement and with less as- 102 GEFUHLSEMPFINDUNGEN surance of carrying conviction. For before the question can be settled, we must, I think, know a great deal more than we do of sense-imagery, and particularly of the range of images of organic sensations. In an early number of Wood- bridge's Journal I gave an account of my own organic images, such as they are; expressed myself as rather sceptical of any great freedom or variety of organic imagery in general; and urged the importance of further work.^^ Let me repeat here that further work — systematic observa- tion by competent observers — is badly needed. Stumpf, now, takes a positive attitude, while he admits the fact of individual differences. "Die Schmerz- und Lustempfindungen, die durch Hautreizungen oder durch die Tatigkeit der vegetativen Organe bedingt sind, hinterlassen zweifellos auch Gedachtnisbilder, blosse Vorstel- lungen." ^^ He is apparently speaking from personal experience, since he says later: "auch mir^scheint z. B. die Vorstellung eines Stich- schmerzes moglich, und zwar mit dem Charak- ter einer reproduzierten Vorstellung in dem- selben Sinne, wie wir von Farben- und Tonvor- stellung reden." ^^ I wish that we had been given more details. The passage from which the first quotation is taken goes on to point out the occurrence of hallucinations of pain, — another IS THERE AN AFFECTIVE IMAGE? 103 subject, surely, which myites psychological in- vestigation. The paragraph ends, rather curi- ously, with the words, ''ahnliches auch bei den Vorstellungen der Wolllistigen." I say 'curi- ously,' because the words ought to mean that the voluptuary has images, or even hallucinations, of his own lust-sensations, whereas it seems obvious that the images must be images of some voluptu- ous situation, and that lust itself is present, not as image, but as sensation. On the topic of what I should call the affective image proper, the image or reproduced idea of the agreeableness and disagreeableness that attach to scents and colours and tones, Stumpf's attitude is similarly positive. He speaks again from personal experience; the affective image which accompanies the memory-image of a major triad or of a Bocklin picture seems to him to be distinct and vivid. ^^ Since he says on the same page, "nattirlich miissen die Falle, in denen offenbar Denktatigkeiten und Affekte mit im Spiele sind, wie das Wohlgefallen an einer Melodic oder einem Bildwerk, beiseite bleiben," we must suppose that he is, in reality, thinking of the constituent tones in the chord and the con- stituent colours in the picture, not of the chord as harmony and the picture as work of art. He further remarks that these affective images "sehr 104 GEFUHLSEMPFINDUNGEN leicht in Gefiihlsempfindungen iibergehen" ; that is, that we are ordinarily liable in their case to affective hallucinations, such as occur only occa- sionally in the case, e.g., of peripheral pain. If you ideate the sound of a friend's voice, or of a musical chord, the affective quality may be as vivid as in actual hearing, while the tonal quality has all the marks of a representation. There are, then, images of the agreeableness and disagreeableness which come with the sensa- tions of the higher senses. They are fleeting; to illustrate them, Stumpf has to cite complex ideas like those of a chord or a picture : ^^ but they exist. Now, then, we may raise the critical question, and ask whether colour image and im- age of agreeableness are separable, — whether, as we put it just now, a colour and its agreeableness are separate in idea. We need not expect, Stumpf says, that the separation will be easy. Think of odours : the people w^ho have smell-images can rarely evoke them without at the same time evoking the memory picture of the flower or fruit or what- ever it is that the scent connects with in sensation. "Ich selbst kann mir u. a. den Heliotropgeruch gut vorstellen, aber nur unter dieser Bedingung." The separation, then, will be difficult: only, it should not be impossible. ^^ COLOURS AND TONES 105 Let us look at the senses in order. For most men, the single colour and the single tone are, in sensation, practically indifferent. "Es wird uns oft recht schwer, zu sagen, ob eine Farbe mehr angenehm oder mehr unangenehm ist, oder ob sie angenehmer ist als eine andere." ^^ Indi- vidual organisation and temporary disposition may afford exceptions to this rule: *'aber im ganzen sind die rein sinnlichen Geflihlswirkungen isolierter Farben (einschliesslich der Graunu- ancen) und isolierter Tone (einschliesslich der Gerausche) relativ gering." I think that Stumpf , even with the allowances that he makes, is here arguing too schematically. No doubt, a patch of red on the book-shelf and the sound of middle C from the piano, as they break into our everyday consciousness, leave us "most uncommon calm." But many men — Wundt is an example, among psychologists — are extraordinarily sensitive to the affective value of single colours and single tones ; and one is surprised that Stumpf himself should so lightly brush aside the Tongefilhle. Again, when colours and tones are presented methodically, as by the serial method or the method of paired comparisons, it is the ex- ception that they are indifferent; the rule is definitely the other w^ay. Perhaps, however, this is what Stumpf has in mind, when he 106 GEFUHLSEMPFINDUNGEN refers to " augenblickliehe Nervendisposition," as he may have Wundt and himself in mind when he refers to **individuelle Organisation." Take, then, more obvious cases. If isolated patches of colour and isolated tones are usually indiffer- ent, what of masses of colour: the carmine or purple or orange sweep of an uniformly coloured sunset sky ? the high blue of a tropical sea, the white of extended snow, the yellow-brown of a sandy shore, the dull lead or slate of lake or ocean ? I am not thinking of landscape and sea- scape, of emotion and reflection, but of the col- ours themselves, given sensibly to the eye. And what of certain noises : harsh, rough, grating, scraping, crunching, sickening noises ? Surely, there is case upon case, instance upon instance, in which colours and sounds possess a high degree of agreeableness or disagreeableness. Stumpf ignores them all, and concludes that the affective tone is too weak, in sensation, to come separately into consciousness as idea. But — cannot weak sensations, then, be ideated ? Fechner believed and Ebbinghaus believes that you can ideate the just noticeable sensation and the just noticeable difference between sensations ; ^^ is there any- thing weaker.? And apart from that, which is a technical matter, is it not a fact of daily expe- rience that weak sensations may be imaged ? TASTE AND SMELL 107 Think now of a diminuendo on the violin, of the faint anticipatory glow of a rising moon, of the suspicion of a breath of garlic in a savoury salad : if you have images in these sense-departments at all, you will have no difficulty in imaging such weakest sensations. Stumpf , nevertheless, has disposed of sight and hearing. He turns to taste and smell, and first of all quotes Nagel's very definite statement with regard to smell. "So leicht es mir ist," we read in the Handbuch, *' das mit einer Geruchsemp- findung verbundene Lust- oder Unlustgefuhl zu reproduzieren, so unmoglich ist es bezuglich der eigentlichen Geruchsqualitat." ^^ The state- ment is definite, but none the less ambiguous ; for the term 'reproduction' may cover a multi- tude of possible experiences. Stumpf finds, by personal inquiry, that NageFs ' reproductions ' are always mediated by association ; Nagel can call up, e.g., the agreeableness of the smell of tar, without reproducing that odour itself, but the pleasantness appears to depend *'auf den, wenn auch unbewussten, Nachwirkungen von Schif- fahrtserlebnissen." ^^ The revised statement is not wholly clear ; but Stumpf concludes that the feeling in question is rather a mood, a Stim- mung, than an elementary Gefiihlsempfindung, and so rules it out of the discussion. Nagel's 108 GEFUHLSEMPFINDUNGEN reproductions of the disagreeableness of odours are always connected with other images (the disagreeableness of ammonia, e.g., with prick- ing and stinging in the nose), so that they furnish no proof of separability in idea. And notice that, all through, we have no strict guarantee of the arousal of an affective image; Nagel's * re- productions' may very well be renewals, rein- statements of feeling, — what Stumpf calls * hal- lucinations.' That is all that we hear about smell. As for taste, Stumpf suggests that the mere sight or name of oysters may arouse in the epicure "einen Anflug des korperlichen Wohlbehagens, das sonst mit dem Genuss verkniipft ist, ohne dass der Geschmack selbst ihm zum Be- wusstsein kommt." ^^ It might, of course, be objected that the feeling in this case is not sepa- rately ideated ; it is ideated along with visual sen- sations, the sight of the shell-fish or of the printed word. Stumpf replies, and from his standpoint justly, that at all events the feeling is ideated separately from the taste, and that that is the important point. ^^ However, he invalidates his own example in a way that, I confess, I should not have thought of: "es ist mir nicht sicher," he says, "dass das namliche Gefiihl, wie es an die Geschmacksempfindung oder Geschmacks- FEELING OR EMOTION? 109 vorstellung geknlipft ist, auch die blosse Gesichts- vorstellung begleitet"; " he is not sure that the affective image aroused by sight is the same affec- tive image that accompanies taste. My own objection would be, again, that there is no evi- dence of affective image, but only of affective reinstatement. Be this as it may, we hear no more of taste. In his eagerness for further instances, Stumpf quotes the colour-feelings aroused in the artist by the sight of an etching, the feelings which attach to poetic expressions, and various expe- riences of himself and of his co-workers in the sphere of tones, — feelings accompanying the sight of musical phrases and harmonies and rhythms, the effort to recall a modulation, the mere thought of the tone-colour of different pianos. ^^ Now it is perfectly clear that in most of these cases we are dealing, not with the affec- tive tone of sensation, but with something much more complicated, — with AffeJd or Gemilts- bewegung. Stumpf, it is true, avers that the look of a " nichtswlirdige Tonverbindung" gives him "einen Stich" of sensory disagreeableness, and that the look of *'langgelialtene konsonante Akkorde" affects him in somewhat the same way as a warm bath. But it is, surely, very difficult to think that we are here in presence of anything 110 GEFUHLSEMPFINDUNGEN else than aesthetic feelings, — feelings that, from long and expert familiarity with the subject- matter, have the promptness and immediacy of sense-feelings, but that nevertheless are in origin sesthetic. How can you tell, by eye, that a Ton- verbindung is nichtswilrdig , save as the result of musical training? The T onverbindung is not a sensation, and no amount of practice can make it a sensation. And how can you recognise long-drawn consonant chords in the musical score ? Not by sensation. Stumpf , however, is very far from dogmatic. "Alle diese Beobach- tungen fiihre ich mit einer gewissen Reserve an, da in Fallen, wo die Tonvorstellung nicht merk- lich ist, doch auch das sinnliche Gefiihl meist nur fliichtig und schwer zu fassen ist," and contrari- wise.^^ Affective image, that is, is at least very largely a function of tonal image ; clear-cut separation of the two is doubtful. I conclude, then, — I have no choice but to conclude, — that the proposed demonstration of the separateness, in idea, of sensation and sense- feeling has broken down. There is no atom of reliable evidence. Remember that the refusal to consider the 'higher' feelings, the rigorous restriction of the argument to the isolated, single sensation, are Stumpf 's refusal and Stumpf 's restriction, not the critic's. Stumpf marked out THE DUALISTIC THEORY 111 his own ground; and though, in my judgment, he has more than once shifted his position,^^ he finds himself obliged to retire. He retires, however, with a very clever riposte. There are psychologists, he says, — Kulpe is one of them, — who posit a simple duality of feeling, a single quality of pleasantness and a single quality of unpleasantness. Now we have seen that cu- taneous pain, which is unpleasantness, may be isolated and imaged ; and we have seen that the cutaneous pleasure-sensations may be isolated and imaged. Ergo, these psychologists must admit, in general, the possible occurrence of the separate affective image as of the separate affec- tion itself ; when they reproduce the unpleasant- ness of a bad smell, they have an image of cutaneous pain ; and if they wish to know, pre- cisely, what the unpleasantness of a bad smell is, per se, they have only to isolate a cutaneous pain-sensation.^^ It is needless to work out the reply. Pain and itch, for these psychologists as for Stumpf himself, are sensations. Only, for that very reason, they are not affections. Stumpf has covered his retreat ; but we must not let ourselves be blinded to the essential thing, — the fact that he has retreated. Now for the rally ! Stumpf rallies — on what ? on descriptive psychology ? Not at all ! — on 112 GEFUHLSEMPFINDUNGEN psychophysics. Our principle of scientific econ- omy forbade us to make a separate class for the affective elements unless the facts positively constrained us to that conclusion. Now let us grant that the agreeableness of colours and tones and scents is separable from the colours and tones and scents themselves neither in sensation nor in idea. We are not really forced to the admis- sion ; but let us bow to the equivocal nature of the evidence, and make it. Are we thereby con- strained to recognise the independence of the affective process.^ Surely not: the purporting affective element may still be a concomitant or accessory sensation of central origin.^" The central physiological mechanism may be of such a kind that the excitation of colour-feeling necessarily implies the coexcitation of colour- sensation. This view becomes, indeed, physio- logically probable if we assume, as many do, that the feeling-qualities of any one sense-de- partment are different from the feeling-qualities of the rest, that the colour-feelings and the tone- feelings are qualitatively distinct. Ebbinghaus, who places the affective elements in a class of their own, nevertheless regards them, physio- logically, as "Nebenwirkungen derselben Ur- sachen, die den begleitenden Empfindungcn und Vorstellungen zugrunde liegen." ^^ It is but SUMMARY 113 a short step from this to the view that makes them sensations, ''zentrale Mitempfindungen." To appreciate this final stand, we must look back over Stumpf s whole essay. He started out on a question of descriptive psychology; we were to hear nothing of genetic psychology or of psychophysics. He began by examining three of the alleged criteria of affection — the three that he himself took to be the most note- worthy — and found them wanting. He then turned to the consideration of the sense-feelings in detail ; dealing first with pain, and the cu- taneous and vegetative pleasures, and secondly with the *aftective tone' of the remaining senses. He had no difficulty in showing that pain, itch, tickle, lust, and so forth are sensational in character, — though, as I pointed out, his inter- pretation leaves many facts out of account (the varying pleasantness and unpleasantness of itch, the possible pleasantness of pain, etc.). He had no particular difficulty, on the side of pain, with the affective tone of intensive sensations of the other senses, — though I showed that there were distinct difficulties on the side of pleasure. His real difficulty, the difficulty which he himself feels and acknowledges, arose in connection with the affective tone of moderately and weakly intensive sensations of sight and sound, taste 114 GEFUHLSEMPFINDUNGEN and smell. And this difficulty, insurmountable on the plane of descriptive psychology, is twice avoided by appeal to speculative psychophysics. Now, then, if I ask you: 'What is Stumpf's psychological evidence for the existence of a class of Gefuhlsevipfi7idungen which shall replace the affective elements of current psychology?' — what have you to reply ? Stumpf offers his whole article in evidence. I grant that the article is subjectively persuasive and objectively im- portant; otherwise I should not have devoted this hour to its criticism. But I affirm that, when critically reviewed, it contains no stronger evi- dence than the principle of economy and the demonstration that, as our knowledge of nerve- physiology goes, the existence of centrally excited accessory sensations is a psychophysical possi- bility. The persuasiveness of the essay, then, I discount altogether. Its objective importance lies, I think, not in what it has shown, but in the example which it has set. Stumpf lends the weight of his name to a sensationalistic theory of affection ; and we may expect in the near future, both from adherents and from opponents of that theory, an industrious collection of psychological facts, psychological observations, which will finally sway the balance in the one direction or in the other. APPLICATIONS 115 However, the proof of the pudding is in the eating. Stumpf adds a final section, in which he deals with 'applications' of his psychophysical theory.^ In the domain of sensation we have, he says, well-developed methods and well-estab- lished results. If, then, we can only bring our- selves to look upon affections as sensations, we can attack them directly by sensation-methods, and can check or control our data by sensation- results. In particular, we may expect by this means to bring light into the dark places of genetic psychology.^ The point is well taken. If, as a matter of fact, the theory of central concomitant sensations helps us to a stable affective psychology, then let us welcome it gladly, without waiting to ask whether its foundation is in psychology or in physiology, and whether its author has or has not adduced at the outset valid arguments in its favour. A good working hypothesis is valuable for its own sake, and the facts whose discovery it assures soon become strong enough to furnish the required corrective. Stumpf, now, devotes seven pages — one-seventh of his whole paper — to the test of the theory from this point of view. Let us see what the outcome is. In the first place, Stumpf points out that the theory accounts for the various analgesias, for 116 GEFUHLSEMPFINDUNGEN the cases of anaesthesia in which pain-sensation persists, and for the occurrence of * unnoticeable ' sensations of pain and pleasure, — sensations which are above the limen of stimuhis but below the limen of attention.^® As we agree that pain, itch, tickle, etc., are sensations, we have no quarrel with him on their behalf. It need only be said that his theory has, here, no advantage of any kind over the orthodox affectional theory. Secondly, Stumpf instances the fact of 'indif- ferent ' sensations. An affectional theory, he says, has, by a sort of a ^priori necessity, to postulate the presence of affective process in conscious- ness, even where introspection is unable to dis- cover it, — witness Lotze's doctrine of the " All- gegenwart der Gefuhle." If, on the other hand, that process is an accessory sensation, "so liegt nicht der mindeste Grund vor, warum eine solche Begleitung vollig allgemein und ausnahmslos den Empfindungen zukommen mlisse"; while we can readily understand how it comes about that extensive and intensive stimulation of any sort brings the accessory sensation into consciousness .^^ This argument, however, is unconvincing. An affectional theory is a theory of affective facts ; and the fact that some sensations are indifferent is ordinarily explained by reference either to habituation or to insufficient intensity of stimulus. APPLICATIONS 117 It seems to me that there is small choice, on this topic, between the opposing views. One might argue, against Stumpf, that a concomitant sensa- tion which is rarely if ever isolable in conscious- ness, which can hardly be separated from its companion, whether as sensation or as image, ought a priori always to accompany its pair; and, indeed, I am inclined to think that this objection is stronger than that which Stumpf urges from the other side. Lotze's view is in- trospectively grounded, and may, perhaps, have been due to " individuelle Organisation."^^ Thirdly, Stumpf declares that his theory accords better than its rival with the facts of the dependence of affective tone upon the quality of sensation, as set forth, e.g., by Ebbinghaus.^^ I might reply that this dependence is itself in dispute ; Kiilpe, e.g., denies it.'^° All that Stumpf could then assert would be that his theory accords with a particular view entertained by Ebbing- haus, — a view, be it remembered, which Ebbing- haus himself regards as compatible with an affec- tional theory. If, however, this is unduly to press the sense of the term 'dependence,' — though Ebbinghaus heads his section, * Die seelischen Geflihlsursachen " ! ^^ — and if Stumpf has in mind simply the factual connections of sensation and affection, then I do not see, and 118 GEFUHLSEMPFINDUNGEN Stumpf makes no effort to show, in what way his theory is superior. The facts are "eine weitge- hende Konstanz, . . . aber doch wieder Aus- nahmen von dieser Regehnassigkeit." To say: '*die Auffassung der Sinnesgefiihle als Mitemp- findungen . . . fiigt sich diesen Bedingungen ohne weiteres," as if there were something in the very nature of concomitant sensations that indi- cated a rule with salient exceptions, is of little avail ; we must know in detail the conditions of the rule and the conditions of the exceptions. The argumentation on these two points is very much 'in the air.' It may well be the case that Stumpf has thought out his position with all necessary fulness, and that as he writes a crowd of confirmatory associations press in upon him. But the statements actually made are schematic to a degree. I have tried, working on the hints which Stumpf has earlier given, to think out a physiological mechanism that should behave, naturally and normally, as the substrate of the concomitant sensations is required to behave : but the deeper I go, the farther do I seem to travel from anything like our current conception of the substrate of sensation. There is a fourth point. Stumpf thinks that his theory brings us nearer than any other to an understanding of the vast and unsettled prob- APPLICATIONS 119 lems of affective genesis, of the individual de- velopment and generic evolution of the sense- feelings, — and also of the related problem of the striking diversity of affective reaction to the same stimulus. "^^ He illustrates this thesis by reference to the senses of taste ^^ and hearing. ^^ As the psychology of tone is Stumpf's special field, and as he gives more space to hearing than to taste, I may confine myself to his discussion of the tonal feelings. We have at our disposal a mass of facts from history, from individual psychology, and from ethnography. We have also a number of facts gained from psychological experimentation : **the existence and the peculiar character of the * feel- ing of purity' with consonant intervals, the shift of this feeling within certain limits under the influence of sesthetic and other motives, as well as its dependence upon recently formed habits; the great secular changes as regards the pleasant- ness of consonances at large, the origin and de- velopment of the modern feeling of harmony, the possibility of its temporary annulment by intensive occupation with divergent tone-struc- tures." ^^ Now, Stumpf says, the theory of concomitant sensation gives us the right atti- tude to all these facts. For brief periods of time, the same stimuli will evoke a constant affective 120 GEFUHLSEMPFINDUNGEN reaction ; but in long periods of time the sense- feelings are exposed to transforming influences, both of an individual and of a generic sort, more especially to the influence of habitual direction of attention, of disposition of judgment, of habits of all kinds. Let such factors operate through generations, and we may have inheritance, con- nate peculiarities of feeling-effect. These pecu- liarities are perhaps, in some instances, residua of sense-feelings which originally appeared in connection with emotions; and Stumpf here broaches a subject of fascinating interest, — the possibility of expert reconstruction of those emotions themselves. " Die Ausfuhrung muss freilich einer anderen Gelegenheit vorbehalten bleiben." '' Here I am less inclined to criticise than to re- gret. We are given a skeleton, an outline — less than that : a bare suggestion of Stumpf 's doctrine of the *Ton- und Musikgefiihle.' '^ Would that we had the completed work ! Until that appears, it is hopeless to argue under this heading, whether for the affective element or for the concomitant sensation. The paper ends with a brief comment on the inadequacy of the teleological principle as a principle of explanation.'^ On this matter I not only agree with Stumpf, but I should even APPLICATIONS 121 be inclined to go farther, and to rule the teleo- logical principle out of affective psychology altogether. — Has, then, this section on 'applications' shaken our previous conclusion ? My own feeling is that Stumpf's presentation would have been stronger without it. These brief and summary statements read like the formularies of a faith; their dog- matism stands in marked contrast to the careful and elaborate argument that has gone before. On the evidence, we must still say that the theory of concomitant sensation, as a psycho- logical theory, has little to commend it. When the evidence is all in, and the explanatory power of the theory has been tested along the whole line of observed fact, then I, for one, shall be ready to revise, and if necessary to reverse, this judgment. May the day come quickly that brings us the long-delayed volume on the Tonge- filhle ! IV THE TRIDIMENSIONAL THEORY OF FEELING LECTURE IV THE TRIDIMENSIONAL THEORY OF FEELING * A LECTURER who had expounded Wundt's elementary doctrine of feeling in the year of grace 1893 would have called attention to two principal points : the status of feeling in con- sciousness, and the number and nature of the affective qualities. Feeling, Wundt says in the fourth edition of the Physiologische Psychologies is a third attribute of sensation, "eine dritte Eigenschaft der Empfindung." "Neben In- tensitat und Qualitat begegnet uns mehr oder minder ausgepragt in jeder Empfindung ein drittes Element. . . . Wir nennen diesen drit- ten Bestandtheil der Empfindung den GefiXhlston oder das sinnliche Gefilhl." And feeling or affective tone ranges between qualitative oppo- sites, which "wir als Lust- und Uiilustgefilhle bezeichnen." Pleasantness and unpleasantness are the ultimate simple forms of sense-feeling, the irreducible qualities of the pure affective tone which is immanent in the simple sensation. At the same time, the terms * pleasantness ' and * un- * This Lecture has been prmted in the American Journal of Psychology, April, 1908. 125 126 TRIDIMENSIONAL THEORY OF FEELING pleasantness' are not adequate to describe the affective tone of any and every sensation that we obtain by psychological analysis. The qualities of the higher senses, sight and hearing, play an important part in the compound ideas which appeal to the aesthetic side of our nature. Prob- ably for this reason, their affective colouring is approximately, anndhernd, identical with that of such compound ideas; they have taken on a Stimmungscharakter, "der nicht mehr schlecht- hin auf Lust und Unlust zurlickgeflihrt werden kann, sondern in andern, in gewissen x\ffecten deutlicher ausgepragten Gegensatzen einen ada- quateren Ausdruck findet." Tones, e.g., may be grave or cheerful, colours may be calming or exciting. The passage from pure affective tone, pleasantness or unpleasantness, to these sesthetic, emotional shades of feeling may be traced through the series of the senses. Touch and the common sensations show pleasantness- unpleasantness with only a trace of "qualitative Farbung"; tastes and smells are predomi- nantly pleasant or unpleasant, but nevertheless ad- mit of **verschiedenartigere Gefuhlsfarbungen." Tones and colours, which are strongly pleasant or unpleasant to children and savages, have al- most lost these attributes for the civilised adult, — though even for us the seriousness of deep THE THEORY OF 1893 127 tones and of black surfaces leans towards un- pleasantness, and the excitement of high tones and of white towards pleasantness, — and have assumed an affective colouring whose general affinity to pleasantness-unpleasantness is, in extreme cases, proved only by its movement between qualitative opposites/ That, then, was Wundt's doctrine, taken at the purely descriptive level : sensations with an immanent attribute of pleasantness-unpleasant- ness, the original simplicity of which appears clearly enough in the lower sense-departments, but in the higher is obscured by aesthetic or quasi- sesthetic reference. Now suppose that, as the novelists say, three years have elapsed, and that the same lecturer is discussing the same subject in 1896. He has in his hands the first edition of Wundt's Grund- riss der Psychologie, And there he reads of *'zwei Arten psychischer Elemente, die sich als Producte der psychologischen Analyse ergeben, . . . Empfindungselemente oder Empfindungen [und] Gefiihlselemente oder einfache Geflihle." The constitutive attributes (" unerlassliche Be- st immungsstiicke") of sensation are quality and intensity. Affection, too, possesses these attri- butes. But there is a difference. While sensible qualities are limited by maximal differences. 128 TRIDIMENSIONAL THEORY OF FEELING affective qualities range between maximal oppo- sites. While the number of sensible qualities is fixed, by the differentiation of the sense-organ, the number of affective qualities is indefinitely large; for simple feelings are the subjective complements, not only of simple sensations, but also of compound ideas and of still more complicated ideational processes. And while sensations fall into a number of separate systems, there is but one affective system ; tone and colour, warmth and pressure, are disparate, but "alle einfachen Geflihle bilden eine einzige zusammen- hangende Mannigfaltigkeit, insofern es kein Gefiihl gibt, von dem aus man nicht durch Zwischenstufen und Indifferenzzonen zu irgend einem andern Gefiihle gelangen konnte." Do, then, all these many affective elements fall within *'dem allgemeinen Rahmen einfacher Lust und Unlust" ? By no means ! There are three HauptricJitungen der Gefiihle, three di- mensional categories, "innerhalb deren unend- lich viele einfache Qualitaten vorkommen." These are pleasantness-unpleasantness, excite- ment-inhibition or excitement-tranquillisation, and tension-relaxation. As a rule, Wundt says, psychologists have paid regard only to pleasant- ness and unpleasantness, and have relegated the other two affective classes to the emotions. THE THEORY OF 1896 129 But as emotions arise from the combination of feelings, the fundamental types of emotion must be preformed, vorgebildet, in the affective ele- ments.^ In cases like this, I always want to trace the motive. Like the lawyer in David Copper- Jield, I assume that in all such cases there is a motive. Wliat was it, then, that led Wundt to his change of opinion ? If my reading of Wundt is correct, the changes that he has made, from time to time, in his various systematic works have never been due, in any real way, to external causes, but have always represented the climax or culmination of a stage of internal development. The germs of the changes are invariably, I think, to be found in the prior Wundt, and the changes themselves are but the full and self-conscious maturity of ideas that had long been 'incubated,' had long been held in the obscure margin of consciousness. On the other hand, it is possible, at least in most cases, to point with a fair degree of probability to the external cause that brought these obscure ideas to the attentive focus. In the present in- stance, that external cause appears, very obvi- ously, in the publication of Klilpe's Gru7idriss, Let me be clear on this matter, even if I am repe- 130 TRIDIMENSIONAL THEORY OF FEELING titious ! I believe that Wundt would have for- mulated his new affective theory in any event ; the theory was implicit in him and in his pre- vious writings. If Kiilpe had not given the touch that led to crystallisation, some one else would, sooner or later, have performed the same office. In fact, however, Kiilpe undoubtedly did furnish the external stimulus, — so that, indeed, we have to thank him, not only for his own Grundriss, but in a certain special and limited sense for Wundt 's as well.^ Let me take you, now, to the first edition of the Physiologische Psychologies the edition of 1874. In general, the exposition is very like the exposi- tion of 1893. But, in 1893, we are told that the affective tone of sensations of the higher senses is a Stimmungscharakter, a colouring that they have 'taken on' in virtue of their constant par- ticipation in aesthetic ideas. In 1874, the ref- erence to aesthetics comes at the end of the dis- cussion; the fact that sight and hearing have freed themselves of sense-pleasurableness and sense-unpleasurableness fits them to serve as elements in sesthetic effect. They are not grave and dignified and happy and cheerful because they have been aesthetically employed, but their gravity and cheerfulness are what enables us to employ them with aesthetic result. **Lust und THE THEORY OF 1874 131 TJnlust," Wundt concludes, ''sind, wie es scheint, nur die von der Intensitdt der Empfindung her- riihrenden Bestimmungen, wahrend an die Quali- tdten Gegensatze anderer Art gekniipft sind, welche zwar zuweilen in eine gewisse Analogic mit Lust und Unlust sich bringen lassen, an sich aber doch von diesen letzteren nicht beriihrt werden." Here is the doctrine of the plurality of affective dimensions plainer and more definite than it was twenty years later ; here is, evidently enough, the germ of the doctrine of 1896.* Once more : the chapter from which I have been quoting is entitled, in 1893, *'Gefuhlston der Empfindung," — in 1874, "Sinnliche Ge- flihle." Is not that significant also ? Affection, in 1874, is not an attribute of sensation; it ap- pears in that role for the first time in 1880. Affection, in 1874, is a relation, the relation which sensation sustains to consciousness at large. **Als ein nach Qualitat und Intensitat bestimmter Zustand ist die Empfindung nur im Bewusstsein gegeben ; in Wirklichkeit existirt sie daher auch immer nur in ihrer Beziehung zu demselben. Diese Beziehung nennen wir das sinnliche Gefiihl." ^ Clearly, then, the whole of the new affective theory is implicit in the original edition of Wundt's great work. So far from suddenly 132 TRIDIMENSIONAL THEORY OF FEELING GROSSE KLANGSTARKE KJange mit tiefen Obertunen. Klange mit tiefen und Klange mit hohen ObertOnen. hohen Obertonen. Tiefe Tone. Hohe Tone. Klange mit tiefen Obertonen. Klange mit tiefen und hoiien Obertonen. Klange mit hohen Obertonen. GERINGE KLANGSTARKE Fig. 3. Wundt's Schema of the 'System der Klanggefiihle' : Physi- ologische Psychologie, 1874, 446. "Jedem dieser Ton- und Kianggegen- siitze entsprechen Contraste des Gefiihls, die alhnahg durch vermit- telnde Z^dschenstufen einem Indifferenzpunkt sich niihem, durch welchen sie in einander iibergehen. Den tiefen Tonen und Klangfarben zur hnken Seite entsprechen die ernsten, den hohen zur rechten die heiteren Stimmungen, bei grosserer Klangstarke sind alle Stimmungen mit einem gehobenen, energischen, bei geringerer Klangstarke mit einem gedampften, sanften Gefiihlston verbunden. Da zwischen den hier herausgegriffenen Strahlen alle moghchen L^ebergange sich denken lassen, so kann man sich vorstellen, alle durch die Klangfarbe bestimmten Gefiihlstone seien in einer Ebene angeordnet, deren eine Dimension, dem Continuum der einfachen Tone entsprechend, die Contraste von Ernst imd Heiterkeit mit ihren Uebergangsstufen cnthalte, wiihrend die zweite, welche die Starke der Theiltone abmisst, die Gegensatze des Energischen und Sanften vermittelt." — Cf. ihid., ii., 1902, 327 f. THE THREE DIMENSIONS 133 reversing his attitude to affective processes, he has, in reality, returned to his first systematic position. In other words, the problem with regard to Wundt is not so much that he now makes affection an independent element with a plurality of dimensions and qualities, as rather that he ever did anything else. This problem, too, can be solved ; but it is foreign to our pres- ent consideration. We are to examine, in this Lecture, the theory which I briefly outlined a moment ago on the basis of the Grundriss of 1896. The theory has been widely and variously discussed, and I can- not attempt to cover the whole of the relevant 'literature.' I shall refer, for the most part, to the earliest statements of it, in the Grundriss of 1896 and the Vorlesungen of 1897, and to the latest systematic statement in the Physiologische Psychologie of 1902. First of all, then, how does AVundt arrive at his three affective dimensions ? How does he prove that there are three, and that these three are pleasantness-unpleasantness, excitement-in- hibition, and tension-relaxation ? Well ! his main reliance is on his own introspection. Wundt is a man of keen sensibility. He writes of feeling con amore ; he is fond of quoting 134 TRIDIMENSIONAL THEORY OF FEELING indigobL^ Fig. 4. Wundt's Schema of the 'System der Farbengefiihle' : Physi- ologische Psychologic, 1874, 448. The affective opposition, "der einer- seits im Gelb, anders^its im Blau am starksten ausgepragt zu sein scheint, . . . ist der der Lebhaftigkeit vmd der Ruhe . . . Zwischen dem Gelb und dem. Blau gibt es aber zwei Uebergiinge : der eine durch das Griin, der andere durch die rothlichen Farbentone. ... In dem Roth und den ihm verwandten Farben ist die Bewegung des Gelb und die Ruhe des Blau zu einem zwischen Bewegung und Ruhe hin- und herwogenden Zustand der Unruhe geworden. Diese Vermittlung durch den Zwiespalt ist am deuthchsten in den blaurothen Farbentonen, wie im Violett, rep- rasentirt. Das Griin dagegen driiekt ein wirkhches Gleichgewicht aus. Im Vergleich mit dem erstarrenden Blau und dem erregenden Gelb verbreitet es ein befriedigendes Ruhegefiihl." These two modes of tran- sition make the series of simple colour-feelings a closed curve, analogous to the 'colour triangle.' "Mit Riick.sicht auf ihre Bedeutung als Ueber- gangsstimmungen %\'ird aber hierbei dem Griin angemessener das Violett als das Purpur gegeniiberzustellen sein, und es werden dem entsprechend Roth vmd Indigoblau, Gelb und Blau einander gegeniiber zu liegen kom- men. . . . Denken wir uns die den verminderten Sattigungsgraden der Farben bis zum Weiss entsprechenden Gefiihle ahnhch angeordnet, so bilden sie alle zusammen die von der Farbencurve umschlossene Ebene, in welcher der Punkt des Weiss die indifferente Stimmung bezeichnet, wne sie die einfache, weder durch besondere Starke oder Schwache des Lichts noch durch einen Farbenton modificirte Lichtempfindung hervor- bringt. Rings herum liegen die matteren und darum auch durch kiirzere Uebergange vermittelten Gefiihlstone der weissUchen Farben." "Fiir jede Farbe gibt es also drei Uebergange der Stimmung zu einer Farbe von entgegengesetztem Gefiihlston : der harmonische durch das ruhige Griin, der contrastirende durch das zwiespaltige Violett und der indifferente durch das gleichgiiltige Weiss." The complete schema of \'isual sensations is, however, tridimensional ; the vertical axis shows the colourless sensations aroused by the "Intensitatsgrade des Lichts." These have their own feelings. "Zwischen den Gegensatzen der Helhg- keit, dem ernsten Dunkel und dem heiteren Lichte, existirt nur der eine Uebergang durch das indifferente Weiss von mittlerer Helligkeit. In- dem die Lichtstiirke der Farben zu- oder abnehmen kann, konnen sie auch an diesen Gefvihlstonen der Helligkeit Theil nehmen." — Cf. ibid., ii., 1902. 329 f. FEELING AND EMOTION 135 Goethe's Farhenlehre; feeling has played a larger and larger part in his psychological sys- tem as time went on; as early as 1874 he had systematised, thrown into diagrammatic form, his affective reactions to colours and tones. So the new theory appears in the Grundriss without preface or apology, — ''wird einfach als Tat- sache eingefiihrt," Orth plaintively remarks,® — takes its place in the exposition with all the assurance of established fact. Remembering its genesis, its deep-rooted and slow growth in Wundt's mind, we need not be greatly surprised. Wundt had said in 1874, ''Gelb . . . regt an, blau stimmt herab " ; and had emphasised *'das eigenthiimliche Gefuhl des Aufmerkens" which appears "im Zustande des Besinnens oder der Spannung." ^ No doubt it seemed obvious to him in 1896 that the introspective evidence, though not expressed, would be un- derstood, — if indeed the thought of expression ever occurred to him. Now, after several years of criticism, he is more explicit; the Physiolo- gische Psychologie introduces the theory by way of definite introspective analysis.^ Even in the Grundriss, however, Wundt is not simply dogmatic. He explains (a) that a triple classification of the affective elements is required 136 TRIDIMENSIONAL THEORY OF FEELING for the distinction of the fundamental types of emotion. Later on, it is true, he declares that a psychological classification of the emotions '*nur auf die Qualitat des Gefiihlsinhaltes gegriindet werden kann." The argument has a circular look : affections are classified by reference to emotion, emotions by reference to affection. I think, however, that it is formally sound. Theoretically, emotions may be classi- fied by quality, by intensity, and by temporal course. In practice, intensity and temporal course fail to furnish reliable criteria : hence emotions must be classified by quality. Quali- tative analysis then reveals certain fundamental types of emotion, which must, of course, be pre- formed in the affective qualities. Emotive classi- fication thus points us back to a particular classification of affections, while affective classi- fication, to be adequate, must necessarily point forward to emotion. Formally, this reasoning is rather a matter of what Fechner would call the 'solidarity' of a system than an instance of merely circular argumentation.^ Whether it is materially sound is another question, — a ques- tion which Stumpf, e.g.^ would answer with an emphatic negative.^ ^ Wundt also brings evidence of an objective sort, the evidence (h) derived from the method THE METHOD OF EXPRESSION 137 of expression. He lays but slight stress on pulse- correlation in the Grundriss : ''es ist unzulassig, die Ausdrucks- der Eindrucksmethode in Bezug auf ihren psychologischen Werth gleichzuord- nen." ^^ In the Vorlesungen, too, the pulse- records are introduced to prove the physiological relationship of the 'lower' to the 'higher' feel- ings, some time before we reach the distinction of the three affective dimensions/^ It is not until 1900, in the Bemerkungen zur Theorie der Gefilhle, that the changes in innervation of heart, vessels, and respiratory mechanism — "ein tiber- aus feines Reagens auf die leisesten Aender- ungen der Starke wie Richtung der Gefuhle"^^ — are given anything like an independent place in Wundt's argument. Do not fear, now, that I shall plunge you into the technical intricacies of the expressive method, and that the remainder of the hour will be filled with sphygmograph and plethysmograph, pneumograph and dynamo- graph ! Even if that method came into our discussion, I could pass it over with the re- minder that, not so long ago, I gave a critical review of it from this platform.^* But it does not come into our discussion. Grant every- thing that the most ardent disciples of the method demand, and then ask yourselves : Where is the evidence, in these correlations, that 138 TRIDIMENSIONAL THEORY OF FEELING we are dealing with elementary mental pro- cesses ? What have pulse-curves to say to the question of the irreducibility, the ultimateness in consciousness, of the experiences of excite- ment-inhibition, tension-relaxation ? Wundt him- self is careful, in psychological connection, to differentiate **specifisclie Beschaffenheit " and **elementare Natur." ^^ How can pulse and breathing be relied upon to make the same dis- tinction ? ^® Let us, then, dismiss the expressive method, and come back to the Grundriss. Had Wundt stopped short at the point which we have now reached ; had he stated his theory, shown its usefulness in systematic regard for the classifica- tion of emotions, and indicated the correlated differences in the pulse-tracings : his position would, I think, have been stronger than it actu- ally is. But he attempts, further, (c) to connect the three dimensions of affection with the three relations in which a given feeling may stand to the temporal course of mental processes at large. Pleasantness and unpleasantness denote a de- terminate modification of our present mental state; excitement and inhibition exert a de- terminate influence upon the next succeeding state ; and tension and relaxation are qualita- tively determined by the preceding state. **Diese TIME-RELATIONS OF FEELING 139 Bedingungen machen es zugleich wahrschein- lich, dass andere Hauptrichtungen der Gefiihle nicht exist iren." ^^ And yet — quality is the criterion for the classification of emotions, and the classification of the emotions requires three ultimate affective dimensions ! Here, surely, we have the fallacy of too many proofs ! Wundt, it is true, offers in the Bemerhungen a defence of his dual argument. **Es handelt sich hier um Momente, die selbst wieder mit einander zusammenhangen : " *' [es] kommt hier uberall nicht ein Verhaltniss von Ursachen und Wirk- ungen, sondern lediglich ein solches von Bezie- hungen und Bedingungen in Frage, die sich wenigstens vorlaufig durch eine vollstandige Analyse aus der Gesammtheit der complexen Bedingungen nicht isoliren lassen." ^^ If I un- derstand these passages aright, Wundt's meaning is as follows : * Consciousness is always exceed- ingly complex, so that the affective processes are given in complex relations and appear as vari- ously conditioned. Causal analysis is, at pres- ent, beyond our powers. We can, however, trace certain relations and follow up certain part-conditions ; and our results, different or even incompatible as they may look, are really abstractions from — represent moments of — a single system of causal interrelations. Hence 140 TRIDIMENSIONAL THEORY OF FEELING they may safely be set down side by side.' In the abstract, all this may be granted. Still, how- ever, I do not see, in the concrete, how the three affective dimensions can be guaranteed both by temporal relation to the course of consciousness and by qualitative differences in emotion. The latter are enough, in themselves : the former is, at the best, a matter of reflection, of analysis above the elementary level ; and its obvious superfluity tends to cast doubt upon the results of qualitative analysis proper, with w^hich it is brought into agreement. For the rest, it is significant that, in his later writings, Wundt has dropped this principle of temporal relation as a means of affective classification. In the Voiiesungen of 1897 a new principle makes its appearance. After distinguishing the three dimensions of pleasantness-unpleasant- ness, excitement-tranquillisation, tension-relaxa- tion, Wundt says : "dass es noch andere Haupt- richtungen ausser diesen gebe, scheint mir nach der subjectiven Beobachtung nicht wahrschein- lich. Audi diirften die genannten den allge- meinsten Bedingungen entsprechen, unter denen Gefuhle liberhaupt entstehen." ^^ The dimen- sions are guaranteed first by introspection, and secondly {d) by the threefold character of affective conditions. The conditions are found THE CONDITIONS OF FEELING 141 in the " Empfindungs- und Vorstellungsele- mente, an die [die Geflihle] gebunden sind." ^^ Pleasantness-unpleasantness represent a quality- dimension; excitement -tranquillisation, an in- tensity-dimension; tension-relaxation, a time- dimension. *'Die Bedeutung von Lust und Unlust als * Qualitatsrichtungen ' liegt darin, dkss vorzugsweise in ihnen die Wirkungen der quali- tativen Eigenschaften des gesammten Bewusst- seinsinhalts zum Ausdruck kommen": and similarly with the other two dimensions. ^^ In- trinsically, of course, every affection is a quality, qualitatively different from every other. But the affective qualities of the three dimensions reflect, express, are determined by the quality, intensity, and temporal properties of sensations and ideas. I am not here concerned with the correctness or incorrectness of Wundt's correlation. He has himself changed it, in the Physiologische Psij- chologie of 1902, where pleasantness-unpleasant- ness represents an intensive, and excitement- tranquillisation a qualitative dimension, ^^ — just the reverse of what was said in 1897. I am con- cerned with the correlation as a principle of classification. There are, Wundt declares, three general conditions of the arousal of feeling : the quality, the intensity, and the temporal relations 142 TRIDIMENSIONAL THEORY OF FEELING of conscious contents. And the threefold char- acter of the conditions furnishes, along with introspection, evidence that there are but three dimensions of affection. What, then, has become of the spatial relations of conscious contents.^ The chapter-headings of the Physi- ologische Psychologie tell us that Sinnes- vorstellungen are of three kinds : intensive, spatial, temporal. Spatial and temporal ideas may be grouped together as extensive ; intensive ideas differ from sensations by the composite nature of their intensity and quality.^^ These intensive ideas are therefore responsible for two affective dimensions, the intensive and qualita- tive; the temporal ideas are responsible for a third dimension, the temporal ; only the spatial ideas are excused from affective duty. I argue, then, in this way. In so far as affective classifi- cation is dependent upon the various forms of idea, in so far Wundt's classification is inade- quate ; for the spatial form of idea is as impor- tant, in the mental life, as the intensive or the temporal.^* And if there is no such thing as a spatial dimension of affective qualities, then we may justly doubt whether the principle of classi- fication is sound, and whether any conclusion as to the number of affective dimensions may be deduced from it. Remember, I am not arguing THE CONDITIONS OF FEELING 143 on a matter of fact; I am considering the ap- plication of a principle. Wundt replies, in the Bemerkungen, that he has left spatial ideas out of account for two rea- sons : first, " weil sich mir Beziehungen derselben zu bestimmten Geflihlsrichtungen weder in der unmittelbaren subjectiven Beobachtung noch bei der Analyse der Ausdrucksbewegungen dar- boten;" and secondly, "weil es mir scheint, dass man sehr wohl bei jedem Affect qualitative, intensive und zeitliche Eigenschaften unter- scheiden kann, wahrend ich mit dem Ausdruck, der Zorn oder die Freude habe irgend eine raumliche Ausdehnung, keinen rechten Sinn zu verbinden weiss." ^^ The first of these arguments misses its mark for the reason that, in the Vorlesungen, the distinction of three general conditions of feeling, their connection with three forms of idea, is offered as additional evidence, over and above * subjective Beobachtung,' for the finality of Wundt's classification. '' Auch dlirften die genannten Hauptrichtungen den allgemeinsten Bedingungen entsprechen, unter denen Gefiihle tiberhaupt entstehen." I object to Wundt that the one of his two criteria is in- valid, and he rejoins that the other is valid ! The second argument goes equally wide. I did not assert that an emotion possesses spatial 144 TRIDIMENSIONAL THEORY OF FEELING attributes, that an affection may be 'extended/ but that certain ideas possess spatial attributes and relations, — and that, if we are classifying affections by reference to the forms of ideas, then these spatial characteristics must be taken into account, as well as the intensive, qualitative, and temporal. I use the phrase 'spatial dimension of affective qualities ' precisely as Wundt uses the phrase 'temporal dimension,' — to signify affec- tive qualities that are dependent upon ideational extension. I acquitted Wundt, just now, of the charge of circularity; I am afraid that I must here charge him with the logical error which is know^n in the vernacular as 'missing the point.' In sum, therefore, Wundt 's three affective dimensions are supported, primarily, by his own introspection, while he has appealed, further, to the necessities of emotive classification ; to the results of the method of expression; to the temporal relations of the affective processes; and to their general conditions in consciousness. The first use of these arguments I take to be sound, both formally and materially, though I do not arrive by it at the conclusion which Wundt has reached. The second must be pronounced irrelevant ; the third has been given up by Wundt himself ; the fourth we have seen to be logically defective and psychologically indefensible. TERMINOLOGY 145 We have now to consider the theory on the basis that remains for it : introspection of the simple sense-feelings and qualitative analysis of the emotions. I find a difficulty, at the outset, in Wundt's terminology. You may have been surprised that, when I have had occasion to mention Wundt's category of 'excitement,' I have paired it with 'inhibition' or 'tranquillisa- tion,' rather than with the more usual term 'depression.' I have, throughout, been quoting Wundt's own words ; but it is true that, in the Grundriss, 'depressing' is given as an alterna- tive to 'tranquillising,' and that in the Physi- ologische Psychologie 'Depression' is suggested for the higher deorrees of ' Beruhiorunor.' ^e Wundt can, of course, do no more than take language as he finds it. But I think that his actual choice of words bears witness to a conflict, in his thought, between two purposes : the purpose of transcribing his introspections, and the pur- pose of maintaining the typical affective move- ment betw^een opposites. Pleasantness and un- pleasantness. Lust and Unlust, are opposite in name, as well as in nature. What of Spannufig and Lbsung ? In English, ' relaxation ' — which, I suppose, is the nearest equivalent of Losung — suggests rather the remitting or resolving of tension than its qualitative opposite : this latter 146 TRIDIMEXSIOXAL THEORY OF FEELING would be better expressed by 'relief.' Possibly Losung has for Wundt an implication of positive relief, of Erleichterung, — though it has not for me, nor for German friends of whom I have made inquiry. Wundt speaks also of the Befriedigung, the fulfilment, of expectation ; ^^ but that term brings us perilously near to Be- ruhigung. The chief difficulty, however, arises in connection with the remaining dimension. What is the opposite of Erregung? Sometimes Wundt says Hemmwig, sometimes he says Beruhigung, sometimes Dejyression. The an- tithesis Erregung- HemmuJig comes from nerve- physiology ; ^^ Erregung-Depi'essioii comes, evi- dently, from observation of the emotions, normal and pathological ; Erregung-BeruJiigung appears to be the analogue of Spannung -Losung and to convey the same suggestion. But what is, in introspection, the felt opposite of Erregung ? I cannot myself identify the feelings of Hemmimg, Depression, Beruhigung ; I cannot feel them as degrees of the same thing, as lying in the same affective dimension ; I cannot always distinguish between Beruhigung and Losung. Erregung^ 'excitement,' seems to me to feel very differently in different contexts, to be an equivocal term. It is easy to say that such considerations are mere ' Wortklauberei ' : but I am trying to express a real introspective difficulty. VARIANTS OF THE THEORY 117 If, then, I am to judge others by myself, this uncertainty in the meaning of terms may be at least a partial reason for the fact that Wunclt's classification, despite its claim to finality, does not always command the assent even of those who agree with its spirit and intention. Gure- witsch, e.g., in his Theorie der sittlichen Gefilhle, makes a fourth affective category for Streben- Widerstreben.^^ ^ogi, again, ranges feelings of activity and passivity alongside of pleasantness- unpleasantness, arousal-depression, tension-re- laxation.^^ Wundt identifies Strebwigsgefuhl with Thdtigkeitsgefuhl, which he regards as a total feeling, compounded of strain and excite- ment.^^ Royce, on the other hand, is disposed to think that two dimensions — pleasantness- unpleasantness and restlessness-quiescence — are adequate to the facts of the affective life.^^ I do not at all mean that these differences of opinion are fatal to the theory. But they testify — do they not ? — to a lack of precise formula- tion. Royce throws two of Wundt 's dimensions into one; Vogt and Gurewitsch split the same two into three. The single dimension about which Wundt himself seems, from the first, to have felt no doubt is that of Spanming-Losung. The other two dimensions, as I pointed out just now, have 148 TRIDIMENSIONAL THEORY OF FEELING actually exchanged places in his system. x\nd the same uncertainty characterises certain of his observations in detail. Let me give you an instance. In the Bemerkungen of 1900, Wundt writes: "ich wusste . . ., wenn ich vor die Walil gestellt ware, irgend einen dieser Ein- driicke dem andern vorzuziehen, absolut niclit zu sagen, ob mir das rein spektrale Blau oder das Roth . . . angenehmer sei." This does not mean that the two colours would be equally pleasant. *'Ich wtirde eben einem solchen Ver- langen immer nur die iVussage gegenliberstellen konnen, dass diese Eindriicke an sich mit Lust und Unlust nichts zu thun haben." ^^ The passage is a little startling, when one remembers that work had already been done upon colours — and colours that were not spectral colours — by the method of impression ! ^* Two years later, now, we have the following: "wenn ich zuerst ein spektralreincs leuchtendes Roth und dann ein ebensolches Blau im Dunkelraum betrachte, so kann ich nicht umliin, beide als im hohen Grad erfreuende, also lusterregende Eindriicke zu charakterisiren." ^ True, the sen- tence is concessive; the next begins with a 'gleichwohl'; but it is, nevertheless, in flat con- tradiction to the former quotation. If two im- pressions are highly pleasant, they can be com- TPIE THREE DIMENSIONS 149 pared as regards pleasantness, and a judgment of greater, less, or equal can be passed upon them. Similarly conflicting statements are made concerning high and low tones. ^^ I readily acknowledge, again, that these minor incon- sistencies are in no sense fatal to the theory; indeed, Wundt has so often emphasised the im- portance for feeling of the "ganze Disposition des Bewusstseins " ^^ that I feel reluctant, as it were a morsel ashamed, to dwell upon them. Still, they are there ! And it is not reassuring to find that the dimension Spannung-Losung owes its exceptional position, the stability of which I spoke above, to its systematic connection with the doctrine of apperception. It must have occurred to many of you, when, earlier in the Lecture, I was arguing the claims of space as a condition of feeling in consciousness, to ask : What, then, after all, are the claims of time ? Since, in the psychology of sensation, duration and extension are, both alike, to a very large extent equivalent to, interchangeable with, in- tensity, why should they not be bracketed with intensity as the conditions of one and the same affective dimension ? We should then have something like Royce's classification : pleasant- ness-unpleasantness, conditioned upon all the 'qualitative' attributes of sensation, and ex- 150 TRIDIMENSIOXAL THEORY OF FEELING citement-quiescence, conditioned upon all the * intensive.' Now Wundt recognises the equiva- lence, under certain circumstances, of intensity and duration. ''Insbesondere kann die Lust- Unlustcomponente [bei langerer Einwirkung auf das Bewusstsein] ganz dieselben Veranderungen erfahren, die audi die Steigerung der Intensi- tat mit sich fiihrt." ^^ But feelings of SiJan- nung and Losung are "die specifischen, fur die Aufmerksamkeitsvorgange charakteristischen Elemente." ^^ "Da aber Apperception und Aufmerksamkeit zeitlich sich entwickelnde Vor- gange sind, die zugleich in einer bestimmten zeitlichen Folge wechseln, indem jede Losung eine vorangegangene Spannung fordert, und eine neue Spannung wiederum nur auf Grund vorangegangener Losungen einsetzt, so sind diese Gefiihlscomponenten enger als die iibrigen an den zeitlichen Ablauf der Bewusstseinsvor- gange gebunden." ^° Any serious doubt, there- fore, about Wundt 's doctrine of attention and apperception must at the same time jeopardise this third dimension of simple feeling. So far, I have spoken only of the three affec- tive dimensions ; I have said nothing of the multitude of elementary qualities which the di- mensions are held to include. "Die qualitative THE PLURALITY OF QUALITIES 151 Mannigfaltigkeit der einfachen Gefiihle ist un- absehbar gross und jedenfalls viel grosser als die Mannigfaltigkeit der Empfindungen." ^^ So the Grundriss, — which proceeds to give two rea- sons. First, every sensation of the muhidimen- sional sensation-systems belongs to more than one affective dimension. Secondly and more importantly, the feelings that attach to sensa- tion-complexes, intensive, spatial, and temporal ideas, and to certain stages in the temporal course of emotion and volition, are nevertheless them- selves irreducible, and must therefore be counted among the elementary affective processes. You will notice that these reasons are phrased in the language of a special psychological system, though the appeal to introspection is implied. Later on, the appeal becomes explicit ; we are reminded that, e.g,, the feeling of gravity, Ernsts "in verschiedenen Fallen in seiner Qualitat wieder variiren kann." ^^ In the Vorlesungen, the doctrine of the multiplicity of affective quali- ties follows naturally from the doctrine of the TotalgefiXhl}^ The Physiologische PsycJiologie relies upon an *aufmerksame Selbstbeobach- tung.' ^^ We are apt to overlook the great variety of the feelings, partly because they are intimately bound up with the objective contents of con- sciousness, partly because we have no words to (S 152 TRIDIMENSIONAL THEORY OF FEELING express them. ''Angesichts der [an der Hand des vergleichenden Verfahrens der Eindrucks- methode] ausgefulirten Analyse scheint es mir in iiberwiegendem Masse wahrscheinlich, dass die seclis Gnindformen . . . eben nur Grundformen sind, von denen jede einzelne eine selir grosse Mannigfaltigkeit im ganzen verwandter, aber dabei doch von Fall zu Fall nuancirter Ein- zelgefiilile unter sicli begreift." ^^ There can be no manner of doubt that, in this matter of the number of the affective qualities, the psychological pendulum has been swinging, of recent years, in the direction that Wundt has taken. Ladd emphatically repudiates the view that '"pleasure-pains' are exhaustive of the en- tire quality of the feeling-aspect of conscious- ness." The theory is simplicity itself: *'but simplicity, in the interests chiefly of biological and experimental psychology, *gone entirely mad.'"^® I do not know whether Ladd felt pleased or pained that he had written this last sentence, when two years later he read Wundt's Grundriss. He says himself, however, that "almost all mental states which are marked by strong feeling in the case of developed minds are mixed feelings." *^ At any rate, he works resolutely through the sense-departments, in 1894, and makes out a long list of elementary THE PLURALITY OF QUALITIES 153 processes. James, in the same year, remarks that "there are infinite shades and tones in the various emotional excitements, which are as dis- tinct as sensations of colour are, and of which one is quite at a loss to predicate either pleasant or painful quality." ^^ This position is, of course, entirely compatible with a dual view of Lust-Unlust, of "the primary GefiiJilston'' ; in- deed, the two doctrines seem to me to appear, side by side, in James' own exposition. Never- theless, the passage may fairly be cited in the present connection. Lipps, again, working as it were from the opposite pole to Wundt, has arrived, as we all know, at a very complicated classification of the feelings. ^^ Stumpf has ex- pressed the opinion, as against Kiilpe, that "sinnliche Annehmlichkeit " and "sinnliche Un- annehmlichkeit " cover "eine grossere Mannig- faltigkeit von Gefuhlsqualitaten." ^^ This array of convictions is imposing, even if there are authorities — Hoffding, Kiilpe, Jodl, Ebbing- haus, Lehmann, Rehmke ^^ — upon the other side. The fact is, of course, that the ultimate ques- tion of our second Lecture, the question of the criteria of affection, has not been settled. The parties to the present controversy do not really * feel ' differently ; but they approach the problem 154 TRIDIMENSIONAL THEORY OF FEELING with a certain attitude towards affective process, with a certain general view of the status of feel- ings in consciousness. Ebbinghaus says outright that Wundt and Jodl, e.g., are 'not talking of the same things.' ^^ Orth believes that Wundt 's theory is the outcome ** seiner ursprlinglichen Auffassung des Verhaltnisses zwischen Empfin- dung und Gefuhl." ^^ Ladd writes with a sort of ethical, even religious, atmosphere upon him : how can you compare the pleasure of cheese and beer with the pleasure of seeing a good Hamlet ? ^* Lipps considers the feelings as modes of reference to the self; feelings are "Ichinhalte oder Ich- qualitaten."^^ Stumpf adopts a sensationalist view of the sense-feelings ; and in sensation qualitative differentiation is obvious enough. James is concerned with the varieties of emotive experience, and his protest against the 'hack- neyed psychological doctrine' that pleasure and pain are the essence of emotion comports, as I have pointed out, with a strictly dualistic view of the affective qualities proper. It is not that our affective experience is radically different, but that w^e approach it from different directions, see it under different angles, assimilate it in terms of our systematic associations. I do not mean that the point at issue is a mere Etikettenfrage. It is much more than that. THE TOTALGEFUHL 155 Our decision 'makes a difference,' as the prag- matists say, to the whole structure of our psy- chological system. And it must be remembered that Wundt does not acknowledge any other methods than those employed by the dualists, and would not acquiesce in the statement that his results are of another order. He comes within our universe of discourse ; he invites argu- ment. I therefore proceed to argue ; and I take as ground for argument an illustration which he employs on more than one occasion, — the feeling which attaches to the common chord c-e-g .^^ Let me remind you, first, of Wundt's doctrine of the Toialgejuhl}'' A compound feeling, a feeling due to the confluence of a number of elementary feelings, is always psychologically simple in the sense that it has its own irreducible quality, but may also permit the distinction of its various components. *'In jedem derartigen Gefuhl lassen sich Gefuhlscomponenten und eine Gefuhlsresultante unterscheiden." The compo- nents Wundt terms 'partial feelings,' the re- sultant, 'total feeling'; we have had an in- stance already in the 'feeling of activity' which results from the compounding of tension and excitement. The compound feeling thus bears a close resemblance to the formation which, in 156 TRIDIMENSIONAL THEORY OF FEELING the sphere of tonal sensation, is called a fusion ; Wiindt speaks, in the Physiologische Psychol- ogie of 'affective fusions.' There are degrees of affective, as there are degrees of tonal, fusion ; the partial feelings may appear simply as an un- differentiated colouring of the resultant, or may maintain their individuality, though in a sub- ordinate position, alongside of the total feeling. After this preface, we are ready to listen to the three tones. To prevent a swamping of the partial feelings by the total feeling, — the high- est degree of affective fusion, — we take the tones separately in succession, and observe how they *feer in isolation. The tone c, heard by itself, affects us, Wundt says, by way of a * calm seriousness ' or a * quiet cheerfulness ' ; it brings out feelings of two dimensions, pleasant- ness-unpleasantness and excitement-tranquilli- sation. The other two, e and g, will do the same, — though the affective qualities will be somewhat different. If, now, we put the tones together in pairs, every pair will give us a com- pound feeling: we have the three total feelings of ce, eg, eg, accompanied or coloured by the partial feelings which we have compounded. And if the conditions are favourable for obser- vation, we should be able to distinguish a five- fold feeling in connection with every pair; the AN EXPERIMENT 157 two dimensions of the two partial feelings, and the total feeling. Now let us sound all three tones simultaneously. We have the total feeling of c-e-g; we have three relative total feelings, or 'partials of the second order,' as Wundt calls them, — the feelings of ce, eg, eg; and we have the 'partial feelings of the first order,' the six elementary feelings aroused by c, e, and g. The feeling of c-e-g is a tenfold complex. Do not forget that such a feeling is, for Wundt, an " einheitliche Mannigf altigkeit " ; do not forget that the partial feelings may, more or less com- pletely, have forfeited their independence. But, with all allowance made, ask yourselves if you experience anything like the body of feeling that, on Wundt's theory, you 'ought' to experience. Suppose that, in spite of our precautions, affec- tive fusion has reached its highest degree; let the partials of the first order disappear alto- gether, as separate components, and let them remain only as a vague colouring of the whole affective impression. Now your compound feel- ing should be a fourfold complex. Surely, it is not; surely, the feeling lacks the depth, the solidity, that a feeling thus compounded must possess; surely, you can describe the chord in no other terms than 'slightly pleasant,' 'mod- erately agreeable.' 158 TRIDIMENSIONAL THEORY OF FEELING I think that it is fair to test the theory in this way, by the judgment of a group of psychologi- cally trained observers, seeing that Wundt has laid the observation before the psychological public in two of his books. I have, for myself, repeated the test often and again, and have varied it in half-a-dozen ways : always, while the chord remains a single impression, a sensible fusion out of musical setting and so far as pos- sible freed from musical significance, I get the same meagre affective result. ^^ If, now, Wundt retorts that in this and like instances we are feeling-deaf and feeling-blind, may we not suggest, on our side, that he is organically anaesthetic .^ The lack of interest that Wundt shows in the organic sensations has always been a source of wonderment to me. Take the new edition of the PJujsioIogische Psychologie. Here is a total of 2035 pages. Of these, 45 are given to Tast- und Gemememp- findungen; the Gemeinemjjfindungen alone, which I now have principally in mind, receive four, two and a half of which are devoted to pain ! ^^ Of course, there are all sorts of scat- tered references. But look in the index under Organempjindungen, Gemeinem'pjindungen, Nie- dere Sinne, Geleiikevipfindungen, Muskelsinn, — what you can think of. Aside from Bewe- THE ORGANIC SENSATIONS 159 gungsempfindungen and Augenbewegimgen there is surprisingly little. Meumann makes a similar complaint with regard to Nagel's Handbuch. ''Vermisst hat der Referent, dass den inneren Empfindungen (Organempfindungen) kein aus- fiihrlicheres Kapitel gewidmet wird; die gegen- wartige Physiologie scheint sich mit der Frage der Sensibilitat der inneren Organe nicht mehr viel zu beschaftigen." ^^ Now I personally be- lieve that the organic sensations play an im- portant part, not only in feeling and emotion, but in many other departments of the mental life : in the formation of sensory judgments, in the mechanism of memory and recognition, in motives to action, in the primary perception of the self. It is true that, as compared with what w^e know^ of sight and hearing, our knowledge of the organic sensations is scrappy in form and small in amount ; that is why I have said, in another connection, that **of all problems in the psychology of sense w^hich are now before us, the problem of the nature, number, and laws of connection of the organic sensations appears to me to be the most pressing." ^^ Let me add, now, that if any one of you is thinking of a piece of work in this general field, he would do far better, in my opinion, to start out from the side of the organic sensations than to succumb to 160 TRIDIMENSIONAL THEORY OF FEELING the fascinations of pneumograph and sphyg- mograph. Well ! I believe that organic sensations are responsible for the dimensions of excitement- depression and tension-relaxation. On this point I can claim the support of Ebbinghaus ^" and, I suppose, of all those who accept the James- Lange theory of emotion. Stumpf, too, declares that he cannot regard them as ''Elementarer- scheinungen," though he offers no further analysis.^^ But I believe, also, that organic sensations are responsible in certain cases for a Nuancirung, a shading and colouring, of feelings in the dimension of pleasantness-unpleasantness. I say 'in certain cases,' for two reasons. First, it is entirely possible that this Nuaiicirung is a matter, not of simple sense-feeling, but of association, of emotive residua. ^^ Secondly, however, I do not think that the colouring and shading is as universal as Wundt asserts. Vogt, whose method of suggestion led him to the dis- tinction of four pairs of feelings, is unable to discover it.^ Orth cannot find it, in the intro- spections that he educes by the Reizinethode .^^ Storring's observers, on the other hand, report a qualitative difference between Stimmungslust and Em/pfindiingslust ; but though this is, so to say, a gross difference, the expressions used are THE METHOD OF IMPRESSION 161 singularly disappointing. We read, in some detail, of extensive differences, differences in intensive fluctuation, differences of excitement and passivity; but on the side of quality we have only ''Stimmungslust ist gleichartiger," and the dogmatic statement ''Zwischen Stimmungs- lust und Empfindungslust besteht qualitative Differenz."®^ I myself have never observed a qualitative differentiation of pleasantness-un- pleasantness, under experimental conditions; and when I observe a difference in everyday life, — a difference on the level of the sense-feel- ing, — I seem to find a reason for it in concomi- tant organic sensations. I have sought, on two occasions, to put Wundt's theory to an experimental test. The method employed was the method of impression, in Colin 's form of paired comparisons. The procedure, in brief, is as follows. A series of stimuli — tones or colours or rhythms — is laid out, and the stimuli are presented to the observer two at a time, care being taken that every mem- ber of the series is paired with every other mem- ber. The observer has to decide which of the two stimuli shown him is the more pleasant, the more unpleasant, the more exciting, the more depressing, and so on. If colours are exhibited, he points to right or left, as the case may be; 162 TRIDIMENSIONAL THEORY OF FEELING if tones are used, he notes down *1' or *2,' according as the first or second stimulus is pre- ferred. The work is laborious, and the method consumes a large amount of time. We have, however, the great advantage of a twofold con- trol, objective and subjective. The subjective control is afforded, of course, by the introspection of the observers. The in- trospective task is extremely simple; the ob- server has merely to be passive, to let himself go, to allow the stimuli to take affective pos- session of him ; and then to indicate, in the par- ticular instance, which of the two makes the stronger impression. Moreover, since the in- trospective experience within a series is cumu- lative, all of the same kind, the observer is able, in the intervals betw^een successive series, to give a general account of his method of judg- ment, of the nature of his affective reaction. The objective control is afforded by the course of the affective judgments themselves. If, e.g.^ pleasantness and unpleasantness are really affec- tive opposites, then the * curves' or tracings which indicate the distribution of judgments in parallel * pleasant' and * unpleasant' series should be diametrically opposed : a colour which stands high on the scale of pleasantness should stand low on the scale of unpleasantness. THE METHOD OF IMPRESSION 163 and contrariwise. If excitement-depression and tension-relaxation also denote affective oppo- sites, then their 'curves' should be similarly opposed. The stimuli chosen were colours, musical tones, and groups of metronome beats given at varying rates. The two former had been speci- fied by Wundt as productive of excitement-de- pression, the latter as productive of tension- relaxation. My idea was, on the subjective side, to test by their means the immediacy of reaction in these dimensions. In the case of pleasantness-unpleasantness, you cannot say what the basis of your judgment is, otherwise than that it resides in the stimulus; the one of two colours or two tones is more pleasant than the other, just as directly as it is bluer or louder. Suppose, then, that colours and tones bring out equally prompt and unmediated judgments of excitement-depression, and that metronome in- tervals bring out equally prompt and unmediated judgments of tension-relaxation : then we shall have some ground for the acceptance of the two new affective dimensions. Suppose, on the other hand, that the judgments of excitement and ten- sion are forced or difficult, mediated by associa- tions or by organic sensations : then we shall have an introspective differentiation of these 164 TRIDIMENSIONAL TPIEORY OF FEELING judgments from those of pleasantness-un- pleasantness. On the objective side, I argued in much the same way. Suppose that the curves, of which I spoke just now, show typical differences, — so that the distribution of judgments of pleasant- ness takes one course, that of judgments of ex- citement another, and that of judgments of tension a third, — while still the curves of pleas- antness and unpleasantness, of excitement and depression, and of tension and relaxation are related as opposites : then, again, there will be ground for the acceptance of Wundt's dimen- sions. Suppose, on the contrary, that the curves of excitement and of relaxation agree with the curve of pleasantness and the curves of depres- sion and of tension with the curve of unpleas- antness : then, since the pleasant-unpleasant dimension is not in dispute, we have a strong indication that that alone is fundamental and that the other two dimensions are affective only because and in so far as pleasantness and un- pleasantness are involved in them. The results of the first investigation, in which colours and musical tones were tested for pleasantness-unpleasantness and excitement-de- pression, and metronome intervals for pleasant- ness-unpleasantness and tension-relaxation, were THE METHOD OF IMPRESSION 165 published in the Wundt Festschrift; those of the second, in which the same tones and intervals were tested for all three of the Wundtian dimen- sions, were published by Hayes in the American Journal of Psychology. They may be summed up under three headings. ^^ (1) Judgments of pleasantness and unpleas- antness are direct, easy, and natural. The qualities themselves appear to the observers to be simple and homogeneous, identical through- out the experiments. Their opposite character is vouched for both by introspection and by the course of the curves. (2) Judgments of excitement are less direct, and the term is equivocal. If it is taken as the opposite of depressing melancholy, its curve agrees with that of pleasantness; if it is taken as the opposite of tranquillity or soothing calm, its curve agrees with that of unpleasantness : the reverse curves then agree with those of un- pleasantness and of pleasantness, respectively. If, in default of special instruction, the observer vacillates between the two meanings of the word, the curve shows a vacillating character, — partly 'pleasant' and partly 'unpleasant'; the period and nature of the affective oscillation are vouched for by introspection. Judgments of depression are, in their turn, distinctly less direct than those 166 TRIDIMENSIONAL THEORY OF FEELING of excitement, and are often associatively medi- ated. There is no evidence of a dimension of excitement-depression, still less of a number of exciting and depressing qualities. (3) Judgments of tension are easy; but ten- sion is described, throughout, in kinaesthetic terms. Increasing tension means, uniformly, in- creasing unpleasantness, and the curves of the two classes of judgment correspond. Relaxa- tion may be taken as the opposite of unpleasant tension, in which case its curve agrees with the curve of pleasantness, or may be identified with depression. Nowhere is there evidence, in this third case, either of a new affective dimension or of specific qualities. Naturally, these results are not 'conclusive.' For one thing, the experiments are too few. For another, they were obtained in a single laboratory, and that a laboratory from which criticism of Wundt's doctrine had already pro- ceeded. For a third, the argument upon which the experiments rest is not demonstrably valid. It w^ould, I think, be a very strange thing if three sets of stimuli should affect a number of ob- servers by way of excitement-depression (or tension-relaxation) precisely as they do by way of pleasantness-unpleasantness, — but nobody can prove that such a state of affairs is, on the SUMMARY 167 plural theory, impossible. Were I a champion of affective plurality, I should unhesitatingly urge these objections to the work, and I have no desire to slur them over because I am on the other side. Nevertheless, the results are experi- mental evidence; Wundt cannot, in the future, appeal to the method of impression as confi- dently as he has appealed in the past.^^ And if our investigations are compared with those of Brahn and Gent, upon which Wundt relies in the Physiologische Psychologie, it will appear, I am very sure, that the critical sauce meted out to the goose must be considerably strengthened for the gander. '^^ If now, in conclusion, I may give, with all due modesty, my own reading of the situation, ^^ it is this: that Wundt's tridimensional theory of feelings shows, as it were in typical form, the peculiar features that distinguish his psychology at large. Wundt has, in an eminent degree, the power of generalisation, and his generalisations cover — as generalisations oftentimes do not ! — an encyclopaedic range of detailed knowledge. But the exercise of this very power leads him to put a certain stamp of finality upon his theories, as if questions were settled in the act of systema- tisation. You know what I am thinking of : 168 TRIDIMENSIONAL THEORY OF FEELING the theory of space perception, the theory of attention, the definition and demarcation of psychology itself. The affective theory which we have been discussing is typical, then, both for good and for bad. It is good, in that it gives rounded and complete expression to a psychological tendency that, in many minds, has been struggling for utterance. It is bad, in that it offers a solution, ready-made, of problems which in actual fact are ripe only for prelimi- nary and tentative discussion. Like those other theories of attention and of space-perception, it represents the culmination of an epoch of psycho- logical thought ; but, like them again, it is rather the starting-point for further inquiry than the statement of assured psychological result. On the whole, I take it as matter of encouragement that generalisation has been at all possible. What has been done, provisionally, at a lower level of knowledge, can be done again, and bet- ter done, at a higher. In the meantime, we must not be dogmatic, we must not be too im- patient for results, we must not set theory above observed fact; recognising to the full the diffi- culty and the merit of constructive effort, we must use all the weapons in our critical armoury against ourselves as against others, and against others as against ourselves. V ATTENTION AS SENSORY CLEARNESS LECTURE ^ ATTENTION AS SENSORY CLEARNESS I SUPPOSE that every experimental psycholo- gist has, at one time or another, been con- fronted with the sceptical question : ' What, after all, has the experimental method done for general psychology ? ' As a rule, it is not easy to find an answer : first, because the questioner, both by the fact and by the manner of his asking, betrays an ignorance of psychology at large ; but sec- ondly, and more especially, because the influence of the experimental method has, as a matter of fact, made itself felt over the whole extent of the psychological system, and instances fail you by the very number and urgency of your associa- tions. " Wenn ich zusammenfassend sagen soil," — this is Wundt's reply to the question, — "was ich selbst an psychologischen Einsichten der experiment ellen Methode verdanke, so kann ich nur antworten : Alles, was ich auf diesem Gebiete fiir richtig und zum Theil fiir unum- stosslich halte." ^ That is largely and positively said. But if we want details, I think that the experimentalists may justly point to three prin- 171 172 ATTENTION AS SENSORY CLEARNESS cipal achievements : the complete recastmg of the doctrine v f memory and association,^ the creation of a scientific psychology of individual differences,^ and +he discovery of attention/ To say, however, that experimental psy- chology * discovered ' attention is to make a fairly sweeping claim, and a claim that you may reasonably incline to dispute. What of Hamil- ton ? You will remind me that Hamilton gives a long discussion of "attention as a general phenomenon of consciousness";^ you may even recall the fact that I myself, in a previous Lecture, quoted this discussion.* What of James Mill, and the twenty-fourth chapter of the Analysis ? ^ What of Bain, and the theory of attention that we find in The Emotions and the Will ? ^ Well ! I make you a present of Hamilton and Mill and Bain. I will do more; I will cite a strongly worded sentence from Braunschweiger. "It would be hard," says this author, a special student of the history of atten- tion, "to find a single idea or thought that can contribute in any sort of way to the solution of this important problem, which does not appear at least m nuce during the eighteenth century." ^ No doubt ! — and we are told, in the same man- ner, that Darwinism goes back to the philosophy * P. 75. THE DISCOVERY OF ATTENTION 173 of Ancient Greece. But what I mean by the 'discovery' of attention is the explicit formula- tion of the problem ; the recognition of its separate status and fundamental importance; the realisation that the doctrine of attention is the nerve of the whole psychological system, and that as men judge of it, so shall they be judged before the general tribunal of psychology. In this sense, surely, experimental psychology discovered attention. And as we connect the name of Helmholtz with the doctrine of sensible quality, and the name of Fechner with that of sensible intensity, so must we connect the name of Wundt with the doctrine of attention, — which, as I see it, is that of sensible clearness. The experiments which Wundt carried out in the early sixties are the beginning of the whole matter ; ^ and the system which Wundt has wrought out is informed and infused with atten- tional theory. The veriest beginner knows that, if he goes to Wundt, he must read about apper- ception ! ^^ It is true that the discovery of attention did not result in any immediate triumph of the experimental method. It was something like the discovery of a hornets' nest : the first touch brought out a whole swarm of insistent problems. We have only to travel beyond the limits of 174 ATTENTION AS SENSORY CLEARNESS Wundt's system, and we find that * chaos' of which Pillsbury complains. *'Die Aufmerk- samkeit," says Ebbinghaus, '*ist eine rechte Verlegenheit der Psychologic." ^^ I think that he has felt the Verlegenheit himself; there is a marked difference between his accounts of sen- sation and association, on the one side, and his treatment of attention, on the other. A char- acteristic feature, both of Ebbinghaus' sections and of Pillsbury' s recent book, is the constant appeal to casual introspection, to the occur- rences of everyday life; and though the appeal is useful, as sustaining the reader's interest, it is none the less a confession of scientific weak- ness. We expect the illustrations in a modern work on electricity to lead us, beyond themselves, to a severely technical exposition; we do not expect to stop short with the illustrations. I shall begin my own discussion of attention with an attempt to lay a very ancient ghost, — the ghost that stalks through current statements of psychological method. Kant told us, more than a century ago, that psychology could never rise to the rank of an experimental science, be- cause psychological observation interferes ^ith its own object.^' We have bowed down before this criticism; and, because the facts were con- THE METHOD OF PSYCHOLOGY 175 tinually against it, we have tried in all sorts of ways to get round the facts, and to save Kant's infallibility while we still went on with our ex- periments. Let us, now, look the objection squarely in the face. Is there anything peculiar, anything fatal, about attention to mental processes ? We are agreed, I suppose, that scientific method may be summed up in the single word ' observation ' ; the only way to work in science is to observe those phenomena which form the subject-matter of science. And observation means two things : attention to the phenomena, and record of the phenomena ; clear experience, and communication of the experience in words or formulae. We shall agree, further, that, in order to secure clear experience and adequate report, science has recourse to experiment, — an experiment being, in the last resort, simply an observation that may be repeated, isolated, and varied. What, then, is the difference be- tween natural science and psychology ? between experimental inspection and experimental intro- spection ? We may set out from two very simple cases. (1) Suppose that you are shown two paper discs, the one of an uniform violet, the other composed half of red and half of blue. Your problem is, 176 ATTENTION AS SENSORY CLEARNESS SO to adjust the proportions of red and blue in the second disc that the violet which appears on rotation exactly matches the violet of the first disc. You may repeat this set of observations as often as you will ; you may isolate the observa- tions by working in a room that is free from other, possibly disturbing, colours; you may vary the observations by w^orking towards the equality of the violets first from a tw o-colour disc that is distinctly too blue, then from a disc that is dis- tinctly too red : and so on. (2) Suppose, again, that the chord c-e-g is struck, and that you are required to say how many tones it contains. You may repeat this observation ; you may isolate it, by w orking in a quiet room ; you may vary it,' by sounding the tones first in succession and then all together, or by striking the chord at different parts of the scale. It is clear that, in these cases, there is no difference between introspection and inspection. You are using the same method that you would use for counting the swings of a pendulum, or for taking the readings from a galvanometer scale, in the physical laboratory. Now let us take some instances in which the material of introspection is more complex. (3) Suppose that a word is called out to you, and that you are asked to observe the effect which this stimulus produces upon consciousness : how THE METHOD OF PSYCHOLOGY 177 the word affects you, what ideas it calls up, and so forth. The observation may be repeated ; it may be isolated, — you may be seated in a dark and silent room, free from disturbances; and it may be varied, — different words may be called out, the word may be flashed upon a screen instead of spoken, etc. Here, however, there does seem to be a difference between intro- spection and inspection. The observer who is watching the course of a chemical reaction, or the movements of some microscopical creature, can jot down from moment to moment the .successive phases of the observed phenomenon. But if you try to report the changes in conscious- ness, wdiile these changes are in progress, you interfere with consciousness; your translation of the mental processes into words introduces new factors into the experience itself. (4) Sup- pose, lastly, that you are observing a feeling or an emotion : a feeling of disappointment or annoyance, an emotion of anger or chagrin. Experimental control is still possible; situations may be arranged, in the psychological laboratory, such that these feelings may be repeated, isolated and varied. But your observation of them interferes, even more seriously than before, with the course of consciousness. Cool consideration of an emotion is fatal to its very existence ; 178 ATTENTION AS SENSORY CLEARNESS your anger disappears, your disappointment evaporates, as you examine it. To overcome this difficulty of the introspective method, students of psychology are usually recommended to delay their observation until the process to be described has run its course, and then to call it back and describe it from memory. Introspection thus becomes retro- spection ; introspective examination becomes fost mortem examination. The rule is, often- times, a good one for the beginner; and there are cases in which even the experienced psycholo- gist will be wise to follow it. But it is by no means universal. For we must remember, first, that the observations in question may be repeated. There is, then, no reason why the observer to whom the word is called out, or in whom the emotion is set up, should not report at once upon the initial stage of his experience : upon the imme- diate effect of the ward, upon the beginning of the emotive process. It is true that this report interrupts the observation. But after the first stage has been accurately described, further observations may be taken, and the second, third, and following stages similarly described ; so that presently a complete report upon the w^hole experience is obtained. There is, in theory, some danger that the stages are artificially sepa- THE METHOD OF PSYCHOLOGY 179 rated; consciousness is a flow, a process, and if we divide it up, we run the risk of missing cer- tain intermediate links. In practice, however, this danger has proved to be very small, — wit- ness the stress laid by many psychologists upon * fringes' and 'relational feelings'; and we may always have recourse to retrospection as an auxiliary method, and compare our partial results with our memory of a like experience un- broken. Moreover, — and this is a point too often lost sight of, — the practised observer falls into an introspective attitude, has the introspec- tive habit, so to say, ingrained in the texture of his mind ; so that it does become possible for him, not only to take mental notes while the ob- servation is in progress, without interfering with consciousness, but even to jot down written notes, as the histologist does while his eye is still held to the ocular of the microscope. Let me cite a parallel case. All of us who are engaged in intellectual work, in the study and the teach- ing of a science, are obliged to read a very great deal, and to read critically and discerningly, in the state of selective attention. Now the ex- perience that I wish to bring to your minds is this : that, as one is reading, one is able to take mental note of passages to be remembered and employed, without appreciable pause in the pro- 180 ATTENTION AS SENSORY CLEARNESS cess of reading itself, and without even momen- tary loss of the thread of the writer's argument. I am not concerned here with the analysis of this experience, but with the mere fact, — with the fact that, when we close the covers of a book after two or three hours' reading, we have marked down half-a-dozen passages for further use with- out interruption of the main current of conscious- ness. That is the technical, critical attitude; and the introspective attitude is akin to it. There can be no doubt, again, from the results of such experiments as those of Solomons and Stein,^^ that the waiting of notes, brief catch- w^ords and symbols, need not in any way inter- fere with the introspection of the moment. And if we refer the disappearance of affective processes to the incompatibility of affection and attention, — I have spoken of this matter earlier, — rather than to the impossibility of direct introspection in general, we have, I think, made out our case all along the line; there is no difference, in principle, between inspection and mtrospection.^* So far, then, as my own psychological thinking is concerned, I do not believe that that ghost will walk again. Attention in psychology and atten- tion in natural science are of the same nature and obey the same laws. But now — what is attention ? THE PROBLEM OF ATTENTION 181 The analytical study of attention has been subject to two adverse influences : the pressure of popular psychology, and the obviousness of application. Popular psychology regards atten- tion, indifferently, as faculty and as manifesta- tion of faculty. It is a faculty, whose operation produces or prevents certain changes in the mental life ; it is also the activity or the state — the activity of remarking, noticing, observing; the state of sustained concentration — which manifests and attests that operation. ^^ Scientific psychology has, in very large measure, fought itself clear of the theory of faculties ; but the in- fluence of the popular conception is still shown in the tendency to treat the attentive* conscious- ness as a whole, to synthetise objective and sub- jective, incidental and essential, in a single view. There is, of course, a typical attentive conscious- ness, as there is a typical memorial or imaginative or^expectant or habituated consciousness. Nev- ertheless, the road to assured result lies through the elements of consciousness, and has conscious- ness itself as its goal; ^^' short cuts to synthesrs, however promising, end always in one-sided theory. The intrinsic tendency of psychology to deal with attention in the large has been further strengthened by the practical importance of 182 ATTENTION AS SENSORY CLEARNESS attention, its importance in educational regard. Here, if anywhere, the passage from theory to practice has seemed to be short and easy ; here, if anywhere, a sound psychology might be of immediate service to the responsive teacher. Since, however, the problems of education are necessarily formulated in terms of a completed psychological system, and since they are of the kind that requires speedy solution, this obvious- ness of application has been a real hindrance to psychology; it has held us to the old paths, and has discouraged that work of scattered exploration by which alone a science is enabled to advance. I think that these two things — tradition and application — are mainly responsible for the unsettled state of attentional psychology. But I think also that, in spite of these two things, analysis has gone far enough to furnish us with a clue to the attentional problem. It seems to me beyond question that the problem of attention centres in the fact of sensible clearness. Let me call my witnesses ! There are two men who have a special claim to be heard in this matter. The first is Wundt, and his claim is of long standing. Now Wundt declares that there are two " wesentliche Bestand- theile," two essential factors — and the word ATTENTION AS SENSORY CLEARNESS 183 'essential' is used in the sense of 'necessary/ not of ' important ' — in every process of atten- tion : first, the increased clearness of a particular idea or group of ideas, which is connected with the characteristic feeling of activity; and secondly the inhibition of other available impressions or memory-images. Attention, in other words, ^tneans a redistribution of clearness in conscious- ness, the rise of some elements and the fall of others, with an accompanying total feeling of a characteristic kind. That is the statement of the Physiologische Psychologie of 1903; and the discussion of attention in 1874 opens, in the same spirit, with the now familiar analogy of the Blickfeld and Blickjpunkt}'^ , Our second wit- ness is Pillsbury, who writes, without theoretical prepossession, from a general review of what had been said and done in the field of attention up to 1903. Pillsbury's statement is that " I'atten- tion accroit la clarte des sensations sur lesquelles elle porte." He goes on : "il est tres difficile de preciser . . . ce que Ton entend par clarte. Pourtant tout le monde salt ce que le terme sig- nifie et tout le monde a eprouve le changement qui s'opere pendant Tattention." ^^ If, however, we rank clearness as one of the intensive attri- butes of sensation, this difficulty is accounted for; we can no more define clearness, in the strict 184 ATTENTION AS SENSORY CLEARNESS sense of definition, than we can define intensity itself. You will not thank me, now, if I bring up a regiment of psychologists, in single file, each to deliver his testimony and disappear. I will fol- low a less tedious method. Baldwin's Diction- ary disiinguishesiiye types of attentional theory ^^ : let us, then, find a typical definition of attention, under each one of these five headings, and see if our emphasis of clearness is confirmed. First come the affectional theories, represented by Ribot. What is the definition.? ''L'atten- tion," says Ribot, " consiste en un etat intellectuel, exclusif ou predominant," — '*est un monoide- isme intellectuel," — *'avec adaptation spontanee ou artificielle de I'individu." A monoideism with adaptation implies, of course, a good deal more than clearness, but it very certainly implies clearness; and we read in Ribot's text of 'une idee maitresse,' *une representation vive,' 'un etat de conscience devenu preponderant,' — phrases in which the reference becomes explicit. "° The theories of ' psychical energy ' or of ' original activity ' come next in order ; here we may quote Ladd. An *act of attention in its most highly .complex form' is defined as "a purposeful voli- tion, suffused with peculiar feelings of effort or strain and accompanied by a changed condition ATTENTION AS SENSORY CLEARNESS 185 of the field of discriminative consciousness, as respects intensity, content, and clearness." Later on Ladd speaks of a "focussing of psychical energy upon some phases, or factors, or objects, of consciousness, and the relative withdrawal of such energy from other phases, factors, objects. "^^ So Stumpf, while he defines attention as a special kind of feeling, *die Lust am Bemerken,' notes that the primary effect of attention is ''die langere Forterhaltung [des bezliglichen Inhaltes] . . . und die aufmerksame Fixirung wahrend dieser Dauer"; or, rather, the primary effect is "ein Bemerken," while the longer duration is ''ein selbstverstandliches Mitergebnis der fort- gesetzten Urteilstatigkeiten, in welche der In- halt verflochten wird." ^" This Fixirung, Be- merken is, evidently, our 'clearness.' Thirdly, we have the 'conative' or 'motor' theories. Stout says that "attention is simply conation in so far as it finds satisfaction in the fuller presentation of its object, without actual change in the object." ^^ Baldwin defines attention as "the act of holding a presentation before the mind"; it increases the intensity of sensation and "the vividness of representative states. "^^ The next group, theories of 'intensity' and 're- enforcement,' is represented by Bradley. "At- tention (whatever it may be besides) at any rate 186 ATTENTION AS SENSORY CLEARNESS means predominance in consciousness. . . . Not theorising but applying descriptive metaphors, we may call attention a state which implies domi- nation or chief tenancy of consciousness. Or we may compare it to the focussing of an optical instrument, or to the area of distinct vision in the retinal field." ^^ Lastly, for the theory of 'inhibition,' we may quote Ferrier. "Just as we can at will fix our gaze on some one object out of many appealing to our sense of vision, and see this clearly while all others are indistinct or invisible, so we can fix our intellectual gaze, or concentrate our consciousness, on some one idea or class of ideas to the exclusion of all others in the field of intellectual vision." ^^ You will understand that I am not here con- cerned with the validity of the classification of theories given in the Dictionary, or with these theories themselves considered as explanations of the attentive consciousness, or with the authors' total descriptions of the state of attention. My point is simply this: that, wherever you look, you find some form of reference to clearness; clearness is, so to say, the first thing that men lay hands on, when they begin to speak about attention. I do not want to press the point ad nauseam. I will add, only, that if you take the quite recent books, those that have appeared ATTENTION AS SENSORY CLEARNESS 187 since Pillsbury completed his review, you will find just the same thing. *'Die Aufmerksam- keit," says Ebbinghaus, "besteht in dem leb- haften Hervortreten und Wirksamwerden ein- zelner seelischer Gebilde auf Kosten anderer, fur die gleichwohl audi gewisse Veranlassungen des Zustandekommens vorhanden sind."^^ '*The fact that consciousness always has a focal point, which reveals the momentary activity of the mind, is what is meant by the fact of attention, so far as it can be described in terms of the content of consciousness"; that is Angell's statement. ^^ And Judd, though his standpoint is different from ours, comes to the same conclusion. "The word 'attention' refers more especially to the selective character of the organising process, whereby one particular group of sensory factors is emphasised more than any other group"; "at- tention is merely a name for various phases of selective arrangement within experience." ^^ Em- phasis and selective arrangement are, again, our fact of 'clearness' translated into systematic terms. Finally, Meumann describes the ' Grund- erscheinung des Auf merksamkeitsvorganges ' as follows : "in dem Masse, als einige bestimmte Bewusstseinsinhalte oder Tatigkeiten in den Auf- merksamkeitszustand geraten, haben diese ho- here Klarheit, hoheren Bewusstseinsgrad, wer- 188 ATTENTION AS SENSORY CLEARNESS den vorlibergehend der Mittelpunkt des ganzen psychischen Lebens . . . wahrend in demselben Masse der iibrioje Bewusstseinsinhalt in niederem Grade bewusst ist und seinen Einfluss auf den Gang der psychischen Tatigkeit verliert."^° All this cataloguing is dry work; I can plead only that it was necessary. Indeed, I may claim your gratitude that there is not more of it; for I have, myself, been obliged to turn up a small library of references, in order to make sure that my position is well taken. With that assurance gained, let us proceed to the study of clearness as an attribute of sensation. Under what condi- tions does a sensation appear with maximal clearness in consciousness ? We shall do best to approach this question empirically, without theoretical bias, and without attempt at a systematic classification. Begin- ning in this way, we find the most obvious condi- tion of clearness in (1) the intensity of stimulus, and its sensory equivalents. Loud sounds, bright lights, strong tastes and smells, severe pressures, extreme temperatures, intense pains, — all these things are clear in virtue of their intensity; they attract or compel our attention, as the phrase goes, in spite of ourselves; they force their way to the focus of consciousness. THE CONDITIONS OF CLEARNEbS 191 whatever the obstacles that they have to over- come. A like value attaches to long duratioix-- and wide extensions, in so far as these are the equivalents of a high degree of sensible intensity. The qualification is important, because it reminds us of the phenomena of adaptation and fatigue. The first really hot days of summer, and the first really cold days of winter, constrain our atten- tion; but we soon grow accustomed to summer heat and winter cold. Enter a family circle, one member of which is partially deaf, and you are embarrassed by the loudness of the voices; but at the end of a week you will cease to notice anything unusual. Tire yourself out, and a stimulus that would ordinarily attract your atten- tion passes unregarded; under the conditions, it is no longer an intensive stimulus. Duration, then, if it is to mean clearness, must be the psy- chophysical equivalent of intensity, as it is, e.g., in certain forms of auditory rhythm. ^^ And the same thing holds of extension. I said just now that the appeal to casual intro- spection is a confession of scientific weakness, and the remark applies in the present connection. We do not know at what average degree of inten- sity clearness makes its appearance, and we do not know within what quantitative limits the psychophysical equivalence of intensity and du- TENTION AS SENSORY CLEARNESS 188 A 1 jn, intensity and extension, obtains. The jneral dependence of clearness upon intensity of stimulus is an evident fact, but it is a fact that we must leave in the rough. ^" It is natural to pass from intensity, duration, and extension to the quality of stimulus. And I think it cannot be denied that (2) form or quality of stimulus is one of the conditions of clearness of sensation. I gave some illustrations in my first Lecture. There are certain pains, by no means intensive, that are nevertheless urgently, insistently, importunately clear, — pains that we * cannot get away from' by any ordinary distraction of attention. There are certain organic complexes, also, which in my own experience have this power to compel the atten- tion ; they are intimate, worrying, wicked things.* The taste of bitter, the smell of musk, the sight of yellow belong, for me, to the same category; the least trace of them fascinates me. No doubt, there is here a wide range of individual differ- ence. But I cannot doubt that some sensible qualities are, intrinsically, clearer than others. James comes at least very close to this doctrine * My general name for all these experiences is ' quick' — not in the sense of 'fast/ but in that of 'intimately vital.' In my child- hood's speech 'the quick' was the tender flesh beneath the finger nails, and the wider use of the term is evidently based upon this association. THE CONDITIONS OF CLEARNESS 191 in his chapter on Instinct, and Mliller in his refer- ences to Eindringlichkeit. "Es erscheint mog- lich, class sich zwei Empfindungen, falls sie von verschiedener Qualitat sind, hinsichtlich der Eindringlichkeit anders zu einander verhalten, als hinsichtlich der Intensitat. . . . Man kann zwei Empfindungen verschiedener Qualitat, z. B. eine Rotempiindung und eine Grauempfin- dung, zwar hinsichtlich ihrer Eindringlichkeit einigermassen mit einander vergleichen, hat hin- gegen nicht in gleicher Weise ein Urteil darliber, ob der Abstand vom Nullpunkte flir diese oder jene beider Empfindungen grosser sei. " Ebbing- haus brings the facts under the heading of inter- est, the "Gefuhlswert der Eindriicke"; but the category is evidently too large for them.^^ In the third place we may consider (3) the temporal relations of stimulus, and especially repetition and suddenness. A stimulus that is repeated again and again is likely to attract the attention, even if at first it is altogether unre- marked. Pillsbury instances the case of a man absorbed in work; you may call his name once, and he will not hear you, — but call again and again, without raising your voice, and he will presently respond. Experiences of this sort are common enough, though their analysis is not quite easy. There is always the possibility, e.g.. 192 ATTENTION AS SENSORY CLEARNESS that the stimulus may operate at a moment when consciousness is free, so that it produces its effect less by sheer repetition than by suddenness or intensity. On the other hand, there seems to be no reason a priori why summation of stimuli should not be a condition of clearness. Ebbing- haus cites, in this connection, the fact of practice ; but that is, surely, a phenomenon of a very differ- ent order. **Der gelibte Kliniker sieht an einem neuen Fall, der gelibte Teclmiker an einer neuen Maschine sofort eine INIenge von ihm bekannten und gelaufigen Dingen, die der Ungelibte erst allmahlich oder audi gar nicht bemerkt." But the previous cases and the older machines were attentively examined. No amount of repeated visual stimuli would make a surgeon or a tech- nician; expert knowledge presupposes attention. The question here, however, is whether repeti- tion as such renders a stimulus clear, brings it to the focus of attention. I think that it may; but I should like to have experimental proof. It would also be interesting to know at what point the summation-effect gives way to habitua- tion, and whether habituation itself is ever pos- sible without foregone attention.^* Sudden stimuli and sudden changes of stimu- lus exert a familiar influence upon attention. As regards the latter. Stern tells us, "ganz allgemein. THE CONDITIONS OF CLEARNESS 193 dass die Veranderungserregbarkeit mit abneh- mender Geschwindigkeit abnimmt." ** Lang- same Veranderungen sind weniger geeignet als schnelle, . . . eine Reaction der Aufmerksam- keit . . . herbeizufiihren." The law rests upon a fairly large body of experimental results, ob- tained in various sense-departments.^^ The mention of change leads us, however, (4) to a fourth condition of great moment, the condition that Pillsbury sets in the first place : movement of stimulus. I quote a few instances from Stumpf. "Sternschnuppen, deren Bild auf seitliche Netzhautteile fallt, werden doch in Folge ihrer raschen Bewegung sofort bemerkt. Halt man einen Bleistift in solcher Entfernung von einer brennenden Lampe, dass sein Schatten auf einer weissen Papierflache auch im directen Sehen eben nicht mehr erkennbar ist, so wird er sofort wieder erkennbar, wenn man den Bleistift bewegt. Beim Tastsinn fand E. H. Weber, dass innerhalb der sg. Empfindungskreise, in welchen gleichzeitige Beriihrungseindriicke nicht mehr unterschieden werden, doch Beweguiigen noch leicht wahrnehmbar sind." Movement, indeed, is a stimulus of such individuality that some authors — Exner, e.g.^ — speak outright of move- ment sensations, and Heller and Stern distin- guish direct and indirect touch, as we all dis- 194 ATTENTION AS SENSORY CLEARNESS tinguish direct and indirect vision, in terms of sensitivity for resting and moving stimuli. At all events, there can be no doubt that the stimulus which moves in the field of vision or of touch has a remarkable power to draw the attention. Since our classification is empirical only, we may follow Stumpf 's example, and include under the present rubric the phenomena that Kiilpe describes as 'partial tonal change,' the "con- tinuous or discrete intensive and qualitative variation of a tone or clang within a connection of tones or clangs." A tone that beats, or re- curs intermittently, or fluctuates in pitch, within a chord or compound clang stands out clearly from its background. "Everyone must have noticed how strongly the attention is attracted in a concert by the voice which carries the melody. The singer's voice, even if compara- tively weak, can be heard without special effort above a full orchestral accompaniment, in pas- sages where it alone has to rise and fall, to execute trills and runs. . . . The same voice is obscured at once, if it is allowed to rest upon a single note." '' — We began our list, naturally enough, with a reference to the attributes and elementary rela- tions of stimulus. It is clear, however, that we are breaking away from stimulus. If the * mov- THE CONDITIONS OF CLEARNESS 195 ing' tone can be so named only by analogy to touch and sight, ' movement ' itself has a psycho- logical significance that extends far beyond its formal definition in terms of space and time. And 'suddenness,' in the same way, is more than a temporal relation ; the sudden stimulus is likely to be the surprising, the unexpected stimulus. These remarks apply, now, with still greater force to the fifth condition of clearness, — (5) the novelty, rarity, unaccustomedness, strangeness of stimuli. The value of this category is not undis- puted. "Ein ungewohnlicher Sinnesreiz," says Mliller, "muss, um in besonderer Weise auf uns zu wirken, in Folge seiner Starke oder anderer Momente die sinnliche Aufmerksamkeit bereits auf sich gezogen haben, so dass er und seine Neuheit und ungewohnliche Eigenthiimlichkeit uns zur Wahrnehmung kommt." While, how- ever, there is truth in this statement, I think that the truth is partial. Novelty and unaccustomed- ness mean, in psychological terms, *non-asso- ciatedness.' The novel impression is the im- pression that lacks associative supplements in consciousness; that stands alone, in isolation. Such an impression, provided that it is at all intensive, seems to me to become clear in its own right; it is 'startling,' just as the sudden stimulus is 'surprising' and the moving stimulus 196 ATTENTION AS SENSORY CLEARNESS disturbing.' For the rest, the effect of novelty is acknowledged by James, Kiilpe, Ebbinghaus, and Pillsbury.^^ A sixth condition of clearness, a condition of the very widest range, is (6) that described by Ebbinghaus as "the presence in consciousness of corresponding, i.e. similar ideas," and by Mtiller — under two separate headings — as ''the likeness of the incoming sensation to the idea, sensation, or image already present in the mind" and the "associative relationship between the incoming sensation and the existent idea, or more generally between the sensation and the whole circle of ideas dominant at the moment." The condition is of great systematic importance, since, for some psychologists, it forms the bridge that leads from passive to active, from invol- untary to voluntary attention. It is also, as I said, of the very widest range; for it covers all cases, from precise duplication, so to say, of in- coming sensation by preexistent image, up to the appeal of stimulus to a dominant psychophysical tendency which, at the time, may be unrepre- sented in consciousness. Classical illustrations of the first kind are afforded by Helmholtz' experiments upon stereo- scopic vision and the hearing of partial tones. Helmholtz found that, when the two halves of THE CONDITIONS OF CLEARNESS 197 a stereoscopic slide were illuminated in a dark chamber by the electric spark, he was able to see double images at will, "wenn ich mir vorher lebhaft vorzustellen suche, wie sie aussehen miis- sen." So, in the case of partial tones, he devised a method to ''obtain a series of gradual transi- tional stages between the isolated partial and the compound tone, in which the first is readily retained by the ear. By applying this process I have generally succeeded in making perfectly untrained ears recognise the existence of upper partial tones." Apart from these experimental results we know that, in everyday life, the man who finds is the man who knows what to look for; the sailor at the masthead, the hunter on the trail, the pathologist at the microscope, are all cases in point. At the other end of the scale stand the per- manent adult interests, connate or acquired, which — even when not represented in idea — are ready to be touched off by a casual stimulus. The collector, the inventor, the expert are aroused to keen attention by stimuli which the rest of the world pass without notice. I have already men- tioned the psychological attitude, the introspec- tive habit, which so grows on one with time and experience that at last everything — novels and games and children's sayings and the behaviour 108 ATTENTION AS SENSORY CLEARNESS of an audience in a lecture-room — becomes tributary to psychology, and one can no more help psychologising than one can help breath- ing. "Some years ago," Jastrow writes, *'I became interested in cases of extreme lon- gevity, particularly of centenarianism, and for some months every conversation seemed to lead to this topic, and every magazine and news- paper offered some new item about old people. Nowadays my interest is transferred to other themes; but the paragrapher continues quite creditably to meet my present wants, and the centenarians have vanished." It is the vanishing, of course, that is the source of danger. If you are 'favourably impressed' by a scientific theory, the facts that support the theory crowd in upon you, while the outstanding facts, those that can- not connect with the trend of consciousness, fail to present themselves ; you mean to be impartial, and the conditions of attention make you one- sided. I said, in a previous Lecture, that scien- tific theories sit more lightly upon their defenders than opponents are apt to suppose.* I must add, then, that this result is secured by the cultivation of the critical attitude, the scientific habit j)ar excellence, which becomes as potent as any other in the control of attention. ^^ * p. 48. THE CONDITIONS OF CLEARNESS 199 Our list of conditions has led us from attributes of stimulus to psychophysical disposition. We might, possibly, bring under this latter head- ing (7) the accommodation of the organs of sense, though I incline to think that an empirical classification would rank it as a separate factor. Wundt gives fixation as one of the external con- ditions of visual clearness. Klilpe expresses himself more sceptically. "We shall, perhaps, be more correct in supposing that [these motor conditions] are only indirectly conducive to the apperception of particular contents, as determin- ing the attributes of the contents themselves." But is not clearness precisely one of these attri- butes.?^ It is true that the accidental conver- gence of the eyes upon some object in the field of vision, while we are mentally occupied with other things, does not bring that object to the focus of consciousness. Nevertheless, in so far as the phenomena of 'fluctuation of attention,' of which we speak later, are referable to peripheral condi- tions, we must admit that accommodation of the sense-organ is at least a negative condition of peripheral clearness. ^^ I come, finally, (8) to the much-discussed cases in w^hich the absence or cessation of stimulus con- strains the attention. We do not notice the ticking of the clock upon our wall, but we notice 200 ATTENTION AS SENSORY CLEARNESS . its silence. We do not notice the ordinary noises from the street, but we notice the unusual quiet after a snowfall. Fechner gives some sa- lient instances. *' Der Miiller erwacht, wenn der Gang der Muhle stockt, der Schlafer in der Kirche, wenn der Prediger zu sprechen, das von der Amme eingesungene Kind, wenn die Amme zu singen aufhort, der bei Nachtlicht zu schlafen Gewohnte, wenn das Nachtlicht erlischt, der im Wagen Fahrende, wenn der Wagen still steht." How are these effects to be explained ? Notice, first, that they are not simply instances of the cessation of unnoticed stimuli. In every case, foregone attention, and prolonged or fre- quent attention, is presupposed. We do, e.g., attend to the ticking of the clock, again and again, in the course of the day; we hear it when we look up to see the time, and we hear it, with all plainness, in intervals of thinking and reading and writing. The miller is interested in the run- ning of his mill ; he has listened, often enough, to make sure that things are in good order. The traveller, before he dozed off to sleep, was made very uncomfortable by the jolting of the coach ; he wished more than once that he was at home in his comfortable bed. On the other hand, the cessation of an unnoticed stimulus, of a stimulus that you have not attended to, is not THE CONDITIONS OF CLEARNESS 201 necessarily remarked. Flowers may be put upon your mantelpiece, curtains hung across your win- dows, — put in place, and taken away again, — without your observing either their coming or their going. But more than this : I doubt if any really un- noticed stimulus attracts the attention by its cessation. The appeal to sleep is very doubtful, and Fechner's examples are general at the best. Suppose that the maid breaks an ornament in the drawing-room, an ornament that you have long ago ceased to think of; you do not notice its absence. If it was large, and stood in a con- spicuous position, you are struck by the novel look of that part of the room, and you cast about for an explanation. If it was small and incon- spicuous, you do not discover your loss until some chance association recalls it to mind, and you search and fail to find it. Once more : if objective cessation may attract the attention, subjective cessation may persist under circumstances that would normally bring the stimulus to clear consciousness. Delboeuf tells us that he was once staying at a country house which stood near a waterfall. The noise was so great that, on the first day of his stay, he was hardly able to follow the conversation at table. However, he soon grew accustomed to it. 202 ATTENTION AS SENSORY CLEARNESS Waking up one night, about a week after his arrival, he was surprised to find that he could not hear the water, "meme en y pretant une at- tention soutenue "; only after he had got up and looked out of window did he succeed in recover- ing the auditory perception of the fall. Here was psychophysical disposition, but no clearness ! And Delboeuf reminds us of a very common ex- perience of the same kind : the experience of waking at night, and listening for the tick of the clock. Is there anybody w^ho has got out of bed, under these circumstances, in the assured con- viction that watch or clock has stopped ? Evidently, these several cases must be clearly distinguished, and referred each to its own special set of conditions. If we go back to our original instances, the ticking of the clock and the noises from the street, it seems to me that w^e notice their cessation, for the most part, only when we are looking for them ; as we glance towards the clock, as we pause in our work and listen for the familiar noise, we become aware of the si- lence. It is not that the absence of stimulus commands attention, but that an expectant atten- tion, a psychophysical predisposition, is disap- pointed, baffled by the silence. This view is borne out by the observation that the clock may have been silent for a long while before we notice THE CONDITIONS OF CLEARNESS 203 that it has stopped. If, however, you think that the explanation goes too far, let us try another. It has been shown, experimentally, that we attend best under a slight distraction ; maximal clear- ness requires a little * effort,' as we say, for its attainment. The clock and the street noises may be considered as distractions, stimulating distractions, of this kind. Their removal would then, after the current * spurt' of energy had ceased, make itself felt as a general restlessness or unsteadiness, a widespread complex of organic sensations. I think you will agree that this general restlessness sometimes appears, and that we work from it to the cessation of the familiar stimulus; though I myself do not find it as fre- quently as I find a baffled expectation. A third and very similar interpretation derives from Lehmann's law of the ' indispensableness of the habitual.' According to that law, you will remember, the removal of an accustomed stimulus leaves a need, a Bedilrfnis. This comes to con- sciousness, in organic terms, as uneasiness or discomfort; and again we have a positive start- ing-point for the attention.*" We must content ourselves with general ex- planations, since this little group of facts has never been brought under experimental control. Indeed, what impresses one most strongly, all 204 ATTENTION AS SENSORY CLEARNESS through the review which we have now completed, is the need of detailed experimental work. I have, of course, omitted a good many experimental ref- erences that I might have given ; but those of you who know what I have left out will realise how very much more there is that I could not put in. Even as things are, however, there is a ray of day- light. Just as we found the various theories of attention held together by the central fact of clear- ness, so we find that all these empirical conditions of conscious clearness may be grouped together as conditions of a powerful impression of the nervous system. Let us look at them in order. Intensive stimuli — and their equivalents in space and time — must, naturally, set up psycho- physical processes of relatively great strength ; and intensive excitations will not be easily in- hibited or obscured by the other excitatory pro- cesses of the moment. So the qualitative stimuli that are effective for clearness must make appeal to some peculiar susceptibility of the nervous system, general or individual. — Repeated stimuli produce a cumulative effect, and thus take their place, as regards nervous excitation, alongside of intensive. Sudden stimuli impinge upon nerv- ous elements that have hitherto been free from stimulation of their particular kind, i.e. upon nervous elements of a high degree of excitability ; THE CONDITIONS OF CLEARNESS 205 and it is probable that the excitations which they set up suffer less dispersion and diffusion, within the nervous system, than the excitations resulting from gradual application of stimulus. — Moving stimuli arouse different nervous elements in quick succession, so that there is no possibility of fatigue or of sensory adaptation ; in a sense, therefore, the effect of the moving stimulus is cumulative. — Novel stimuli are isolated stimuli; they have neither to share their effect with asso- ciates nor to hold their own against rivals. The excitation set up by the novel is thus of the same order as that set up by the sudden. — As for the effect of the anticipatory image, it is clear that, the more nearly the excitation correlated with the given stimulus coincides with a psycho- physical excitation already in progress, the more easily will it make its way within the nervous system and the more dominant will it become. And, in the same way, excitations that coincide with modes of excitatory activity habitual to the particular nervous system, excitations that are in the line of a 'psychophysical disposition,' will evidently have a greater effect than others that are less accustomed. — Lastly, peripheral accommodation opens the gateway to the cortex, and permits the stimulus to operate at its full strength from the first. ^^ 206 ATTENTION AS SENSORY CLEARNESS I do not think that it is worth while, in an elementary discussion, to go further into physio- logical theory. Nor shall I attempt to recast our empirical classification of conditions, and make it scientific. The list stands very much as Lotze left it. It is true that Lotze himself, and later psychologists from Wundt to Pillsbury, draw a distinction, in more or less definite lines, between physiological and psychological, external and in- ternal, objective and subjective conditions. But what is external in one system becomes internal in another, and it is not difficult to argue either that all alike are objective or that all alike are subjective. All are objective, in that they oper- ate by way of the nervous system ; all are subjec- tive, in that the specific organisation of the nerv- ous system determines their effect. I will only suggest, then, that the common element which, empirically, holds all the conditions together — the ultimate condition of clearness at large — may be designated as nervous disposition, pre- disposition of the nervous system and its sensory attachments.^^ It is the task of genetic psy- chology to classify the determinants of attention in the order of time, as ordinal, generic, individ- ual; it is the task of experimental psychology to delimit and quantify their influence; and it is the task of physiology to exhibit the mechanism of their nervous operation. I VI THE LAWS OF ATTENTION: I LECTURE VI THE LAWS OF ATTENTION: I MY last Lecture was, in effect, a plea for a sim- plification of the psychology of attention. Kiilpe tells us, in his Grundriss, that psycholo- gists have tended to find "the real object of in- vestigation into the psychology of space, not in the spatial attributes, but in the spatial relations. . . . The result has been an almost total neglect of the perception of extension and figure, and an al- most exclusive regard of the perception of dis- tance and position.'* And he remarks further that, in the psychology of time, ''interval has been given the preference over duration with as perplexing results as follow from the preference of distance over extension in the psychology of space." ^ I believe that much of our diflSculty in the psychology of attention arises, in the same way, from our concessions to tradition and to practical demands; and that we should do well to sit down in serious earnest to a psychology of clearness, — considering clearness as an attri- bute of sensation, conditioned upon nervous predisposition, just exactly as quality is an attri- p 209 210 THE LAWS OF ATTENTION: I bute of sensation, conditioned upon nervous differentiation. How far this elementary psy- chology of attention could be carried it is, evi- dently, impossible to predict; but the number of experimental problems suggested by the pre- ceding Lecture shows that there are many and definite points of attack. However, a science does not advance accord- ing to any prearranged logical plan, but haltingly and unevenly, as the interests of individual workers prompt, or the claims of practical utility dictate. And the experimental psychology of attention centres, as a matter of fact, about some half-dozen large problems, — in part relatively new, in part handed down from the empirical psychology of the eighteenth century, — which have been discussed again and again, to the neg- lect of other and equally important questions. We are still inclined to speak, not of ' the ' experi- mental psychology of attention, but of Wundt's or Stumpf's or James' or Muller's views upon attention. I shall not attempt, now, to lay out an ideal programme for further work; the at- tempt would be overbold, and the programme would not be followed. I desire rather to review what we know, what has already been done ; and I shall therefore treat the elementary psychology of attention topically, under those half-dozen CLEARNESS AS SENSATION ATTRIBUTE 211 headings. For the sake of clearness, 1 shall throw each headmg into the form of a law, a general statement of the behaviour of conscious contents given in the state of attention. But the statement is not to be understood dogmati- cally, for we shall be largely occupied with argu- ments and results that make against its universal validity; the *law' is rather a challenge, an appeal to the bar of fact. My first *law,' in this sense of the term, runs as follows: (1) clearness is an attribute of sensa- tion, which, within certain limits, may be varied independently of the other concurrent attributes. What are the facts ? In the first place, there can be no doubt of the independent status of clearness as sensation- attribute. As Wundt says: *'Klarheit und Starke der Eindrlicke sind durchaus voneinander verschieden " ; "das Klarer- und das Starker- werden eines Eindrucks sind . . . subjectiv wohl zu unterscheidende Vorgange." ^ There are, in my experience, very few departments of psy- chological observation in which the distinction of clearness from the other attributes of mental processes offers appreciable difficulty. Nevertheless, independent status does not necessarily mean independent variability. It is 212 THE LAWS OF ATTENTION: I true that most sensible qualities may be present at any degree of clearness ; but, as I have already pointed out, the rule is not universal, — there are qualities that appear to be bound up with a determinate clearness, or at any rate to admit of only a very narrow range of clearness-degree. And when we turn to intensity, we are upon de- batable ground from the start. Is clearness ever independent of intensity ? or, in popular phrase, do we ever attend to a sensation without thereby making it stronger ? You will find all sorts of opinion : that atten- tion intensifies sensation, that attention leaves sensible intensity unaffected, that attention re- duces the intensity of sensation. On the whole, however, the trend of psychological belief just now seems to favour an interdependence of the two attributes. Pillsbury, who devotes a good part of his first chapter to a balancing of the evi- dence, pro and con, ends with a no7i liquet: "il semble done que cette discussion sur les rapports entre la clarte et Tintensite reste sans conclusion."^ I think, though, that his own leanings — if one may presume to read between the lines — are towards a coupling of the two attributes, at least within certain limits. This is also Kiilpe's view in the Grundriss. " Within certain narrow limits, . . . contents are really intensified in the state CLEARNESS AND INTENSITY 213 of attention." ''The sensation of a loud sound, inattentively experienced, may seem equal . . . to that of a faint sound, attentively experienced. Again, it is interesting to note that the alteration of judgment by inattentive observation is al- ways precisely the same as the alteration produced by a reduction of the intensive, spatial, or tem- poral values of the impressions, except that it is somewhat more uncertain. . . . This fact re- quires further investigation." ^ Wundt writes, to the same effect, ''dass beide [Eigenschaften] einen gewissen Einfluss auf einander aussern konnen. ... So bemerkt man, wenn ein Reiz das Bewusstsein bei grosser Unaufmerksamkeit trifft und dann in gleicher Starke wiederholt wird, wie z. B. beim unerwarteten Stundenschlag einer Thurmuhr, dass der zweite Eindruck entschieden nicht bloss deutlicher, sondern scheinbar auch intensiver wahrgenommen wird. Das namliche zeigt sich, wenn man sich willklirlich anstrengt, Erinnerungs- und Phantasiebilder zu erwecken und moglichst intensiv im Bewusstsein festzu- halten." ^ Ebbinghaus admits that the experi- mental evidence is doubtful, but argues, from general experience, that "eine allgemeinen Erho- hung der Empfindungsstarke durch Zuwendung der Aufmerksamkeit durchaus wahrscheinlich ist." He gives the illustrations that Wundt had 214 THE LAWS OF ATTENTION: I given before him : the varying intensity of the bell strokes, heard with attention and with inatten- tion, and the hallucinatory character of images in a state of sustained and concentrated attention.^ Pillsbury and Ebbinghaus both reply to the ob- jection that an intensifying effect of attention would falsify our perceptions, would jeopardise the validity of Weber's Law. Pillsbury suggests that the increase of intensity is not absolute, not the addition of a constant amount, but relative, proportional to the intrinsic intensity of the stimulus; and Ebbinghaus points out that nor- mal intensity is, after all, intensity in the state of attention. *' [Es] beziehen sich alle genaueren Angaben tiber Empfindungen, tiber ihre Eigen- schaf ten. Sell wellenwerte U.S. w., . . . durchweg auf eine erhohte ihnen zugewandte Aufmerk- samkeit. Verschiedenheiten aber, die nun noch etwa durch verschiedene Grade einer solchen er- hohten Aufmerksamkeit hervorgebracht werden konnten, werden als unerheblich betrachtet wer- den durfen." My impression is that views of this sort are gaining ground in psychology, as against, e.g., Stumpf's doctrine that only weak sensations are intensified by attention. Stumpf, you will remember, looks at the operation of attention from the negative side ; the weak sensation rises, by the removal of counter-influences within the CLEARNESS AND INTENSITY 215 nervous system, to the full (or approximately the full) intensity which it would have possessed in its own right had those adverse influences been absent.^ However, I will quote an authority on the opposite side. **[Eine] verstarkende Funk- tion der Aufmerksamkeit," says Miinsterberg, "giebt es nicht; neuere Experimente bestatigen die schlichte Erfahrung, auf die schon Fechner hinwies, dass ein graues Papier an der Stelle, der sich die Aufmerksamkeit zuwendet, nicht heller erscheint ; das schw^ache Licht w4rd nicht intensiver, ein Gewicht nicht schwerer, eine Linie nicht langer, ein Ton nicht lauter, wenn unsere Aufmerksamkeit die Vorstellung erfasst."^ Miin- sterberg' s statement is in flat disagreement with those which I have just been reading. Experiment must decide; but direct experi- ment is very difiicult. Let me remind you of an historical incident which has always seemed to me to be characteristic for psychology at large, and — if looked at in the right way — encourag- ing to the student of psychology. Mach and Stumpf sat down together before a harmonium, in the physical laboratory at Prague, to decide the question whether attention to one of the com- ponent tones in an ordinary musical chord does or does not strengthen that particular tone. The chord was sounded, and the two men listened. 216 THE LAWS OF ATTENTION: I ''Wiihrend Mach die Verstarkung ganz deutlich zu horen angab," says Stumpf, who tells the story, *'konnte ich nichts davon finden." *'So- viel ist sicher, dass bei ganz unveranderten Um- standen eine Verstarkung starker Tone neben anderen gleichzeitigen starken Tonen fur mich nicht wahrnehmbar ist, wahrend Mach sie auch dann wahrzunehmen erklixrt." And he con- cludes, resignedly, that individuals differ.® I do not know about the individual differences: but I call the observation characteristic, because it may stand as a typical instance of divergent introspections ; and I call it encouraging, because the student may take heart from it to hold by his own introspective conviction on unsettled points. No doubt, both Mach and Stumpf heard what they say they heard. *No doubt,' I say, though I myself cannot hear as Mach hears. But since we have, in science, to pass beyond individual experience, the direct method must be given up for an indirect ; we must seek to arrange condi- tions in such a way that the introspective dis- crepancies disappear. The experiments made by the method of distraction are exceedingly interesting.^^ But though they are not very numerous, I cannot here attempt to review them. Criticism of opinion may be condensed into relatively few words; CLEARNESS AND INTENSITY 217 criticism of experimental method needs time and detail. I will rather give you a brief account of experiments recently carried out in the Cornell Laboratory by my colleague Professor Bentley,^^ — experiments which point quite definitely to the positive conclusion that, even in the case of strong stimuli, attention has an intensifying effect. The stimuli, which were presented in pairs, were the sounds produced by the ordinary gravity phonometer. To the one stimulus of each pair, the observer was maximally attentive ; from the other he was distracted. The distrac- tion was effected by means of odours, which we had found in previous investigations to be su- perior to such things as counting, adding, mul- tiplying, etc. Suppose, then, that two sounds, a weaker and a stronger, are given ; and that the weaker is the sound attended to, the stronger the sound distracted from. If the observer judges the two sounds to be of equal intensity, still more if he judges that the objectively weaker sound is the more intensive of the two, we have an over- estimation of intensity in the state of attention. Out of 300 preliminary experiments, 285 were successful. Of these 285 judgments, 136 showed an overestimation of the stimulus attended to, 40 showed an underestimation of that stimulus, and 109 reported the relation of the two stimuli \ 218 THE LAWS OF ATTENTION: I correctly. Further experiments came out in the same way. And experiments with two pairs of intensive stimuli, weak and loud, which might be supposed — under Weber's Law — to measure equal sense-distances, gave almost identical re- sults ; the overestimations wdth the loud stood to the overestimations with the weak stimuli in the ratio 39 to 40. Pillsbury's conjecture is thus confirmed. Lastly, it was found that the per- centage of * right cases ' with distraction was prac- tically the same as that with continuous atten- tion. Puzzling at first, this result becomes clear if we remember that the error of distraction would operate as often to increase as to decrease the differences of sensible intensity ; so that its effect must appear rather in the distribution than in the number of incorrect judgments, rather under the headings of overestimation and underesti- mation than in the column of wrong cases. It is clear, then, that strong as well as weak sounds are intensified by attention, or, if you prefer the negative statement, are reduced in intensity by distraction. What this precisely means, physiologically and psychologically, it is at present impossible to say. Professor Bentley's observers certainly did not confuse intensity with clearness; and the intervention of reproductive or affective influences seems to be ruled out both CLEARNESS AS SENSATION ATTRIBUTE 219 by the conditions of the experiments and by the introspections. So far as they go, the results tell directly for what I have called the current psychological view of the relations of intensity and clearness. — What, then, of our law ? Why, the law stands, under the conditions and with the limitations of which w^e have spoken. Clearness is an inde- pendent attribute of sensation. It is also, in some measure, an independently variable attri- bute. It may vary in entire independence of most sensible qualities; it may vary also inde- pendently of intensity, in the sense that a very weak sound may be as clear as a very loud sound. Only it seems bound up with intensity to the extent that change of clearness involves always a change of intensity as well; very weak clear sounds are not as weak as they would be at a lower degree of clearness. How far the converse of this statement is true, within w^hat limits a change of intensity brings with it, normally, a change of clearness, cannot be said, though the correlation probably extends beyond those ex- tremes of intensive stimulation which we dis- cussed in the preceding Lecture. I may add that there is nothing surprising, from the psy- chophysical point of view, in an intimate relation between clearness and intensity; all the condi- 220 THE LAWS OF ATTENTION: I tions of maximal clearness are also, as you will remember, conditions for the powerful impres- sion of the nervous system. I turn, in the second place, to the law which I have named — and the name shows my own bias and opinion — (2) the law of the two levels. It is generally agreed that increased clearness of any one part-contents of consciousness implies the decreased clearness of all the rest ; the ' energy of attention,' as we say, is limited and practically constant. So the question arises : how many levels or degrees of clearness may coexist in the same consciousness ? Opinions are widely divergent. Baldwin, e.g., gives in his Senses and Intellect a * graphic repre- sentation of area of consciousness, after analogy with vision,' in which no less than four levels are distinguished. At the very centre of conscious- ness stands apperception. Beyond that lies active consciousness or attention ; beyond that, again, passive or diffused consciousness ; and be- yond that, the subconscious. The whole series of concentric circles is then enclosed by the un- conscious or physiological, a region of uncertain boundary. "It is well," Baldwin says, "to note the play of ideas through all these forms of transi- tion, from the dark region of subconsciousness. THE TWO LEVELS 221 to the brilliant focus of attention [i.e., to apper- ception]. Images pass both ways constantly, acting varyingly upon one another and making up the wonderful kaleidoscope of the inner life." ^^ There is no question that images pass through a large number of degrees of clearness — certainly many more than four — in their passage from maximal to minimal attention; the question is, however, whether they show all these degrees within a single consciousness. Angell seems to accept Baldwin's view in this strict interpretation. "The field of conscious- ness," he says, ** is apparently like the visual field. There is always a central point of which we are momentarily more vividly conscious than of anything else. Fading gradually away from this point into vaguer and vaguer y^^ 5 Ir~ea ^Conscious- consciousness* is a margin ^^^^- ~ '^- ^^- Baldwin, ^ Handbook of Psychology: of objects, or ideas, of which Senses and Intellect, 1890, we are aware in a sort of mental indirect vision." ^^ Baldwin's diagram is printed in illustration. Kiilpe takes the opposite standpoint. He be- gins his article on The Problem of Attention by * Italics mine. 222 THE LAWS OF ATTENTION: I contrasting physiological with psychological clear- ness. "[As I sit] looking at the flowered pattern of the paper on the wall in front of me, ... I notice that around the spot of clearest vision the pattern loses in clearness, at first slowly, then more and more quickly, until I reach the limit of my field of vision, and cannot make out any pat- tern whatsoever. If I did not know that the whole wall is covered with the same paper, I should suppose that the paper-hanger had chosen less and less pronounced patterns, the farther he moved from the point upon which my eyes are fixed, until finally all pattern and colour were lost in an indifferent gray." I must interject here that I cannot, personally, verify the details of this observation ; I think that Klilpe has read into the wall-paper a good deal of his own know- ledge of sense-psychology. But, at any rate, he refers the observation itself to physiology, and does not use the analogy as Baldwin and Angell use it. On the contrary, he writes of the attentive consciousness as follows: "When we ask how the degrees of consciousness are related to one another, we find, not an uniform grada- tion from the highest to the lowest, but, in most cases, a fairly sharp line of distinction. Certain contents stand at the level of clear apprehension ; and from them our consciousness drops away. THE TWO LEVELS 223 without transition, to the level of obscure general impression, above which the other contents of the time are unable to rise. And the clearer the first group of contents, the more indistinct are all the rest. ... If, therefore, at any given moment we make a cross-section of the stream of consciousness, we shall find represented on it, not all conceivable degrees of clearness, but as a rule just two groups of processes separated from each other by a considerable interval." The statements are cautious; Kulpe puts in the qualifying 'in most cases,' 'as a rule'; but the caution is plainly due to the lack of experimental evidence, and cannot obscure the writer's own opinion.^* Six years earlier, Kulpe had argued in a similar spirit against Kohn. "Die Klarheit und die Un- klarheit wachsen in entgegengesetzter Richtung, und die Zustande, die ihnen entsprechen, konnen somit um so leichter voneinander unterschieden werden, je ausgepragter die Aufmerksamkeit ist. Diese Erscheinung zwingt uns geradezu statt von einer einformigen quantitativen Abstufbarkeitdes Bewusstseins von zwei gesonderten Zustanden desselben zu reden." Kiilpe distinguishes here also between physiological and psychological distribution of clearness. ^^ Ward posits three grades of consciousness in 224 THE LAWS OF ATTENTION: I the wide sense : "a centre or focus of conscious- ness within a wider field, any part of which may at once become the focus," and a third grade of * subconsciousness.' This subconsciousness is, however, literally sub-conscious. The threshold of consciousness **must be compared to the surface of a lake, and subconsciousness to the depths beneath it." "Presentations in sub- consciousness have not the power to divert attention, nor can we voluntarily concentrate attention upon them." " This hypothesis of sub- consciousness ... is in the main nothing more than the application to the facts of presentation of the law of continuity." ^^ Consciousness has, then, not three experienceable levels, but two only; Ward's subconscious presentations are simply Fechner's negative sensations. ^^ Mar- shall, if I understand him aright, inclines to go a little farther. "In the moment of reflection," he says, "we find in all cases what have been called the fields of Attention and of Inattention. We find them and nothing more." Neverthe- less, "the field of inattention seems to resolve itself into an aura, as it were, which aura has now a 'feel' of being fuller, and now of being narrower. . . . The observation that this aura at times seems to be fuller, and again narrower, surely points to the existence of something psy- THE TWO LEVELS 225 chic beyond either the fields of attention or of inattention, points to the existence of mentality out of which consciousness whether of attention or of inattention arises." ^^ I do not find that Marshall is more logical than Ward, — though he does not follow Ward's example of including in consciousness what is by definition below the level of consciousness, — for a ' feel ' of fulness and narrowness must be a conscious feel, and observation of the feel must be introspective observation. However, both Ward and Mar- shall are arguing theoretically : Ward for the law of continuity and Marshall for a form of psycho- physical parallelism ; they are not directly facing our present problem. Helmholtz, who does face that problem in a particular case, afl&rms that **wir fur das Be- wusstwerden ' einer Empfindung zwei verschie- dene Arten oder Grade unterscheiden mussen," the kinds or degrees which Leibniz named per- ception and apperception.^^ And this is, of course, the doctrine that we also associate with the name of Wundt. Indeed, the representa- tion of consciousness in two levels, clear and obscure, is so characteristic of Wundt's psy- chology that I think we sometimes tend to credit him with its invention, — just as we credit him with the metaphor of the Blickfeld and the Blick- 226 THE LAWS OF ATTENTION: I punkt, although Wundt, I suppose, took that from Fortlage, and Fortlage may have taken it from Lotze, and Lotze from some earlier writer ; for it goes back at least as far as Tucker ! ^^ The metaphor itself, with its direct implication of the two staores of consciousness, stands in 1874 at the very beginning of Wundt's section on at- tention, and in 1903 stands second only to the ThdtigkeitsgeJiXhl?^ '*Sagen wir von den in einem gegebenen Moment gegenwartigen Vor- stellungen, sie befanden sich im Blickfeld des Bewusstseins, so kann man denjenigen Theil des letzteren, dem die Aufmerksamkeit zugekehrt ist, als den inneren Blickpunkt bezeichnen. . . . Der innere Blickpunkt kann sich nun successiv den verschiedenen Theilen des inneren Blick- feldes zuwenden. Zugleich kann er sich jedoch, verschieden von dem Blickpunkt des ausseren Auges, verengern und erweitern, wobei immer seine Helligkeit abwechselnd zu- und ab- nimmt. ... Je enger und heller hierbei der Blickpunkt ist, in um so grosserem Dunkel be- findet sich das tibrige Blickfeld." This is fa- miliar teaching. It must be taken in connection with Wundt's refusal to grant any psycho- logical place to the subconscious or the uncon- scious. He freely admits the influence of feeling on the course of ideational association, but he THE TWO LEVELS 227 will not allow the feeling to stand alone in con- sciousness, the counterpart of an * unconscious ' idea. The fact is rather '*dass die betreffende Vorstellung im Bewusstsein vorhanden sei, dass sie aber zu jenen dunkleren Inhalten desselben gehorte, die uberhaupt mehr durch ihre Wirkung- en auf andere Bewusstseinsvorgange als durch ihre eigenen Bestandtheile erkennbar werden." ^^ Hamilton's doctrine of 'mediate association' — "one idea mediately suggests another into con- sciousness, the suggestion passing through one or more ideas which do not themselves rise into consciousness" — is treated in the same way. **Man hat wohl ein Recht von ' unbemerkten ' oder von *dunkler bewussten' Mittelgliedern solcher Associationen zu sprechen, nimmermehr aber von *unbewussten.'" ^^ And Herbart fares in like manner : " es gibt keine ' f rei auf steigende ' Vorstellungen." ^^ Morgan, as we all know, gives a prominent place in his psychology to the distinction be- tween 'focus' and 'margin' of consciousness. "In any moment, . . . there are, in addition to and alongside the dominant elements constituting the summit of full clear consciousness, dimly felt elements which may have little or no direct con- nection with those dominant elements. These we will speak of as subconscious,'''' ^^ — a quite 228 THE LAWS OF ATTENTION: I unnecessary term, since for Morgan the processes in question are all conscious. ** Directly we begin to examine and measure any part of the margin," he says, "it thereby ceases to be mar- ginal and becomes focal;" but we cannot exam- ine and measure the subconscious. You will have been wondering, too, why I have not men- tioned James' chapter on the Stream of Thought.^^ I have omitted it, because I think that it belongs to another part of our subject, to which I come in a moment. I do not know, now, how the *law of the two levels' is to be put to any conclusive test. If Baldwin and Angell find their three degrees of consciousness below the apperceptive level, and if Marshall finds his aura below the second level, of obscure consciousness, I can only fall back upon Stumpf's 'individual differences' and envy those whose minds are richer than my own, I find nothing of the sort. Working from the other end, from the level of clear apprehension, I am accustomed to use the following illustra- tion. Take one of the familiar puzzle pictures, a picture which represents, we will say, a house and garden, and somewhere in which is concealed the outline of a human face. As you search for the face, the contents of the whole picture are at the conscious focus. Suddenly you find it: THE TWO LEVELS 229 and what happens ? Why, as you do so, the picture drops clean away from the focus; the face stands out with all imaginable clearness, and the house and garden are no clearer than the feel of the paper between your fingers. The ex- perience is very striking, as I have described it; it is more striking still if the face baffles you, and you go off on false scents. For every time that you think you have found the hidden out- line, the picture slips from you, — slips, to come back with a mental jerk as you realise your fail- ure. There is no poising of the picture, after the riddle has been read, midway between crest and base of the wave of consciousness. Suppose, however, that we accept the law; suppose that we agree upon the dual clearness of consciousness. The duality will range from maximal to minimal difference, according to the degree of 'concentration' of attention. When we are totally absorbed, we are also absent- minded ; the upper level is admirably clear, the lower is exceedingly obscure. When we are less fixedly attentive, there will be a less marked difference between the two conscious levels. Now, then, the question arises : Do all the part-contents, within the two divisions of consciousness, show one and the same degree of clearness ? The main division, we will assume. 230 THE LAWS OF ATTENTION: I is plain enough ; consciousness is arranged step- fashion. But is the surface of consciousness, at the two levels, smooth and unwrinkled, or are there differences of emphasis both over the area of relative clearness and over the area of relative obscurity ? We will begin at the bottom. Morgan makes his 'subconscious' elements ''subconscious in different degrees"; he speaks of "a short rising slope of dawning consciousness and a longer falling slope of waning consciousness." ^^ I am not sure, however, how far this purports to be matter of observation, and how far it is merely a diagrammatic representation, due to the 'wave' metaphor. Angell, you will remember, speaks of a 'gradual' fading away into 'vaguer and vaguer consciousness ' ; and if he had not accepted Baldwin's fourfold arrangement, we might, perhaps, interpret these words to indi- cate simply relative difference of obscurity within a single general obscurity. Wundt seems, at first reading, to be quite definite. " [Es] lasst sich experimentell mit Sicherheit nach- weisen, dass [die dunklen,] mit dem appercipir- ten Inhalt meist nur in einem losen und aus- seren Zusammenhang stehenden begleitenden Vorstellungen die allerverschiedensten Grade der Klarheit darbieten konnen, von einer oberen THE TWO LEVELS 231 Grenze an, wo sie noch als zwar undeutliclie, jedoch in ihren allgemeinen Eigenschaften noch einigermassen erkennbare Objecte erfasst war- den, bis zu einer unteren, wo nur festzustellen ist, dass iiberhaupt in einem bestimmten Sin- nesgebiet irgend etwas vorhanden war, das im Bewusstsein wirksam wurde, aber schon im Moment, nachdem der Eindruck voriiberge- gangen, nicht mehr zur Apperception gebracht werden kann." ^^ I say that this seems definite; and I think there can be no doubt as to Wundt's general opinion, — that, within the total ob- scurity of marginal consciousness, differences of relative position may be made out. iVt the same time, he speaks of the 'passage of the im- pression,' and a footnote refers us to his dis- cussion of tachistoscopic experiments. I am, then, after all not sure that Wundt is not illus- trating one thing by another, illustrating mar- ginal differences of clearness by reference to dif- ferences within the upper level of consciousness. For myself, I find the issue very difficult to decide. Since the relative clearness of any particular process depends upon conditions, — the conditions that we listed in the last Lecture, — and since our nervous system is an extraordi- narily complex mechanism, and may be very variously affected at any given time, I see no 232 THE LAWS OF ATTENTION: I reason a priori why there may not be differ- ences of obscurity in the lower level as there undoubtedly are differences of clearness in the upper. Nevertheless, I am by no means sure that I discover these differences. Observation in the large is practically impossible; one must catch favourable moments as they occur in the course of experimentation. And I mean by * favourable' moments occasions when the dif- ference between the two main levels is not overgreat, — when, for one reason or another, the observer's attention is less fixed and con- centrated than the experiment properly requires. I have caught myself, time and again, slipping from the prescribed object of attention to some secondary circumstance or obtruding idea; but when I ask whether, a few seconds earlier, that circumstance or idea was clearer amidst ob- scurity than the look of my surroundings or the organic background of consciousness, I am unable to give a definite answer. The problem might, perhaps, be attacked indirectly by an inversion of the method of distraction. If, e.g., we could show that very various degrees of reinforcement were needed to shift the focus of attention to contents that were known to be in the conscious margin, then we might argue that these contents themselves were originally pres- THE TWO LEVELS 233 ent at different degrees of clearness or, rather, of obscurity. It is further possible that the experiments would give opportunity for the introspective verification of the differences thus objectively determined. We have better evidence to go upon when we look at the contents of the upper level, the apperceived contents; for we may appeal to the results of all the experiments upon the 'span of consciousness' or the 'range of atten- tion.' I begin with Dietze's determination of what Wundt still terms the 'Umfang des Be- wusstseins.' ^^ The observer in these experiments listened to series of metronome beats, which were separated into groups by the sound of a bell; the problem was to discover the upper limit at which two successive series might be discriminated, without counting, as of different lengths. We are interested just now, not in the numerical results, but in the state of con- sciousness. Dietze writes on this topic as fol- lows: **Bedingung der Zusammenfassung einer gegebenen iVnzahl von Vorstellungen in eine Reihe ist, . . . dass, wenn nach Ablauf der Reihe eine neue, in gleichem Zeitintervall fol- gende . . . Vorstellung appercipirt wird, in die- sem Moment die erste Vorstellung eben erst auf der Schwelle des Bewusstseins angelangt ist. 234 THE LAWS OF ATTENTION: I . . . Der Grad der Klarheit der gleichzeitig anwesenden Vorstellungen wird nun einmal abhangen von der jeweiligen Entfernung der Vorstellungen vom Blickpunkt des Bewusstseins und zweitens von der Energie, mit welcher die Vorstellungen appercipirt worden sind;'*^^ the ideas show a gradation of clearness, which in part is a simple function of time elapsed, in part depends on the emphasis of subjective accentua- tion. Wundt writes to the same effect, though less cautiously, in the Physiologische Psycholo- gie. '*In dem Moment, wo ein neuer Reiz ... in den Blickpunkt des Bewusstseins tritt, werden stets die vorangegangenen noch in abgestufter Klarheit vorhanden sein." '*Die unmittelbar vorangegangenen Eindriicke sind . . . keineswegs aus dem Bewusstsein, ja die nachsten nicht einmal ganz aus dem engeren Focus der Aufmerksamkeit verschwunden, son- dern sie treten nur allmahlich in den dunkleren Umkreis des inneren Blickfeldes zuriick. Hier verdunkeln sie sich dann um so mehr, je weiter sie durch die inzwischen abgelaufene Reihe von dem momentan appercipirten Eindruck getrennt sind, bis sie endlich bei einem bestimmten Punkte aus dem Bewusstsein verschwinden." ^^ I shall not discuss the general question, how far or in what sense these experiments serve to THE TWO LEVELS 235 measure the span of consciousness. But I must take issue with Wundt's introspective interpreta- tion. We are to suppose that the metronome beats march out of consciousness, in single file, each one growing dimmer and dimmer until it finally crosses the conscious limen and dis- appears. Now look at that statement logically. You remember that the sounds are not all equally clear; some of them are subjectively accentuated. Suppose, then, that an eight- membered rhythmical unit is passing out of consciousness. The first member is relatively very clear; the second member is relatively obscure. Will the first pass out before the second ? It ought not to do this; it is only one place ahead in the order of time, while it is three places ahead in the order of clearness, of rhyth- mical accent. Logically, the prior but clear term should remain after the later but obscure term has disappeared. I know that logic is not psychology ; but then psychology is not, either, illogical. You may, however, urge that the unit in the present in- stance is the complete rhythmical form, not the single metronome stroke ; and that the rhyth- mical units may tail off in consciousness in the way that Wundt describes. Wundt speaks, definitely, of the individual Schallreiz; but I 236 THE LAWS OF ATTENTION: I will accept the amendment. The appeal then lies to introspection : do we actually find the gradations of clearness, as between unit and unit, incoming and outgoing ideas ? Schumann, who engaged Wundt in a controversy on this matter, finds no trace of them. "So oft ich auch bei den obigen Experimenten versucht habe, etwas von den in den dunklen Umkreis des inneren Blickfeldes zuriicktretenden Vor- stellungen zu bemerken, so ist es mir doch nie gelungen und ebensowenig den Versuchsperso- nen [among whom was Mliller], welche ich darauf aufmerksam machte." Schumann's own explanation is couched in terms of sensory and motor Einstellung, feelings of fulfilled or dis- appointed expectation.^^ On the negative side, my own introspection agrees with that of Schumann and his observers ; I cannot put my finger on the train of vanishing ideas. On the positive side, I think that Schu- mann's account holds under certain conditions, — not under all. But even if we accept the theory of a 'simultane Gesammtvorstellung,' there is no need to assume Wundt 's series of graded ideas; all sorts of surrogates are possible. I am sorry to have led you so far afield ; but there was no alternative. If Wundt 's analysis were correct, we should have not only to give THE TWO LEVELS 237 up our law of the two levels, but also, I believe, to recast a good portion of his own systematic teaching. I do not propose to do either, but to utilise Dietze's results in another direction. The important thing for our present purpose is this : that the experiments show, without any question, the coexistence of different degrees of clearness at the higher level of consciousness. While a rhythmical unit may be clear as a whole, its constituent elements vary in clearness-degree. The same thing holds of simultaneously pre- sented visual impressions, all of which fall within the area of distinct vision. ''Man be- merkt," says Wundt, "ausser den deutlich apper- cipirten Eindrlicken zunachst eine Anzahl anderer, die sich als 'halbdunkel' bezeichnen lassen : hier ist man im stande, einzelne nach- traglich durch angestrengte Aufmerksamkeit auf das reproducirte Bild des Gesammteindrucks zu erkennen. Daneben existirt aber immer noch ein weiteres, 'ganz dunkles' Feld, bei dem man nur uberhaupt feststellen kann, dass irgend ctwas da war." ^^ This last, the 'wholly ob- scure ' field, belongs unquestionably — if I may trust my own observation — to the lower level of consciousness ; the ' something there ' is a something of precisely the same kind as the look of the tachistoscope itself, or of the black walls 238 THE LAWS OF ATTENTION: I of the observing tube. But the other two grades, the * clear' and the *half obscure,' belong, as certainly, to the upper level. Let us see pre- cisely what the distinction means. When a tachistoscopic field is exposed for the first time to an unpractised observer, he will very probably fail to 'make out' anything at all; the lines or letters or geometrical figures are seen as a general impression, without dis- crimination of detail. Was, then, the field obscure ? Surely not ! The observer was ' at- tending' with all the concentration he could summon ; the field was the clearest thing in his consciousness. What he failed to do was to cognise. Cognition is not clearness; it is an associative process of the assimilative kind. Apperception and cognition are so usually con- joined, in our adult experience, that we may sometimes forget to separate them; but psy- chologically they are different things. When, then, the practised observer tells us that some of the details in the exposure-field are 'clear' and others 'half obscure,' he means that he has cognised the former and failed to cognise the latter ; all alike were clear, but the clearness did not, in all cases, suffice for cognition. The fact that the half-obscure elements are recoverable in the 'image of reproduction' shows that they THE TWO LEVELS 239 were well within the field of clearness; the fact that they were not directly cognised shows that this field is not uniformly illuminated, that parts of it are more strongly accented than others, — just as, in the temporal field, there are degrees of clearness among the members of a rhythmical unit.^^ We must conclude, therefore, that — whatever is the case at the lower level — there are notice- able differences of clearness in the processes at the upper level of consciousness. It would be strange if there were not ! And one of the most interesting new departures in experimental psy- chology, to my mind, is the work now in progress in the Leipsic laboratory upon this very point. Attempts are being made to measure the differ- ences of clearness in focal contents, whether by determining the limen of change at various parts of a spatial field or by comparing the times of reaction obtained with varying distribution of attention .^^ Here is the beginning of a new chapter in scientific psychology; and here is Wundt handling the problems of attention as masterfully as when he first began to experiment nearly fifty years ago. We are now, I think, at the point where it is fitting to refer to James and his conscious * fringes.' Ordinarily, the distinction which 240 THE LAWS OF ATTENTION: I James draws is taken to be the same as that drawn by Morgan in the terms 'focal' and * marginal/ "The margin of mental processes," says Angell, "outside the focal point of atten- tion, constitutes what James calls the 'fringe of consciousness.'"^^ No doubt there are plenty of passages in which this use may be found. But if we turn to the locus classicus, the chapter on the 'Stream of Thought,' I think it is clear that James is dealing throughout with the upper level of consciousness, the field of attention. He comes to his "psychic overtone, suffusion, or fringe" by way of 'transitive states,' 'feelings of relation,' 'feelings of tendency.' And his point is that "every definite image in the mind is steeped and dyed in the free water that flows round it." I do not understand that the 'free water' is flowing at a lower level, but simply that — within the area of attention — it is less stable and therefore less clear than the 'definite image.' "The fringe ... is part of the object cognised, — substantive qualities and things ap- pearing to the mind in a fringe of relations. Some parts — the transitive parts — of our stream of thought cognise the relations rather than the things." James is considering the 'cognitive function of different states of mind.' "Knowledge about a thing is knowledge of its THE TWO LEVELS 241 relations. ... Of most of its relations we are only aware in the penumbral nascent way of a 'fringe' of unarticulated affinities about it." This penumbra is surely the analogue of the marginal impressions in tachistoscopic experi- ments, the impressions that are apperceived but not cognised. I do not want to labour clumsily at a thing that James has treated with all his accustomed lightness and freshness of touch, — but I think it is pretty obvious that, in this part of his psychology of cognition, James is primarily concerned with the upper conscious level. He is distinguishing degrees of clearness within the clear, not distinguishing clearness from ob- scurity. That distinction he discusses in the following section, in its relation to interest and attention, accentuation and emphasis.^^ — In fine, then, a diagram of consciousness would show, in terms of the foregoing analysis, a two-level formation, broader below and nar- rower above, — the relative width and height of the two stages differing at different times. The surfaces are not smooth; the upper cer- tainly, the lower probably, is creased or wrinkled. The number and depth of the wrinkles will depend upon circumstances: upon the condi- tions of clearness as an attribute of sensation, and upon the more complicated conditions 242 THE LAWS OF ATTENTION : I which govern the degree of clearness of the part- contents of ideas. So much we can, perhaps, say with a fair amount of confidence.^^ But the really hopeful thing, for experimental psy- chology, is the programme of further work, the long array of definite problems that our review, ever so hasty as it is, has already brought to light. We can hardly be on a wrong track if perspectives open as they are opening to-day to the students of attention. We may look, in the third place, (3) at the temporal relations of attention, as expressed in the laws of accommodation and of inertia. *'Die Aufmerksamkeit," says Stumpf, "braucht eine gewisse Zeit, um sich dem Eindruck s. z. s. zu accommodiren, um ihr Maximum zu er- reichen." ^^ The fact here alluded to is familiar to all of us in connection with the reaction ex- periment; for the simple reaction, the optimal accommodation-time — the optimal interval be- tween signal and stimulus — is about 1.5 seconds, while for transit-observations it is apparently a little shorter, about 1 second. Attention is, however, flexible, labile; we are able, as Wundt points out, to adapt ourselves, within certain limits, to rhythms of different period, just as we can adapt ourselves to different intensities and qualities of stimulus.^^ ACCOMMODATION OF ATTENTION 243 It would be quite wrong, however, to identify this * accommodation ' of attention with the rise of a particular sensation to maximal clearness. We have a number of determinations, beginning in the early sixties and extending down to the present day, of the Ansteigen of sensations ; *^ and as they were all made in the state of con- centrated attention, the times which they fur- nish may be taken as the times required for a stimulus, acting under the most favourable con- ditions, to produce its full conscious effect. These, therefore, are the times that a psychology of clearness must analyse and interpret. The accommodation-time is rather the time required for peripheral or central Einstellung, — ''for the accommodation of a sense-organ, or for the establishment of a psychophysical disposition ; it gives us the temporal limen, not of clearness, but of certain conditions of clearness. Let me quote you an observation of Pills- bury's. '*Si, tandis qu'on lit, vient un desir soudain de savoir Theure, les images de la pendule rappelees a la conscience se presentent ^ avant que le mouvement ne commence, et il y a un intervalle considerable entre I'instant ou les yeux sont fixes sur la pendule sans adaptation complete et celui ou I'image est assez nette pour que Ton aitconnaissancede I'heure."^^ Pillsbury 244 THE LAWS OF ATTENTION: I IS arguing that peripheral adaptation is subse- quent to attention itself. If, however, we read 'clearness' for 'attention,' the facts wear a little different appearance. I want to know the time, and I look across the room at the clock. The clock is, at once, the clearest thing in conscious- ness; but it is not yet maximally clear, clear enough for cognition. To see the position of the hands, I must wait for the 'accommodation of attention,' i.e. for the adjustment of the mechan- ism of visual accommodation. This peripheral adjustment is one of the conditions of maximal clearness. Before accommodation is effected, I am in much the same position as one who is listening to a lecturer whose voice is too weak to carry across the room. The sounds heard are, again, the clearest things in consciousness; but they, too, fall short of the degree of clearness necessary to cognition, because intensity — one of the conditions of maximal clearness — is lacking to them. I The law of 'accommodation of attention' is a Ireal law; it covers a large number of facts of / observation. But it is a law of the conditions of clearness, or, if you like, a law of the total attentive consciousness, rather than of clearness itself ; and in an elementary psychology of atten- tion we shall do well to pass it over, and to limit INERTIA OF ATTENTION 245 ourselves to the interpretation of the Anstieg. Very much the same thing, I think, holds of the law of inertia, to which I now turn. The law is iormulated by Fechner as follows : "es behagt uns bis zu gewissen Grenzen mehr, in einer einmal eingehaltenen Richtung und hiemit Beschaftigung der Aufmerksamkeit zu verharren, als sie zu verlassen, die Beschaftigung zu unterbrechen " ; and by Stumpf, in similar terms, as follows: **die Aufmerksamkeit halt leichter etwas bereits Gegebenes fest, als sie etwas zu Suchendes findet."^^ It covers a wide range of experience : that you can follow the movement of a single instrument in the orches- tra better, when there has been solo-playing before, than when the whole number of instru- ments begin together; that you can finish a conversation, once begun, at a distance which would render the words of an unexpected ques- tion altogether inaudible ; that you can trace the upward course of a fire-balloon to a point at which it would otherwise be quite invisible. These are matters of perception ; but there are analogies in plenty in the realm of ideas. It is difficult to break away from a current train of thought, and to give your full attention to a letter or a visitor ; it is difficult to come back to your scientific work when you have been bothered by details of business or administration. 246 THE LAWS OF ATTENTION: I These are important facts; and they have been taken account of, m various ways, by ex- perimental psychology. But they seem to me to be facts which, on the psychophysical side, re- late to the conditions of clearness, — peripheral adaptation, psychophysical disposition, Perse- verationstendenz,^^ — and, on the psychological, to the time relations of the total attentive con- sciousness. They are therefore beyond the range of our present consideration. An elementary psychology will deal with the sensation, under its aspect of clearness; it will determine the least time interval between two maximal clear- nesses in the same and disparate senses ; and it will measure the carrying power of clearness, the amount of fluctuation which may be intro- duced into a continuous stimulus w^ithout im- pairment of sensible continuity. Here, again, there are methods ready for our use, and a body of experimental results awaiting our interpreta- tion.^^ And the law of inertia offers us problems of ever increasing complexity. For, since inertia is the opposite of motility, and the carrying power of clearness is the opposite of our liability to distraction, the determinations which I have just mentioned must be made under all sorts of conditions, and we shall be led on, as it were, by force of circumstance, from sensation to the INERTIA OF ATTENTION 247 simpler complexes, and from these to con- sciousness itself. So far, I think that my proposal of a simplified psychology of attention has been justified, — although I realise to the full the schematic char- acter of this treatment of the subject. In the next Lecture we shall see how it fares with still other laws of attention. VII THE LAWS OF ATTENTION: II LECTURE VII THE LAWS OF ATTENTION: II I HAVE suggested that an elementary psy- chology of attention will deal, not with the facts of attentional accommodation, but rather with the ' rise ' of the single sensation ; that it will begin, not with the gross facts of attentional inertia, but rather with the absolute temporal limen and the carrying power of clearness under simple conditions. I have been careful to say that the results of experiment in these fields must be 'interpreted' by a psychology of attention; the factors that make for clearness must be separated from the other conditions involved, and must if possible be separately estimated or 'weighted.' We get a hint towards this analysis in the fourth law that I shall mention, — (4) the law of prior entry. The stimulus for which we are predisposed requires less time than a like stimulus, for which we are unprepared, to pro- duce its full conscious effect. Or, in popular terms, the object of attention comes to conscious- ness more quickly than the objects that we are not attending to. 251 252 THE LAWS OF ATTENTION : II We have rough-and-ready illustrations of this law in various features of the reaction experi- ment/ Many of the effects that we ascribe to * practice,' in the most diverse kinds of experi- mental work, also fall under the same heading. A strict test, of the elementary sort, w^ould con- sist in the comparative measurement of the Anstieg and of the absolute temporal limen, first with complete predisposition, and secondly under measurable distraction.^ Unfortunately, as we shall see later, measurable distraction is still a problem for the future. In the meantime, we have a qualitative demon- stration of the law of prior entry in Stevens' inversion of the complication experiment.^ The arrangement is very simple. We take a bell metronome, and a cardboard arc whose radius is the length of the metronome pendulum. Scale divisions of 5° are laid off upon the cir- cumference, and the arc — with the zero-point of the scale corresponding to the position of equilibrium of the pendulum — is impaled upon the eye which serves to lock the lid of the metro- nome. The white cardboard thus forms a background, in front of which the pendulum oscillates. A piece of red paper, cut to the shape of an arrow-head, is spitted upon the end of the pendulum. The metronome is set to the PRIOR ENTRY 253 rate of, say, 72 in the one minute, and the bell rings at every complete oscillation. The posi- tion of objective coincidence of bell-stroke and arrow-head may be found, approximately, by slowly moving the pendulum with the hand until the bell sounds; in our instrument, it comes at about 22°. The experiment is then performed in two ways. First, the observer Fig. 6. Simple Complication Pendulum. attends to the moving pointer ; the sound of the bell is secondary, — it floats, so to say, upon the main current of visual change. Under these conditions, the pointer carries the bell out ; an average determination of subjective coincidence is 30°. Secondly, the observer attends to the bell ; the movement of the pendulum is now secondary, — the expected bell-strokes stand 254 THE LAWS OF ATTENTION: II out upon an indifferent shifting field. Under these conditions the temporal displacement of the sound is negative, not positive ; the point of subjective coincidence lies, on the average, be- tween 10° and 15°. It is very clear that the stimulus for which we are predisposed has the advantage over its rival. I call this observation the 'inversion' of the complication experiment, because in it the direc- tion of attention is prescribed, whereas, in the complication experiment proper, there is no preliminary instruction, and attention is appealed to only after the event, as an explanatory prin- ciple. You will, however, expect me to say something about the temporal displacement in complications, — a fact which, until the appear- ance of Geiger's paper in 1903,^^ was one of the most disputed and least understood in the whole range of experimental psychology. I think that Geiger's introspective analyses give us a defini- tive insight into the mechanism of the 'compli- cated' consciousness, although, doubtless, there is work of detail still to be done. You remember the circumstances. A pointer revolves at uniform rate before a scaled clock- face. At some moment of its revolution, un- known to the observer, a bell is sounded. The observer is to report the point of subjective coin- PRIOR ENTRY 255 cidence of sight and sound, the moment of forma- tion of an Herbartian 'complication.'^ Intro- spection varies very considerably with variation of conditions; but the important and curious result is that, under certain conditions, the bell- stroke suffers a negative displacement ; it is conjoined, in consciousness, with a division of the scale which the pointer has already passed when the objective sound is introduced. The sound is 'thrown back'; it is heard *too early.' What is the explanation ? Very different explanations have been sug- gested. One of the first investigators, von Tchisch, sought an analogy in the premature reaction.^ Theoretically, he says, two things are possible : the visual perception may be delayed, or the auditory anticipated. Now there is no assignable reason for delay. "Es ist dagegen leicht zu erklaren, dass momentane Reize vor ilirem Erscheinen appercipirt werden." In re- action experiments with a constant interval be- tween signal and stimulus you find a gradual reduction of the reaction time; the values be- come very small, and finally reach zero, — "die Reactionszeit . . . wird negativ." Just the same thing happens in the complication experi- ment; each recurring bell-stroke is the 'signal' for the next following, the 'stimulus.' "Durch 256 THE LAWS OF ATTENTION: II dieses Wiederholen wird die Apperception nicht nur vorbereitet, sondern dieselbe reproducirt unmittelbar den Eindruck. Mitliin sind das die Bedingungen unter welchen wir horen, flihlen, ehe der Reiz thatsachlich zustandekommt." Ideas of this sort were natural enough in 1885, when the psychology of the simple reaction was still crude, — though von Tchisch might, per- haps, have learned better from the Physiologische Psychologie of 1880.'^ But, whether natural or not, they are psychologically impossible. "The explanation," says James, ** requires us to be- lieve that an observer . . . shall steadily and without exception get an hallucination of a bell- stroke before the latter occurs, and not hear the real hell-stroke afterwards. I doubt w^hether this is possible, and I can think of no analogue to it in the rest of our experience." ^ We may all subscribe to this criticism. Indeed, the ex- planation satisfied nobody ; every later psycholo- gist who has discussed the complication experi- ment has sought to improve upon it. Sometimes, recourse has been had to physiological factors, the quick rise of auditory and the slow rise of visual sensations; sometimes to psychological, the interruption of the visual perception of movement and the substitution therefor of a perception of position; sometimes the explana- PRIOR ENTRY 257 tion has gone still further afield, to the observer's desire to make a good showing, to do as well in observation as his fellows.® All unnecessary labour ! Wundt had given the right cue, in his doctrine of the " Spannungswachsthum der Auf- merksamkeit" ^^; it was only needful to follow up the cue into the labyrinth of observational detail. Suppose that a naive observer takes his place before the complication clock; and suppose that the rate of revolution is moderate, so that the bell-stroke sounds once in every 1.5 seconds. The observer follows the pointer with his eye, and in the very first revolution refers the sound to some region of the circle. Notice that it is a region; the sound seems to spread over, to be coincident with, a fairly wide range of scale marks. The second revolution narrows this region ; the third narrows it still more, — and so on, until finally there are only a few scale divisions, one or two on either side of the objec- tively correct position, with which the bell appears to coincide. In the meantime, attention has been sharpening to the sound; and, more than that, an accommodation of attention has taken place ; the observer is predisposed to hear the bell at a certain instant. The instant ar- rives ; the sound is apperceived, rises to maximal 258 THE LAWS OF ATTENTION: II clearness, in a minimum of time ; and the result is that scale marks which the pointer had trav- ersed before the hammer struck are themselves apperceived, come to the focus of attention, only along with the objectively later complicating stimulus. *' [Der Schall] iiberholt gleichsam [die Theilstriche] auf dem Wege zur Apperception/' ^^ It is all a matter of prior entry, due to definite predisposition of the attention ; and the puzzle arises simply from the continuity of the visual movement. Had there been one sound and one sight, and had the sound come to consciousness before the (objectively simultaneous) sight, no one would have wondered. It is the backthrow of the sound which surprises us ; and yet that back- throw is, under the conditions, the inevitable result of attentional accommodation. I need not go into this question at greater length ; you will find a full and clear discussion in Geiger's article. I quote only a few sentences from the Physiologische Psychologie of the same year. " [Die negativen Zeitverschiebungen] sind natiirlich nicht so aufzufassen, als wenn man einen Reiz wahrnehme, noch ehe er w^rklich stattfindet ; sondern in eine Reihe von Gesichts- eindrlicken, die im Bewusstsein die simultane, aber stetig fliessende Vorstellung eines Zeitver- laufs bilden, tritt ein momentaner Schall- oder LIMITED RANGE 259 Tasteindruck ein, der als solcher nur mit irgend einem einzelnen Punkt dieser Zeitvorstellung associirt werden kann : mit welchem, dies hangt lediglich von den Bedingungen tlieils des Ein- drucks selbst, theils seiner Apperception ab. Je mehr die Aufmerksamkeit auf ihn gespannt ist, um so mehr wird er an den Anfang der ihm zugeordneten Zeitstrecke des Gesichtssinnes ver- schoben, je mehr jene Spannung erschwert ist oder aus irgend einen Ursachen abnimmt, um so mehr riickt er gegen das Ende derselben. . . . Es wird stets gleichzeitig gehort und gesehen; aber der Umfang, in dem die beiden nebenein- ander hergehenden Vorstellungsreihen zusam- men im Bewusstsein anwesend sind, lasst der Verbindung beider einen Spieh-aum, innerhalb dessen nun theils den ausseren Bedingungen theils und vornehmlich der Aufmerksamkeit der entscheidende Einfluss zukommt." ^^ Unless, then, I am unduly optimistic, the negative displacement of the bell-stroke, in complication experiments, need give psycholo- gists no further trouble. I pass to a brief con- sideration of (5) the law of limited range. You recall the facts. If a group of objects, all of which lie within the scope of clear vision, is momentarily exposed by means of some 260 THE LAWS OF ATTENTION: II tachistoscopic arrangement, a practised observer is able to cognise from four to six of them 'by a single act of attention.' It makes no difference whether the objects are lines, or geometrical figures, or numbers, or letters, or short words; under all these conditions, the range of clearness is approximately the same. Hence you find, in the current text-books of psychology, the state- ment that the grasp of visual attention covers from four to six simultaneously presented simple impressions.^^ Ebbinghaus takes issue with this statement. "Die so erhaltenen Werte . . . sind als Um- fangsbestimmungen der Aufmerksamkeit zwei- fellos zu hoch." He points out that "das gleichzeitig aufgefasste immer nur ... als ein Ganzes mit mehreren Teilen erkannt wird"; we are dealing, not with four to six separate objects, but with separately distinguished parts of an unitary whole. He argues, however, from the results of the complication experiment, "dass [die Seele] auf zwei voneinander ganz unab- hangige Reihen einfacher Eindriicke iangere Zeit hindurch gleichzeitig aufmerksam bleiben kann, ja daneben noch imstande ist . . . man- cherlei Ueberlegungen zur besseren Losung der Aufgabe anzustellen." In simple cases, then, **kann die Aufmerksamkeit ohne Schwierigkeit LIMITED RANGE 261 zwei, aussersten Falls vielleicht drei, voneinander ganz unabhangigen Dingeii zuge wandt werden . " ^* If we accept this discussion as it stands, the number of objects simultaneously apprehensible by the attention reduces from four or six to two or three. But I think that Ebbinghaus has mis- read his authorities. One would hardly gather, from his text, that Wundt, in the Physiologische Psychologie, had laid equal stress upon the uni- tary character of the tachistoscopic field. Yet Wundt says, definitely: *'immer bildet dieses Feld der Apperception eine einheitliche Vorstel- lung, indem wir die einzelnen Theile desselben zu einem Ganzen verbinden. So verbindet die Apperception eine Mehrheit von Schallein- driicken zu einer Klang- oder Gerauschvorstell- ung, eine Mehrzahl von Sehobjecten zu einem Ge- sichtsbild." And again : "man bemerkt ubrigens leicht, dass sich die Eindrlicke auch dann, wenn sie nicht Bestandtheile einer schon gelaufigen Vorstellung sind, doch zu einem zusammenge- horigen Bilde vereinigen." ^^ Can anything be plainer.^ Ebbinghaus' ''reichhaltige und ge- gliederte Einheit," which is apprehended as a whole while certain divisions or subdivisions stand out in clear isolation, — *'mehr oder minder deutlich gesondert, " ^® — is no new dis- covery, but a transcript of the Wundtian doctrine. 262 THE LAWS OF ATTENTION: II The other side of Ebbinghaus' exposition is similarly open to criticism. The bell-stroke and the visual impressions of the complication experi- ment are, as stimuli, disparate, addressed to differ- ent sense-departments. But psychologically they are by no means" voneinanderganzunabhangig." On the contrary, the sound, after a very few ob- servations, becomes organically related to the movement of the pointer; the two things seem to go together naturally, to belong to each other ; there is no disparateness in idea. There re- main the 'mancherlei Ueberlegungen ' ; but so far as my experience goes — and Geiger bears me out ^^ — these reflections exist only in Ebbing- haus' imagination. The case stands, therefore, as follows. When- ever in the state of attention two stimuli are given simultaneously or in immediate succession, they form a connected whole ; that is the most general law of association, in Ebbinghaus' phras- ing.^ ^ From this point of view, then, the field of attention is limited always to one complex, a single associated whole. The question of the range of attention thus becomes the question of the conscious articulation of the unitary complex. In psychological terms it runs : How many part- contents are, under the most favourable condi- tions, distinguishable within the whole .^ In TEMPORAL INSTABILITY 263 psychophysical terms : How many stimuli may become clear in consciousness at one and the same time ? And the current answers, although they are liable to experimental revision, may be taken as valid for their day and generation. We are on much more difficult ground when we turn (6) to the law of temporal instability. This law, also, may be approached from the side either of descriptive psychology and the atten- tive consciousness, or of experimental psychology and the attribute of clearness. According to Wundt, attention is discontinu- ous from force of circumstances and intermittent by its very nature. It is discontinuous because ideas come and go in consciousness, and atten- tion grasps but one idea at a time : " zwischen der Apperception je zweier auf einander folgender Vorstellungen wird immer eine Zwischenzeit liegen, in der die eine schon zu weit gesunken, die andere noch nicht zureichend gehoben ist, um klar appercipirt zu werden." It is also in- trinsically intermittent. '' Dauernd eine Vor- stellung mit der Aufmerksamkeit festzuhalten ist, wie die Erfahrung lehrt, schlechthin unmoglich : . . . ein dauernder Eindruck kann nur festge- halten werden, indem Momente der Spannung und der Abspannung derselben mit einander 264 THE LAWS OF ATTENTION: II wechseln. Auf diese Weise ist die Aufmerk- samkeit ihrem Wesen nach eine inter mittirende Function." '^ In his remarks on discontinuity, I think that Wundt has in mind what he himself elsewhere terms a limiting case, the typical associative consciousness of the English school. ^^ For, so far as introspection goes, we may attend, con- tinuously and unremittingly, for very consider- able periods: we read a novel or a scientific monograph at a sitting, we follow a whole act of grand opera, we work at our special subject for two or three or four hours at a time, without sensible interruption of attention. There are objective interruptions, of course : we stop read- ing to pursue a train of thought, to work out a difficulty, to cut the pages of the book ; we look away from the stage to exchange a remark with our neighbour; we get up to verify a reference, or we pause to slip a fresh sheet of paper into the typewriter. And these interruptions illustrate, often enough, the labile, instable character of attention; we drop from our high level of concentration to *take things easily,' to *let our mind wander' for a while. But instability is not discontinuity ; and, in the experiences now under consideration, instability itself is not the universal rule. We may very possibly give our TEMPORAL INSTABILITY 265 full attention, without lapse of any sort, to the question asked of us, or to the accurate adjust- ment of the new sheet; or, contrariwise, we may hokl firmly to our original topic, and speak and act automatically. All these cases demand closer analysis; but on the whole James' statement — that "thought is sensibly continuous" ^^ — seems to me to be nearer the facts, and nearer also to Wundt's general psychological doctrine, than the counter-assertion that attention is discontinu- ous. I should question the appearance of dis- continuity even in extreme instances of successive association. '* Dauernde Aufmerksamkeit," says Ebbinghaus, "gibt es nur bei einem stetigen Wechsel der Inhalte, in deren Hervortreten das Aufmerksamsein besteht."^" The 'nur' we have still to discuss ; but we may surely agree that attention can be sustained, and that the shift of ideas is continuous. What, then, of the Apperceptionsivellen ? Is attention intrinsically intermittent, and is it im- possible to hold a single, simple content steadily in the focus of consciousness ? We must not demand too much. Conscious- ness is always in flux, and 'dauernd' is a relative term. Sensible quality, for instance, cannot maintain itself in consciousness for any length of time; wherever there is sensory adaptation — - 266 THE LAWS OF ATTEXTION : II in pressure, in temperature, in sight, in smell, to some extent in taste — there is also gradual change or disappearance of quality. The condi- tions of clearness, central predisposition and peripheral accommodation, may be given ; but the quality will still fade out. Yet we do not speak of quality as an intrinsically intermittent attribute of sensation ! What is the evidence, then, in the case of attention, of clearness itself ? It is necessary, at this point, to change the venue to the laboratory, because descriptive psy- chology cannot distinguish between discontinuity, due to the come-and-go of ideas, and intrinsic intermittence. The chief and obvious reason that we are unable, under the conditions of every- day life, to hold fast to a single idea is that other, invading and competing ideas oust it from the conscious focus. Only experiment, therefore, can decide recjarding the 'fluctuation' of atten- tion. And experiment, as you know, has been at work with ever increasing frequency since 1875. Investigations have been made by the help of * minimal ' stimuli, — stimuli that are so small or so weak or so little different from their surround- ings that the least slip of attention, the slightest loss of clearness, will mean their complete dis- appearance from consciousness; it is far easier to say that we do or do not hear or see something TEMPORAL INSTABILITY 267 than it is to be sure that what we see or hear has grown more or less clear. Visual, auditory, and cutaneous stimuli have been employed : light and colour, tone and noise, mechanical pressure and the interrupted current. The ques- tions at issue may be formulated, in logical order, as follows : Does fluctuation occur in all sense- departments ? Are the conditions of fluctuation, where it occurs, central or peripheral ? And, if they are central, are they the conditions of 'at- tention ' ? We begin with the question of sense- departments. Lange, who was the first systematically to in- vestigate the subject, found fluctuation in all three, — sight (Masson disc), hearing (watch- tick) and touch (induction current). Let us confine ourselves for the moment to touch. Fluctuations of electro-cutaneous sensations were later observed by Lehmann ; and fluctuations of areal pressure by Wiersma. On the other hand, Ferree and Geissler, working recently in my own laboratory, have been unable to confirm these results. Ferree reports briefly that *'liminal pressure stimuli [very smooth cork wafers sup- porting minimal weights] were applied to several observers, but no fluctuations were expe- rienced"; and that with liminal electro-cutane- ous stimulation at the tip of the tongue "no 268 THE LAWS OF ATTENTION: II fluctuations of intensity could be detected, al- though repeated attempts were made on a num- ber of observers." Geissler, who repeated and extended the experiments of Wiersma and Fer- ree, comes to the same conclusion. *' Under favourable circumstances, attention focussed upon liminal and supraliminal cutaneous sensa- tions remains approximately constant for at least two to three minutes, provided that physiological adaptation of the sense-organ and violently obtruding distractions can be avoided for this length of time." The *two to three minutes' is a conservative estimate; the time was often ex- ceeded, and on one occasion a trained observer reported no fluctuation for ten minutes ! How much longer he might have attended we do not know; at the end of the ten minutes his obser- vation was interrupted by the experimenter, who thought that something was the matter. This absence of fluctuation in the sphere of touch — if we may accept it as fact — strongly suggests that the conditions of fluctuation at large are not central but peripheral. For the skin has no special mechanism of accommodation, and possesses but a poor substitute in the * Tastzuck- ungen' that Czermak noticed in his blind ob- servers. Involuntary tremors in the hand were a minor source of disturbance in Geissler's ex- TEMPORAL INSTABILITY 269 periments. We might argue, therefore, that fluctuation will appear only where there is a peripheral apparatus for accommodation, and that the appeal to central conditions is unneces- sary. Unfortunately, as we shall see, the issue is more complicated.^^ — Aside from touch, we have a reported failure of fluctuation in the case of auditory stimuli. The experimental results are, however, contra- dictory; Heinrich, whose observations upon tones were in part confirmed in the Cornell laboratory, finds no fluctuation, while Dunlap declares that '*the fluctuations were unmistakably observed by each of the five subjects employed." The difference is, apparently, a matter of conditions, which must be further studied. I return to the point presently.^* Our second question was that of the peripheral or central seat of the fluctuations; and, in turning to it, we shall naturally think, first of all, of the mechanisms of accommodation. There can, however, be no doubt that their presence is not essential. Visual fluctuations have been observed by Pace during temporary paralysis of the muscles of accommodation, and by Slaughter and Ferree in the case of aphacic subjects; auditory fluc- tuations occur, according to Urbantschitsch and Eckener, despite the lack of a tympanic mem- 270 THE LAWS OF ATTENTION: II brane. At the same time, the occurrence of fluctuation when the mechanisms are absent does not prove that shift of accommodation has noth- ing to do with fluctuation in the normal organ, where they are present and in working order. And Heinrich maintains, as a matter of fact, that the fluctuation of visual point-areas corresponds with variation in the curvature of the lens, and that the fluctuation of minimal noises is due to **pulsatorische Aenderungen des tensor tympani in seinem Erregungszustande."^^ It is impossible, in the present state of our problem, to take a definite stand for or against Heinrich's explanations. The peripheral theory which he represents obviously requires a central supplement; but that is not a decisive argument. What compels us to a suspension of judgment — perhaps even to a negative attitude, at least so far as vision is concerned — is the appearance of new observations and a new theory. Ferree has published two elaborate investigations, in which he seeks to show *'that the intermit- tences of sensation resulting from minimal visual stimuli . . . are, in reality, simply adaptation- phenomena somewhat obscured by the special conditions.*' " Adaptation is, in itself, a continu- ous phenomenon, but its continuity is interfered with by eye-movement, blinking, etc. Through TEMPORAL INSTABILITY 271 these influences, probably essentially through that of eye-movement alone, it becomes an inter- mittent process, whether the stimulus be liminal or intensive, provided that proper areas be used. The conditions are especially favourable for short periods of intermittence when the stimuli are liminal and of small area." The central idea of this theory, the combination of local adapta- tion and eye-movement, goes back at least as far as 1894 ; but the theory itself, as fitted to the phenomena of visual fluctuation, may justly be described as new. Ferree has worked it out for adaptation and for the converse of adaptation, the negative after-image; he has taken account both of voluntary and of involuntary eye-move- ment; he has made observations in direct and in indirect vision ; he has determined the condi- tions under which fluctuation does and does not occur. In particular, he has been venturesome enough to announce a new discovery in physio- logical optics : "eye-movement . . . determines or influences the washing or streaming over the retina of some material capable of directly affect- ing the visual processes." His chain of evidence is not yet complete, and I may be prejudiced in favour of an investigation which was begun and largely carried out in my laboratory; but it seems to me that Ferree 's principles are likely 272 THE LAWS OF ATTENTION: II to give us a definitive solution of the problem of fluctuation in the sphere of vision. Hammer, like Ferree, appeals to local adapta- tion and slipping of fixation: "doch ist es wohl iiberdies nicht unmoglich," he says, "dass der Adaptationsprozess gleichwie der negativen Nach- bilder seiner Natur nach intermittierend ist." Ferree finds, however, that adaptation is ' a con- tinuous phenomenon,' and that "fluctuation is not grounded in the nature of the after-image process." Hammer then proceeds to report experiments on sound, and concludes — this is the point to which I said just now that I should recur — that there is no such thing as auditory fluctuation. "Auf dem Gebiete des Gehors- sinns existieren tiberhaupt keine Aufmerksam- keitsfluktuationen." If that conclusion could be accepted, our path would be smooth indeed ! — no fluctuations in touch; no fluctuations in hearing, — the whole question of the role of the tensor in fluctuation shelved for ever; fluctuations in sight alone, and due in the case of sight to very special conditions residing in the function of the peripheral organ. But too many observers have recorded auditory fluctuation for us lightly to disregard the positive testimony; all that we may do, again, is to suspend judgment. ^^ — We are not even yet out of the wood. For TEMPORAL IxXSTABILITY 273 peripheral — or at least subcortical — conditions of jfluctuation may be found, not only in the or- gans of sense themselves, but also in those syste- mic changes that are studied by the method of expression. The influence of tlie pulse, for in- stance, is attested by Stumpf , Mach, and Preyer for tones, and by Stumpf for visual stimuli. ^^ The influence of respiration is mentioned by Helmholtz : **ich erinnere daran, dass selbst die Athembewegungen auf das Eigenlicht der Netz- haut einwirken"; it has been traced also by Lehmann and Slaughtcr.^^ And within the last few years a series of investigations, carried out in Pillsbury's laboratory, has emphasised the correspondence of sensible fluctuation with the Traube-Hering wave of blood pressure. ^^ Exner had said, in 1894 : ''es liegt nahe, als Erklarungs- grund aller dieser Erscheinun^en vasomoto- rische Ursachen anzunehmen"; and Pillsbury, in 1903, would explain the 'fluctuations of atten- tion' as "a resultant of two physiological pro- cesses, of the degree of efficiency of the cortical cells, on the one hand, and of the state of excita- tion of the vasomotor centre on the other." ^^ In estimating this position, I feel a strong in- clination to shelter myself behind Pace, and to say simply : "the results thus obtained are obviously of great importance ; and they are certainly open 274 THE LAWS OF ATTENTION: II to various interpretations."^^ For indeed criti- cism at this time and in this place is impossible; it must be criticism of detail, — of the interpreta- tion of records, of the differential value of control experiments, of the probability of rival theories. Slaughter uses the vasomotor phenomena in one way; but Fechner, and Fick and Gurber, and Lehmann, and now Ferree suggest other ways in which they may be turned to theoretical ac- count .^^ I am by no means convinced that Slaughter's hypothesis is the best. — Until we know more about these peripheral conditions it is, I think, useless to appeal to the centre. In particular, it is useless to raise our third question, and to attempt any characterisa- tion of possible central conditions. There are many psychologists who have a predilection for the cortex ; my own leaning is towards the sense- organ. But apart from that — or, if you like, because of that ! — I believe that experimental psychology has always made most progress when it has worked from without inwards: **It is a healthy instinct," I have said elsewhere, **that sends us back and back again to the channels of sense, as we seek an appreciation of the ful- ness and richness of the mental life." I do not deny that the cortex is concerned in the ' fluctua- tions of attention ' ; no one at the present time TEMPORAL INSTABILITY 275 can make such denial. But I look for explana- tion from the behaviour of the sense-organs.^^ At this level of * minimal stimuli,' then, the law of temporal instability will mean that the pe- ripheral conditions of clearness are intermittent. Whether the central predisposition persists or itself oscillates, during peripheral intermittence, is an open question. I think that the predispo- sition is sustained. Geissler is of the same opinion. Even Pace writes in similar vein. The attention, he says, '*must undergo a change of some kind when the stimulus disappears. . . . When the gray ring or band of light vanishes, the attention is divided between the memory-image of that which has just disappeared and the im- pression actually received from the general field. Again, while it may be said that the attitude of the attention in both phases of each fluctuation is one of expectancy, it is also true that the term of this expectation varies : in one phase, the ob- server awaits the disappearance of the stimulus, in the other, he looks for the reappearance of the stimulus." This 'change' of attention, 'divi- sion' of attention, shift of expectant attitude, — all this is very different from a fluctuation of attention, a more or less periodic rise and fall of attentional energy. There is nothing in Pace's 276 THE LAWS OF ATTENTION: II language to bear out the Wundtian statement: "die Aufmerksamkeit ist ihrem Wesen nach eine inter miUirende Funct ion . " ^* Let me add, for the avoidance of misunder- standing, that the law of temporal instability holds, without any question, for central predis- position in the large. The 'instances of continued attention that I gave some pages back were ex- treme instances; and even they, as I said, ** illustrate often enough the labile, instable char- acter of attention."* But these jfluctuations of the total attentive consciousness lie beyond our present horizon. There should, now, be a final law (7) of degree of clearness, — a law that would stand to clear- ness as Weber's Law stands to intensity of sen- sation, and as the various discriminative constan- cies stand to the qualitative, temporal and spatial attributes. "The discovery of a reliable meas- ure of the attention," Klilpe says, "would appear to be one of the most important problems that await solution by the experimental psychology of the future." ^^ The discovery has not yet been made; but we may devote a little space to methods. There seem to be two possible ways, a direct * P. 264. DEGREE OF CLEARNESS 277 and an indirect, of * measuring attention,' form- ing a scale of clearness-degrees, by appeal to the attentive consciousness. The first or direct way is to utilise the observer's introspections of clear- ness itself. Suppose that an observer has at- tained to maximal practice in some field, let us say, of discriminative sensitivity. Maximal prac- tice may, for experimental purposes, be con- sidered a constant. Suppose, again, that we have arranged a series of distracting stimuli, homo- geneous in kind but graded in complexity, such that we are able to reduce the observer's per- centage of right cases from 100 to 95, 90, 85 . . . according to the distraction employed. It is necessary that the action of the distractors be constant ; and it is necessary that they be of the same kind, and therefore exert an influence which differs only, and differs measurably, in degree. Having secured these conditions, we should let the observer decide whether the clearness of conscious contents was distinguishably different under a 5 per cent, and a 10 per cent, distraction, or under a 5 per cent, and a 15 per cent, distrac- tion, or again under an 80 per cent, and an 85 per cent, distraction, and so on, all through the series. We should thus finally obtain a scale of notice- ably different clearnesses paralleled by a scale of measured amounts of distracting stimulus; we 278 THE LAWS OF ATTENTION: II should have the materials for formulating our law; we should have solved the problem of measurement of attention. With this idea in mind, I set a number of my advanced students to work, years ago, upon the preliminary question of the distractor. We dis- covered some interesting things : that distraction may spur instead of distracting; that intermit- tent distractions, the ordinary ' intellectual opera- tions,' are unreliable; that odours are admirably constant distracting material, — if only they could be measured ! and so forth. But we got no farther; and no one else has got any farther by that road. Nevertheless, I am not yet persuaded that the road is altogether impracticable, and we are now making a renew^ed attempt to open it up.^® The other, indirect way of measuring atten- tion is to measure the concomitant sensations of strain, the 'effort' of attention. "Wir besitzen an gewissen begleitenden GefUhlen einer gros- seren oder geringeren Anstrengung . . . ein Mit- tel, uns zu verge wissern, ob unsere Disposition in zwei Fallen annahernd dieselbe sei"; that is a suggestion of Stumpf's. May we not generalise it, and argue that different degrees of effort run parallel to the distinguishable degrees of clear- ness ? Unfortunately, no ! In the first place, the concomitant effort is an indication, not of DEGREE OF CLEARNESS 279 degree of attention, but rather of inertia of atten- tion; strained attention is attention under dif- ficulties; we attend best when effort is small. May we, then, reverse the parallelism, and make degree of effort an inverse measure of degree of clearness ? No, not that either ! For, secondly, experiment has shown that under certain cir- cumstances attention is maximal when we are slightly * distracted ' ; a modicum of effort is favourable to clearness. In a word, the relation of effort to degree of attention is equivocal ; even if we could accurately measure effort, we should have no measurement of clearness.^^ It does not follow, however, because we are as yet unable to measure attention, that we may not devise objective tests which shall inform us, on the one hand, of gross differences in attentional degree or attentional capacity as between ob- server and observer, and on the other hand of approximate constancy or marked fluctuation of attention in the same observer. Four kinds of test offer themselves at once. We may deter- mine the range of attention, simultaneous or successive ; we may have recourse to tests of sen- sitivity and sensible discrimination ; we may determine associability, the rate and stability of association and reproduction ; and we may 280 THE LAWS OF ATTENTION: II measure the promptness of voluntary action, the time of simple reaction. The application of such tests is by no means easy, and I imagine that it can rarely be direct. The test will rather appear as an incidental feature of some more general investigation ; or it will be made an end in itself, and its result then carried over by analogy to investigations whose main purpose is of a different nature. A good deal of work has already been done. Binet subjected two groups of school children, classed by their teachers as intelligent and unin- tellige;it, to a long series of tests: the children were required to discriminate sesthesiometric im- pressions, to count dots by eye and sounds by ear, to memorise letters, to read a word exposed by the movement of an instantaneous shutter, to perform simultaneous additions, to correct proofs, to make reactions, etc. I think that, on the whole, the outcome of the inquiry justified the time and care devoted to it ; but the results, as is only natural, leave us in doubt as to the per- manent value of the individual tests and their precise relation to the attention. Janet, taking the cue from his experience with hysterical sub- jects, proposes a perimetrical test of the degree of attention. Of a somewhat different order are the proposals of Oehrn and Henri, to measure TESTS OF ATTENTION 281 attention by reference to mean variation, and of Wiersma and Pillsbury, to utilise for the same end the duration of noticeability (or the ratio of the periods of noticeability and unnoticeability) in experiments on fluctuation. Oehrn's sugges- tion, in particular, may very well prove to be of value, though it is clear that detailed analysis of conditions, a careful sifting out of contributory factors, must precede its application.^^ The number of possible tests is thus very great. And since all psychological observation is done in the state of attention, and distracting stimuli may always be 'thrown in,' there is no single form of experimental procedure that cannot be made to afford a rough gauge of concentration. It would be strange if, out of this multitude of possibilities, there should be no positive gain for psychology. There will be, — if analysis is pushed far enough, and if recourse is had to the observer's introspection ; otherwise, we shall remain upon the plane of the roughly practical. I gave, at the beginning of this discussion, a schematic outline of a psychological distraction- method. I remark, in conclusion, that it will be wise to combine such a method with the tech- nique of the method of expression. I have no faith in the power of the expression-instruments to tell us of the nature and number of the affec- 282 THE LAWS OF ATTENTION: II tive qualities. But they may help — one never knows ! — towards an objective differentiation of the conscious degrees of clearness.^^ I am at the end of my review. I have done my best to make the review complete, on our elementary level, and to disentangle the really elementary problems from the problems of the total attentive consciousness. In the next Lec- ture we must try to gather up the critical threads and to weave them into a pattern ; we must con- sider the status and the relations of affection and attention within a systematic psychology. VIII AFFECTION AND ATTENTION LECTURE VIII AFFECTION AND ATTENTION THROUGHOUT our discussion of attention, I have been urging that it is advisable, in an elementary psychology, to shift the emphasis from total attentive state to sensation and the attribute of clearness; the discussion has centred about that suggestion. Our treatment of the affective pro- cesses, on the other hand, was almost wholly critical. Construction is always more difficult than criticism, — even when it takes the very modest form of making up one's mind in the face of rival theories. But you have the right to look here also for some positive suggestion ; and, although I have nothing original to say, I shall accordingly begin this final Lecture with a brief outline of an elementary psychology of feeling. Let me assure you again, as I have assured you before, that my position is tentative, provisional, not fixed and dogmatic. We find, in the history of psychology, two op- posed views of feeling, views that I shall dis- tinguish as the intellectual and the affectional. 285 286 AFFECTION AND ATTENTION The intellectual view considers feeling as a form of cognition ; the affectional gives it an inde- pendent place among the mental faculties. On the score of formal expression, the intellectual view is undoubtedly the older; human thought, in the early stages of its activity, is prone to rationalise, and for a long time — for a time, indeed, that extends well into the modern period — it was also dominant. But the affectional view crosses it at many points; and when we come to Kant, the traditional status of the two theories has completely changed, and the affec- tional has gained the day. The faculty of feel- ing is added, as intermediary, to the faculties of knowledge and of desire. Kant's authority was, of course, very great; and the affectional view of feeling held its place in the writings of those later psychologists who escaped the influence of Herbart. Fortunately or unfortunately, however, the main current of modern psychology takes its source from the intellectualism of Herbart and the sensationalism of contemporary physiology. Hence — if I may change the figure — experimental psychology had, from the outset, a strong intellectual bias, a definite leaning towards Gefuhlsempfindungen or a GefiXhlston der Empjindung, We saw, in an earlier Lecture, that Wundt at first resisted HISTORICAL VIEWS OF FEELING 287 the pressure of this tendency, but later for a while succumbed to it.* In a word, then, the intellectual view of feel- ing has been favoured in two ways : by the inertia of a settled philosophical tradition, and by the nature of the sources from which modern psy- chology derives/ But the tradition, after all, merely illustrates an inherent onesidedness of reflective thought ; and we must remember that it was successfully overcome by the psychology of the eighteenth century. As for Herbart and the physiologists, it is — for our immediate pur- pose — nothing more than an historical accident that the succession of the faculty-systems should have devolved upon a rigorous intellectualism. What is significant, again, is this : that modern psychology, just like the psychology of the eigh- teenth century, has finally revolted against intel- lectualism, so that the majority of present-day psychologists recognise the independence of the affective processes, and the doctrine of the * affec- tive tone' has well-nigh disappeared. But, you may ask, has this second movement for affective independence really accomplished anything ? Have we not ourselves admitted and emphasised the unsettled state of the psychology of feeling.^ Are not the opposed camps, of * p. 133. 288 AFFECTION AND ATTENTION majority and of minority alike, split into num- berless factions ? — I will try to answer these questions: premising only that ground once gained, in the history of a science, is never wholly lost, and that our modern movement could not have culminated so quickly, had we not had the example of the eighteenth century before our eyes. And my answer will be twofold. Do not let us forget, in the first place, that the physiological tradition is unbroken. The theo- ries of Bourdon and von Frey can trace their descent, in the spirit if not in the letter, from a long line of workers in * physiological psychology ' ; and the Gefuhlsempfindungeyi, w^hatever may be thought of them by the descriptive psychologist, will not easily yield their claim on the side of explanation. Sensational theories of feeling we shall have always with us. But let us reflect, secondly, that a period which is sterile in obser- vation is, invariably, fruitful in speculation. When, a few years ago, I was classifying the con- tents of the leading psychological journals, I was amazed at the small number of the experimental studies of feeling.^ I had known, as we all know, that a marked interest in feeling is of quite recent growth ; I had not realised how profound was the lack of interest that preceded. No won- der, then, that every psychologist has his own THE STATUS OF AFFECTIOX 289 hypothesis ; no wonder that no two psychologists can agree upon * the definition of feeling ' ! ^ We have a fairly exact parallel in the history of psychophysics. There was a time, in that too, when all the world was writing theory and no- body was doing work, — the time when Merkel prayed aloud that his own experiments might lead, not to "weitere theoretische Discussionen von Seiten der vielen Gegner," ^ but to more accurate tests, made by better men. We have come out of this sterile period in psychophysics, and we shall come out of it also in the psychology of feeling. For a while yet we shall go on wran- gling about opinions; but every experimental study helps to clear the air, and as observations multiply, theory will reshape itself to accord with fact. My personal opinion is, as I have shown plainly enough, that affection must be given elemental rank in consciousness, as a process coordinate with sensation. I rely, primarily, upon the lack of the attribute of clearness; all sensations may become clear, while an affection — however pro- longed or intensive — is never clear, never comes to the focus of attention. I rely also upon the criterion of ' movement between opposites * ; it seems to me that conscious opposition is al- ways a matter of affection, never of sensation. 290 AFFECTION AND ATTENTION And I rely, to some extent, upon the concurrence of these distinguishing characters, and upon their implication — or, at least, suggestion — of cer- tain other differences between the two modes of conscious process. On the w4iole, I am con- vinced that a generic difference can be made out, in the adult human mind, between sensation and affection. I therefore believe that Stumpf's proposal, to treat affective processes as if they were sensations, to bring all the machinery of sensation-method to bear upon the feelings, is a mistaken proposal ; I do not think that it is worth while to assume after-images and memory- images, and contrasts and inductions, in the sphere of feeling; What do you gain by the assumption, if you cannot find the facts ? On the question of the number of the affec- tive qualities, I have no choice but to abide by my experimental results. The situation has its humorous side; for I have tried, I suppose, as hard as any one to discover the pluralists' variety — with * vorgef asste Meinung ' and * leere Schab- lone ' and ' Dogma der Lust- Unlusttheorie ' and ' vollig haltlose Behauptung' all the while buzzing about my ears. I do not know why Wundt should be so severe upon those who differ from him, seeing that his own opinion has more than once changed, and that what he himself terms the A THEORY OF FEELIN"? ^^^ *'erste, vorlaufige Darstellung des dreidimy' ^"^ alen Systems der Gefiihle " ^ was not given to^^^^' world until 1896. I do not think, either, that th2 * Lust-Unlusttheorie ' is a dogma. It has been a dogma; it was allowed to become a dogma by the supineness, not of the dualists (for we were all dualists together), but of the experimentalists in general; and, as I have pointed out, this dog- matic slumber of experimental psychology is re- sponsible for the current hypertrophy of theory. But to charge the dualist with dogmatism, in the year 1908, is simply to charge him with accep- tance, in a modern version, of the traditional doc- trine of pleasure-pain. Is a man dogmatic every time that his experiments lead him into agree- ment with Aristotle ? I will now venture to sketch a theory of feel- ing which seems to me to be sufficiently plausible, and which serves to round out, by explanation, the remarks of the preceding paragraphs.^ It is natural to suppose that the material of con- sciousness, the stuff out of which mind is made, is ultimately homogeneous, all of a piece. Let us make that supposition. The affections then appear — I do not like to say, as * undeveloped sensations,' for an undeveloped sensation is still a sensation ; but at any rate as mental processes of the same general kind as sensations, and as 290 aff;»^ection and attention And I K processes that might, under favourable of tKiitions, have developed into sensations. I iiiazard the guess that the * peripheral organs' of feeling are the free afferent nerve-endings dis- tributed to the various tissues of the body ; * and I take these free endings to represent a lower level of development than the specialised receptive organ. Hence we have peripheral organs of sense, but no 'organs,' in the strict meaning of the term, for affective processes. Had mental development been carried farther, pleasantness and unpleasantness might have become sensatrbns, — in all likelihood would have been differentiated, each of them, into a large number of sensations. Had our physical development been carried farther, we might have had a corresponding increase in the number of internal sense-organs. What does this theory explain ? It explains, first, the obscurity of feeling, the absence of the attribute of clearness. Affective processes are processes whose development has been arrested ; they have not attained, and now they never can attain, to clear consciousness. Affective expe- rience is the obscure, indiscriminable correlate of a medley of widely diffused excitatory pro- * We must, of course, except the free nerve-endings at the periphery of the body, which are probably the 'organs' of pain. I do not think that the exception hurts the theory. A THEORY OF FEELING 293 cesses.'^ The theory explains, secondly, the movement of affective process between opposites, and the relation of this movement to the health and harm, the weal and woe of the organism.^ For the excitatory processes will report the *tone' of the bodily systems from which they proceed, and the report will vary, and can only vary, be- tween * good ' and ' bad.' At this point, of course, the theory takes account of * mixed feelings.' It explains, thirdly, the lack of qualitative dif- ferentiation within pleasantness-unpleasantness. The report of *good' or *bad' may show varia- tion in degree, but cannot change in kind. And, lastly, the theory explains the introspective resem- blance between affections and organic sensations. Genetically, the two sets of processes are near akin ; and it is natural that they should be in- timately blended in experience. I shall not attempt further details. If the theory appeals to you, you will work out details, applications and corollaries, for yourselves. If it does not, you will pursue some other path, — and we shall see presently who has made the wiser choice. A distinguished physicist remarked the other day that theories are matters, not of creed, but of policy; ® and it seems to me that it is better policy to look at the affective processes in the manner here outlined than to think of 294 AFFECTION AND ATTENTION them as apperceptive reactions, or as centrally aroused concomitant sensations, or as indices of the state of nutrition of the cerebral cortex, or as symptoms of the readiness of central discharge. But every one cannot be right ; and where our positive knowledge is practically nil, there is no disgrace in being wrong. I pass on, then, to another question. Let us take it as agreed that affection is an independent mental process, inherently obscure, and evincing a qualitative duality. What, now, is the rela- tion of affection to attention.? When I read Ebbinghaus' chapter on Atten- tion in 1902, I was greatly surprised at the men- tion, among the 'Bedingungen der Aufm^rk- samkeit,' of the affective value of impressions. "Diejenigen Ursachen, die einen stark lustbe- tonten oder unlustbetonten Bewusstseinsinhalt zur Folge haben, setzen diesen Inhalt leichter durch als andere Ursachen ihre indifferenten AVirkungen." I had supposed that * interest' still figured as a condition of attention only in quite popular psychologies; yet Ebbinghaus said in 1902, and repeats in 1905, that "Interesse . . . besonders haufig ein starkeres Hervortre- ten [eines] Eindrucks in der Seele bewirkt." This doctrine implies, first, that feeling precedes ATTENTION AND INTEREST 295 attention, that sensory clearness follows in the train of pleasantness-unpleasantness. Indeed, the point is explicit; for interest hewirht, effects or induces clearness, and interest (Ebbinghaus is properly careful to define it) is itself ** [eine] Lust, die hervorgebracht wird durch das har- monische Zusammengehen eines gegenwartig der Seele nahegelegten Eindrucks mit friiher erworbenen, jetzt durch ilin geweckten Vorstel- lungen." ^^ Pleasure, then, comes first, and atten- tion afterwards. The doctrine implies, secondly, that certain of the conditions of attention are non-affective, that sensory clearness may be established in the absence of pleasantness- unpleasantness. On the former issue, Ebbing- haus comes into direct conflict with Stout. ''The assumption that attention depends on pleasure- pain seems to have no suflScient basis. . . . In- terest and attention do not seem to be related as antecedent and consequent, but rather as differ- ent aspects of the same concrete fact. . . . Feelings of pleasure and pain ... do not de- termine attention as antecedent conditions." ^^ On the latter, he comes into conflict with Wundt's well-known definition of feeling as "the reaction of apperception upon sensations." ^^ Ebbinghaus supports his assertion, that atten- tion is conditioned upon feeling, by reference to 296 AFFECTION AND ATTENTION incidents of everyday life : the attraction of a pretty face or a bad accident, the fascination of anything connected with a man's particular hobby, etc.; he offers no experimental evidence. Now Stout and Pillsbury have analysed a num- ber of precisely such instances, and have been led to precisely the opposite conclusion. Stout I have already quoted. Pillsbury declares, in the same sense : **les choses ne sont interessantes que parce que nous portons sur elles notre atten- tion, et nous ne portons pas sur elles notre atten- tion parce qu'elles sont interessantes." ^^ Unless, then, there are outstanding facts, which refuse to be analysed in this way, we must, I think, decide against Ebbinghaus. Personally, I con- fess that, after the discussion as before, I find it difficult to take his position seriously. The second question that he raises for us is, on the contrary, of very great systematic impor- tance. Do we ever attend without feeling .? Or is it rather true that whenever we attend we feel, and w^henever we feel we attend ? In so far as Ebbinghaus' treatment implies that we may attend without feeling, I am in agreement with it. I am afraid, however, that that bare statement is misleading; and I shall accordingly try to give it a systematic setting. Consider the attentive consciousness in the large : ATTENTION AND WILL 297 what is its place in a psychological system? Wundt, of course, places it under the heading of 'will.' "Die Apperception ist gleichzeitig ele- mentarer Willensact und constituirenderBestand- theil aller Willensvorgange." ^* Stumpf takes a similar view, though he seems to have felt a difficulty (as Wundt does not) in bringing in- voluntary and voluntary attention under a single heading.^^ Ebbinghaus finds that, in voluntary attention, "der Gesammtzustand durchaus gleich dem . . . als Wollen beschrieb- enen ist;" but he distinguishes involuntary from voluntary attention, as he distinguishes Trieb from Wille. Since, however, will is a develop- ment from impulse, — "der Wille ist der voraus- schauend gewordene Trieb," — the distinction is merely terminological/® Other psychologists have held other opinions. I cannot here discuss them ; I can only say that, so far as I see, the term 'will' affords the best general title for tw^o great groups of psychologi- cal facts : the facts of attention and the facts of action. There can, I think, be no doubt that these two groups are intimately related, that ac- tion is simply a special case of attention. But, if that is the case, we may use our knowledge of either one to throw light upon the other. Thus, psychologists and moralists alike have long dis- / 298 AFFECTION AND ATTENTION puted whether 'pleasure and pain' are the sole conditions of action. I do not consider that they can be numbered at all among the condi- tions of action ; I believe that the conditions of action are to be learned from a study of the conditions of attention ; and I should analyse the alleged positive instances on the lines of Stout's and Pillsbury's analyses in the sphere of attention. On the other side, our immediate question — * Do we ever attend without feel- ing ?' — is answered as soon as we appeal to action. Do we ever act without feeling ? Very certainly we do ! Actions may be cla-ssified in various ways, and I shall not try to impose upon you a classification of my own ; but we shall agree that there are many types of action, reflex and automatic and ideomotor and what not, that are performed without the arousal of pleas- antness-unpleasantness in consciousness. In just the same way, as it seems to me, may we have an automatic or instinctive or mechanised attention that is altogether free of feeling. No one will deny that pleasantness and unpleasantness ap- pear often and often again — 'besonders haufig,' as Ebbinghaus says ^^ — as the accompaniments of attention ; it would be strange if our experience were otherwise, since all the conditions of atten- tion are at the same time conditions of a powerful ATTENTION WITHOUT FEELING 299 impression of the nervous system. The connec- tion is, indeed, so obvious and so widespread that it is only natural to regard it as universal; I have myself for many years subscribed to this belief, and have taught that affection and atten- tion are simply back and front, obverse and re- verse, of the same consciousness. But, after all, views must give way before observations; and though I have, for clearness' sake, thrown our discussion into systematic form, my reliance throughout is, as you will have understood, upon observed instances of feelingless attention. Now that I have once noticed such cases, they prove to be of fairly common occurrence; and I am sure that you will have no especial difficulty in discovering their like : — only you must not look for them in the professional spirit, but keep your eyes open to mark them as they come. You may, perhaps, demur to my proposal that all these phenomena of attention and action be brought under the heading of 'will.' Reflex actions and instinctive attentions are, indeed, from one point of view, the antipodes of will. If, however, I adopt the term, it is because I accept the genetic theory of Wundt and Ward. I believe, with Wundt, that '*die zweckmiissigen Reflexbewegungen stabil und mechanisch ge- wordene Willenshandiungen sind";^^ I be- 300 AFFECTION AND ATTENTION lieve, with Ward, that ''volition or something analogous to it" has, in the race as in the indi- vidual, invariably "preceded habit" ;^^ and I believe, with Cope, that even "the automatic 'involuntary' movements of the heart, intes- tines, reproductive systems, etc., were organised in successive states of consciousness." ^^ Argu- ment, on so large a subject, is here out of the ques- tion; but I am glad of the opportunity to recite my credo. And the more strongly you react against it, the more earnestly do I beg you to give it a fair examination.^^ — So far, then, the relation between affection and attention is hardly more than external. Affection reports the tone of the great bodily sys- tems that lack organs of sense ; attention means the clarifying of sensory contents under the in- fluence of powerful nervous stimuli. The or- ganism may, with time, become adapted to these attentional stimuli, — so that, while the corre- sponding sensations appear, at least momentarily, at the conscious focus, there is no felt shock or tilt of the whole living body, no concomitant pleasantness or unpleasantness. We may at- tend without feeling. May we, on the other hand, feel without attending.? Can there be a change in general organic tone, suflSciently marked to reveal itself as feeling, while the NO FEELING WITHOUT ATTENTION 301 sensory contents of consciousness are still ob- scure ? Wundt answers this question with a decided affirmative. ** [Die Gefiihle] konnen auch dann, wenn ihre Vorstellungsgrundhige ausserordentlich dunkel bleibt, eine relativ grosse Intensitat gewinnen." ^^ And again: *'erhebt sich irgendein psychischer Vorgang uber die Schwelle des Bewusstseins, so pflegen die Ge- fuhlselemente desselben, sobald sie die hinreich- ende Starke besitzen, zuerst merkbar zu werden, so dass sie sich bereits energiscli in den Blick- punkt des Bewusstseins drangcn, ehe nocli von den Vorstellungselementen irgend etwas v/ahr- genommen wird." ^^ Here, I think, Wundt is working with 'feelings' where he should be working with organic sensations and pleasant- ness-unpleasantness ; and, in so far forth, his statements are unconvincing. But this objec- tion does not fully meet the issue ; we have to consider the question on its psychological merits. Notice, then, that the question itself takes two forms, a popular and a scientific. I gave them just now as if they were interchangeable ; but a very little reflection brings out their difference. The popular form of the question is : ' May we feel without attending.^' And the implication here is that attention, the attentive consciousness, is something sporadic and occasional; that the 302 AFFECTION AND ATTENTION two-level consciousness alternates with a one- level, wholly 'inattentive' consciousness. Now it is possible that this state of inattention exists, though I confess myself sceptical in the matter; I doubt w^hether inattention, in the waking life, is not always 'attention to something else.' ^^ But, at all events, I do not think it likely that any one will argue for the affective character of inattention ; the very word suggests a state of indifference. In its second and scientific form the question asks whether the obscure contents of a two-level consciousness may be as strongly 'toned' as the clear. Wundt declares that they may; my own analysis leads me to the opposite conclu- sion. I grant that we attend without feeling; and this admission seems to me to bring with it a very welcome 'loosening up' of systematic psychology. But I cannot grant that — in the sense of this paragraph — we feel without at- tending. I incline rather to find a fairly close parallel between degree of clearness and degree of pleasantness-unpleasantness, and thus to regard the relation between affection and atten- tion, on this side, not as external, but as intrinsic. Wundt has missed the organic sensations alto- gether; and we, who emphasise them, must ourselves be careful not to confuse the clearness FEELING AS REACTION 303 of a sensory fusion either with qualitative articu- lation or with definiteness of localisation. Bear- ing this caution in mind, you will surely agree that, whenever we are moved and stirred to feel- ing, the sensible factors in the total process are relatively clear. ^^ It is now necessary to go back a little way, in order to remove a possible misapprehension. I said that the list of conditions of attention, in Ebbinghaus' Grundzilge, implies that we may attend without feeling. And I said that that implication is in conflict with Wundt's familiar definition of feeling as the reaction of apper- ception upon the sensory contents of conscious- ness. How can we attend without feeling, if feeling is generated in the act of attention ? But I have just now quoted from Wundt sentences which affirm that feeling may be present, at a relatively high intensity, while its sensory sub- strate is still 'extraordinarily obscure.' Do not these sentences suggest that feeling may be gen- erated without attention ? And is not Wundt, therefore, inconsistent ? When Wundt wrote, in 1893, that feeling is "die Reactionsweise der Apperception auf die sinnliche Erregung," he meant that statement to be understood in its obvious and literal sense. 304 AFFECTION AXD ATTENTION '* Der Gefuhlston " — we are in 1893 ! — '' kommt liberhaupt nur zu Stande, insofern wir die Empfindungen appcrcipiren, und er kann daher als die subjective oder psychische Seite jenes cen- traleren V oranges der Apperception angesehen werden, der zu der centralen Sinneserregung hinzukommt, wenn sich die Thatigkeit des Bewusstseins ihr zuwendet." ^^ In 1902 he has kept the phrase, — feeling is still the *' Reaction der Apperception auf das einzelne Bewusstseins- erlebniss"; it is of the essence of feeling *'Re- actionsweise der Apperception auf den Bewusst- seinsinhalt zu sein,"^^ — but he has changed its meaning. Feeling is no longer confined to those sensory contents that are the 'object' of atten- tion; on the contrary it may accompany any contents, clear or obscure. '' [Die] centrale Function der Apperception ist in jedem Augen- blick auch fiir den ganzen iibrigen Bewusstseins- inhalt bestimmend, indem dessen sammtliche Elemente nach ihrem Verhaltniss zu den apper- cipirten Elementen geordnet werden. So er- scheinen denn auch die an die einzelnen Bewusst- seinsinhalte gebundenen Geflihle durchaus als subjective Bestimmungen, die jedes einzelne Be- wusstseinserlebniss durch seine Einwirkung auf die Function der Apperception empfangt. In diesem Sinne" — in this new and modified sense THE ATTENTIVE CONSCIOUSNESS 305 — ''ist jedes Gefiihl . . . Reaction, der x\pper- ception auf das einzelne Bewusstseinserlebniss." I do not understand how conscious contents can be reacted upon by apperception without thereby becoming clear. But, however that may be, this reaction, which evokes feeling over the w^iole of the obscure background of consciousness,-^ is something entirely different from the direct reac- tion of attention upon its* object. The tridi- mensional theory of feeling has compelled Wundt to change his exposition ; and he has changed it in such a way that, so far as the phrases go, he is not inconsistent. At the same time, his revised doctrine is, I am sure, only transitional, and I hope that we may presently have an essay in which it is fully worked out. That was a digression. We have next, pick- ing up again the main thread of the discussion, to attempt a rough characterisation of the atten- tive consciousness. Its central feature, the two- level formation, has already been described.* But besides the ''Klarheitszunahme einer be- stimmten Vorstellung oder Vorstellungsgruppe " and the "Hemmung anderer disponibler Ein- driicke oder Erinnerungsbilder," — besides, that is, the appearance of the two levels, — Wundt *P. 241. 306 AFFECTION AND ATTENTION finds an essential constituent of every process of apperception in the concomitant Thdtigkeitsge- fiXhl?^ What are we to say with regard to the feeling of activity ? Wundt has often been charged with circularity of statement. Feeling is the reaction of apper- ception upon sensation ; yet apperception itself comes to consciousness as a feeling. Or again : apperception is the* primitive act of will; acts of will are feelings; yet feeling presupposes the direction of an act of will upon sensation. ^^ I have been accustomed to meet this charge by the reply that the feeling of activity is the dis- covery of introspection. Let me quote a parallel case. James has told us that *' the acts of attend- ing, assenting, negating, making an effort are felt" by him '*as movements of something in the head." '* Whenever my introspective glance succeeds in turning round quickly enough to catch one of these manifestations of spontaneity in the act, all it can ever feel distinctly is some bodily process, for the most part taking place within the head." ^^ Kohn objects to this de- scription that "if the feelings were present w^hile the attention is directed upon some other object, there would be no need at all of the 'turning round' or the 'introspective glance.' We should be conscious of them without this." ^^ To which THE FEELING OF ACTIVITY 307 the obvious rejoinder is that we are conscious of them * without this ' ; otherwise there would be no cue for introspection. We do not attempt to introspect the non-existent. But, when we are giving a psychological account of any contents, we examine it in the state of attention.* Apply that, now, to Wundt. The typical form of attention, if one induces it for purposes of introspection, is voluntary attention. Conscious- ness in the state of voluntary attention is com- posed, in part, of *muskulare Spannungsemp- fmdungen.' When, then, one seeks to introspect the attentive consciousness, one comes naturally upon these sensations of strain ; they are made focal ; and, in the process of their focalisation, a 'feeling of activity' must, on Wundl's view, be struck out. Hence it is impossible to introspect the state of voluntary attention without discover- ing a ThdtigkeitsgefilhL I think that this explanation heads off the charge of circularity ; and it seemed worth while to lay it before you because, as I said, the charge has often been made. It must, however, be pointed out that the independent status of the feelings in Wundt's recent writings changes the whole situation. The Thdtigkeitsgefuhl has now * It must be remembered that James and Kohn use 'feel' and 'feeling' where we should employ 'perceive' and 'perception.' 308 AFFECTION AND ATTENTION to stand on its own feet, without aid from intro- spection ; we either experience it — as feeling — in every instance of attention, or we do not. My own opinion is that we do not.^^ In frequent cases of wliat I have called reflex or instinctive or mechanised attention I find no trace of feeling at all. The depressing Gefilhl dcs Erleidens and the subsequent exciting Gcfuhl der Thdtigkeit are both, in my experience, conspicuous by their absence. Wundt's schema for voluntary atten- tion is expectation, followed by a very brief feel- ing of satisfaction or fulfilment, followed again by the feeling of activity .^^ But everything de- pends, surely, upon what you mean by voluntary attention. Wundt is thinking of reaction ex- periments ; ^^ and there are, I admit, — though with a reservation to be made in a moment, — certain forms of compound reaction in which that sequence of feelings is realised. On the other hand, I have noted many instances of what would pass, in ordinary psychological usage, for voluntary attention, in which one or two or all three of the feelings w^ere lacking. My reservation is not serious; it concerns merely the naming of the processes in question. Expectation and effort are not, in my view, necessarily affective, though both of them may, under certain circumstances, be accompanied by THE MOTOR THEORY OF ATTENTION 309 pleasantness-unpleasantness. It is more im- portant, however, to examine the part that effort actually plays in the attentive consciousness. Such an examination brings us face to face with two large questions : the ' motor ' interpretation of attention, and the distinction of attentive states as active and passive, voluntary and involuntary. On these two questions I can only repeat what I have said elsewhere. I have always regarded, and I probably shall always regard, the motor interpretation of attention as one-sided. We have already seen, on the plane of descriptive psy- chology, that kinsesthetic sensations stand in an equivocal relation to 'degree' of attention.* And it may be seriously doubted whether they are a necessary and integral part of the attentive consciousness. Kinsesthesis is, I suppose, al- ways present in the obscure background of consciousness; but I question if, in states of what I term * secondary passive attention,' these kinsesthetic processes are necessarily intensified, or new kinsesthetic sensations introduced. Wundt remarks that the 'muskularen Spannungsemp- findungen,' which are a frequent partial contents of apperception, may '*fehlen oder von sehr geringer Starke sein." ^^ What we need, in this matter, is less theory and more observations of * P. 279. 310 AFFECTION AND ATTENTION fact. On the side of explanation, I suggest that a like one-sidedness is shown in the constant in- sistence on the reflex arc as the functional unit of the nervous system. It seems to be forgotten that, from another point of view, the oflSce of the cortex may properly be described as the disjunc- tion of the reflex arc, the interposition of resist- ance between sensory stimulus and motor re- sponse. The result of this disjunction is that the attitude of the organism may be typically re- ceptive, typically elaborative, or typically execu- tive. In sensible discrimination, the attitude is mainly receptive ; there are no known muscular adjustments that can keep pace with the just noticeable differences of colour and tone. In concentrated thought the attitude is mainly elab- orative ; and where is the evidence of motor outflow here ? I am not disputing the neurone theory; but I argue that the longer a principal path is made, the more synapses there are in its course, and the more numerous the bypaths be- come, the more difficult w ill it be for an excitatory process to find its way out in the straightfor- ward sensorimotor fashion. McDougall whites, to the same effect, that the "physiological basis of the ' Lebhaftigkeit ' of the presentation" is to be found in ''the complexity of the upper levels" of the nervous system ;^^ and Ebbing- PASSIVE AND ACTIVE ATTENTION 311 haus, too, lays great stress upon the 'Querver- bindungen' in his theory of attention.^^ Whether, then, we consider it psychologically or physiologically, the motor interpretation of attention appears to be one-sided. How does it gain acceptance ? Pillsbury gives a reason ; he suggests that all the motor theories derive, in the last resort, from "la tendance populaire a regarder I'activite accompagnant le processus de I'attention comme sa cause." ^^ If, however, this is correct reasoning, the theories are doubly suspect: they stand committed, from the out- set, to a partial view of the facts. We may cheerfully grant that, for this very reason, they have done psychological service; where the problem is complex, exaggeration in one quarter may be necessary to prevent neglect in another. But this does not mean that exaggeration is it- self laudable ; and I can see nothing but a palpable exaggeration in the definition of atten- tion as a motor reaction. ^^ I must hurry on to our second question, — the question of the distinction between passive and active, involuntary and voluntary attention. "La distinction entre I'attention passive et I'at- tention active est basee," Pillsbury says, "sur I'absence ou la presence de sensations d'effort. . . . Mais comme les sensations d'effort sont 312 AFFECTION AND ATTENTION des accompagnements fortuits, ne correspondant ni aux conditions, ni au degre de Tattention, il semble impossible de retenir une partie de cette classification sans compliquer considerablement la terminologie, et cela sans grand profit." ^^ On the historical issue, Pillsbury is undoubtedly right ; and I have already expressed my agree- ment with the view which he takes of sensations of strain. But I doubt very much whether we can afford to discard altogether the use of the terms active-passive or voluntary-involuntary. Better a poor terminology than the slurring of an observed difference ! And, at any rate, it is interesting to note that neither Wundt nor Eb- binghaus relies for the distinction — which both draw, though in characteristically different ways — upon strain sensations. According to Ebbing- haus, ' sensations of activity ' are marks of atten- tion in general. ^^ Wundt, on the other hand, is so largely occupied with the feelings that sen- sations of strain play a very minor part in his account. How, then, are the two forms of attention dis- tinguished ? For Wundt there is, first of all, the difference of feeling. Further: **die active Ap- perception ist im allgemeinen eine durch die Gesammtlage des Bewusstseins vorbereitete, die passive ist in der Regel eine unvorbereitete.'* PASSIVE AND ACTIVE ATTENTION 313 And this means, again, that passive attention is "im allgemeinen eine Willenshandlung unter der Wirkung eines Motivs, oder . . . eine Trieh- handlungy Active attention, on the contrary, is equivocally conditioned ; it is a WillkiXrhand- lung, subject to the interplay of primary with secondary motives, or a W ahlhandlung , the re- sultant of a conflict of primary motives. The two criteria (prepared, unprepared; univocally conditioned, equivocally conditioned) are coordi- nate, — or rather represent two aspects, the descriptive and the causal, of one and the same general difference.^^ I think that these distinctions hold ; and I think that, if we have recourse to our general law of the rise and fall, the expansion and reduc- tion of conscious formations,^^ and classify atten- tions accordingly as primary passive, active, and secondary passive, we are able to do rough and ready justice to the facts. This classification is, in the first instance, genetic. We may assume that attention, in its beginnings, was a definitely determined reaction — sensory and motor both — ■ upon a single stimulus. As sense-organs multi- plied, two or more disparate stimuli might, each one in its own right, claim the organism's atten- tion; here, in sense-rivalry and the conflict of motor attitudes, we should have the birth of ac- 314 AFFECTION AND ATTENTION tive attention. When, later on, image super- vened upon sensation, conflict and rivalry were largely transferred to the field of ideas, and we find in consequence that separation of the recep- tive, elaborative, and executive attitudes of which I spoke just now. So far, there has been a pro- gressive increase in the complexity of the atten- tive consciousness. At this point reduction sets in; choice and deliberation give way to secon- dary impulses, and active gives way to secondary passive attention. The ground is thus cleared for further growth ; new formations appear in the state of active attention, to be simplified in their turn, — and the cycle recurs, with constant al- ternation of habit and acquisition, so long as the organism retains its flexibility. This account shows, in barest outline, my own systematic use of the distinction ; and you see that the whole schema is implicit in Wundt's doctrine, and fol- lows naturally from it.^^ Genetic psychology lends itself to a summary exposition of this kind; to its wider view the principle stands out, clear of confusing details. But Wundt himself is writing descriptive psy- chology; and descriptive psychology is always in the grip of details. This is a fact that we must bear in mind when we seek to appraise the distinction of voluntary and involuntary atten- PASSIVE AND ACTIVE ATTENTION 315 tion in Ebbinghaus' system. '*Die willkiirliche Aufmerksamkeit," he says, *'ist die voraussehau- end orewordene unwillkiirliche. . . . Sieverhal- ten sicli also zueinander wie Trieb und Wille."^® I remarked, a little while ago, that the difference between Wundt and Ebbinghaus, in regard to the will, is at bottom no more than a difference of terminology. Very much the same thing may be said here, except that Ebbinghaus con- fines himself wholly to description, and rejects the coordinate causal explanation, while at the same time his descriptive distinction is cleaner cut, more dogmatic, than that of Wundt. I should give the preference, on both counts, to Wundt's exposition.^^ It must be remembered that we are dealing with formations of bewilder- ing complexity, with total consciousnesses; and that we have little more to guide us than psycho- logical tradition and the casual observations made in the course of experimental work or in everyday life. Under these conditions, we ought to follow up every clue that offers, and we ought also to leave room in the system for doubt- ful cases, intermediate forms, transitional modes. No doubt, the hostile critic will at once raise the cry of inconsistency. Comfort yourselves with the reflection that the hostile critic is generally superficial ! It is the sympathetic critic who dis- 316 AFFECTION AND ATTENTION covers your real weaknesses, and helps you by showing where they lie; and the sympathetic critic is less likely to charge inconsistency than to probe for its underlying reasons. I am now at an end. I finished writing the last paragraph with a feeling compounded, in Wundtian terms, of pleasantness, relaxation, and tranquillisation. We set out from uncer- tainty and chaos ; and we have at least achieved a fairly definite point of view, and have laid out a programme of experimental work for the future. Unfortunately, affective processes move between opposites : and that first feeling — which in my own poverty-stricken terminology would be merely a feeling of relief — soon gave way to a feeling of unpleasantness, tension, and depres- sion. We know so very little of the subject of these Lectures, and the work that we have found to do will take so long in the doing ! But feel- ings, again, are subject to Abstwnpfung, show the phenomena of adaptation ; and the feeling of de- pression passed as the feeling of relief had passed before it. The professional attitude came to its rights. And that attitude, in the case of the experimental psychologist, is — how shall I describe it ? — an attitude of patient confi- dence. We must be patient, because of all the CONCLUSION 317 objects of human inquiry mind is the most baffling and the most complex; we must expect that the systems of to-day may have only an his- torical interest for the next generation. But we may have absolute confidence in our method, because the method has proved itself in the past ; it has done far more for psychology than is generally acknowledged, far more even than is recognised in the ordinary text-book of psychol- ogy: for the law of attentional inertia holds in science as it holds in ordinary life. There is not the slightest doubt that the patient applica- tion of the experimental method will presently solve the problems of feeling and attention. / NOTES NOTES TO LECTURE I ^ It is, of course, an open question whether the sensation and the image may be bracketed under a single heading (Kulpe) or must be treated as distinct elements (Ebbing- haus). I am not here prejudging this question. Which- ever view one holds, the combined doctrine of sensation and image is set off, in systematic regard, from the doctrines of affection and attention. ^ W. Wundt, Grundziige der physiologischen Psychologies i., 1902, xiv. ; ii., 1902, v. Cf. also i., 353. — H. Ebbing- haus, Grundziige der Psychologies i., 1905, 183 f., 432. The difference might, no doubt, be moderated. Thus Wundt wrote as early as 1896: "Kommt daher auch die Grosseneigenschaft als solche, und zwar im allgemeinen in verschiedenen Formen, namlich als Intensitat, als Qualitat, als extensiver (raumlicher oder zeitlicher) Werth, und eventuell, namlich wenn die verschiedenen Bewusst- seinszustande berlicksichtigt werden, als Klarheitsgrad, jedem psychischen Element und jedem psychischen Ge- bilde an und fur sich schon zu, u.s.w." {Grundriss der Psychologic, 1896, 296 f . ; 1905, 312 [Engl., 1897, 252; 1907, 288 f.]). And Ebbinghaus declares that the general attributes appear " in der Regel je mit mehreren einer be- stimmten Klasse von [den spezifischen Empfindungen] auf einmal" ; difference and multiplicity, for instance, presup- pose at least two sensations (op. cit.y 433 f.). At the same time, the two systems cannot be brought into accord. ^ Vdlkerpsychologie, cine Untcrsuchung der Entwick- Y 321 322 NOTES TO LECTURE I lungsgesetze von Sprache, Mythus und Sitte. I. Die Sprache, i., 1900, 37 ff. Also 1904, 43 ff. It may be urged that, for Wundt's constructive purposes, it matters little whether the dimensions of strain-relaxation and excitement-depression represent "einfache GefUhls- formen" or simple syntheses of affective process with or- ganic sensation. The reply is twofold. If such a change of standpoint is immaterial, then the system cannot be very closely articulated; the superstructure (to change the fig- ure) must sit rather loosely upon its foundations. And again, if the change of standpoint is psychologically neces- sary, then it also becomes necessary to inquire whether there are not other, fundamental and typical syntheses, over and above strain-relaxation and excitement-depression, which have an equal claim to recognition. ^ W. B. Pillsbury, V attention, 1906, v. " Dans Tetat cha- otique ou se trouvent les theories contemporaines de Tattention," etc. Attention^ 1908, ix. ^ E. Mach, Beitrdge zur Analyse der Empfindungen, 1886, 121 f., 134; Die Analyse der Empfindungen und das Verhdltniss des Physischen zum Psychischen, 1900, 180 f., 193. ® See, e.g., the discussion of the Method of Limits in my Experimental Psychology, II., ii., 1905, 99 ff. On the general topic of elements and attributes it may suffice here to refer the reader to A. Meinong, Ueber Be- griff und Eigenschaften der Empfindung, Vjs. f. wiss. Philos., xii., 1888, 324 ff., 477 ff. ; xiii., 1889, 1 ff . ; Be- merkungen iiber den Farbenkorper und das Mischungsge- setz, Zeits. f. Psychol, u. Physiol, d. Sinnesorgane, xxxiii., 1903, 1 ff., esp. § 6 ; E. B. Talbot, The Doctrine of Con- scious Elements, Philosophical Review, iv., 1895, 154; NOTES TO LECTURE I 323 I. M. Bentley, The Simplicity of Colour Tones, American Journal of Psychology, xiv., 1903, 92; M. Meyer, On the Attributes of the Sensations, Psychological Review, xi., 1904, 83. Of the systematic treatises I mention only C. Stumpf, Tonpsychologie, i., 1883, 108. The departure from psychological tradition in H. Munsterberg's Grund- zilgeder Psychologic, l, 1900, is noteworthy; but its discus- sion would take us too far afield. See M. F. Washburn, Some Examples of the Use of Psychological Analysis in System-Making, Philos. Review, xi., 1902, 445 ff. — Fur- ther references are given in later Notes. ^ G. E. MUller, Zur Psychophysik der Gesichtsempfin- dungen, Zeits. f. Psychol, u. Physiol, d. Sinnesorgane, x., 1896, 2 f., 25 ff. ^ G. T. Fechner, Elemente der Psychophysik, i. (1860) 1889, 15. ^ W. James, The Principles of Psijchology, ii., 1890, 136. *^ See, e.g., the discussion in Wundt, Physiologische Psychologic, i., 1902, 14 f. (Principles, i., 1904, 12 ff.), 339 ff., 350 ff. It is needless to multiply references, as the usage of the experimentalists is now strict and con- sistent. James himself often employs the term 'sensa- tion' very loosely (cf. the discussion in Principles, ii., 1), though he offers two definitions. On the one hand, (1) sensation is a limiting form of cognition, the form in which "the object cognised" comes nearest "to being a simple quality like 'hot,' 'cold,' 'red,' 'noise,' 'pain,' apprehended irrelatively to other things" (loc. cit.). Sen- sation is realised only in the earliest days of life; it is impossible, or all but impossible, to adults whose cog- nitive function has passed from acquaintance-with to knowledge-about {ibid., 3, 7 f . ; cf. i., 221 ff., 478 f. ; Text- 324 NOTES TO LECTURE I book, 1892, 12 ff.). On the other hand, (2) sensation is "the object cognised" in this Hmiting form of cognition, "namely, simple qualities or attributes like hard, hot, pain " themselves. In this sense, too, a pure sensation is known to the adult only by way of abstraction (Prin- ciples, ii., 3; cf. i., 195, 224, 478 f. ; Text-book, 40 ff., etc.). It is clear, I think, that on either of these definitions the statement of the text is valid. " With this whole discussion, cf. C. Stumpf, Tonpsij- chologie, i., 1883, 207 ff. ; ii., 1890, 56 ff., 535 ff. Ebbing- haus regards volume merely as a 'characterisation' of pitch, and thus endows tonal sensations with but a single qualitative attribute: Grundzuge, i., 1905, 294 f., 445. ^^ The preceding paragraphs are the outcome of personal observations, taken especially during the year 1906-1907. The views which they embody are stated in more detail in the forthcoming edition of my Outline of Psychology. On the question of itch and its relation to pain I may also refer to L. Torok, Ueber das Wesen der Juckempfindung, Zeits.f. Psychol, xlvi., 1907, 23 ff. ^^ E. Hering, Zur Lchre vom Lichtsinne, 1878, 55 f. ; Grundzilge der Lehre vom Lichtsinn, 1907, 111. F. Hille- brand, Ueber die specifische Helligkeit der Farben, Sitzungsber. d. kais. Akad. d. Wiss. in Wien, mathem.- naturw. Classe, xcviii., 3, 1889, 89. O. Klilpe, Outlines of Psychology, 1895, 30, 114, 119; Ueber die Objectivirung und Subjectivirung von Sinneseindrlicken, Philos. Studien, xix., 1902, 509. E. B. Titchener, An Outline of Psy- chology, 1896, 68, 71, 77. G. E. Muller, Zur Psychophysik der Gesichtsempfindungen, Zeits. f Psychol, u. Physiol. d. Sinnesorgane, x., 1896,30 ff., 411 f . ; xiv., 1897,40 ff., 60 ff. NOTES TO LECTURE I 325 The psychophysical argument which Muller urges against Hillebrand (Zeits., x., 33) is, I suppose, impHed in one form or another by the taper of the colour pyramid. I do not see, however, how it can be translated into psy- chological terms, as an introspective argument for the intensity of visual sensation. Wundt constructs the col- our pyramid from hue, chroma, and intensity {Physiol. Psychol., ii., 1902, 159 ff.) ; but this procedure necessarily leads to confusion. " Duration has been discussed by M. W. Calkins, Attributes of Sensation, Psychological Review, vi., 1899, 506 ff. (cf. xi., 1904, 221 f.), and M. F. Washburn, Notes on Duration as an Attribute of Sensations, ibid., x., 1903, 416 ff. A few additional words may, perhaps, prevent mis- understanding of the text. When I say that extension and duration are the attributes to which we attend when we are asked certain questions, I do not mean that exten- sion, as such, is or has a definite form or a definite magni- tude or a definite local arrangement of parts, or that duration, as such, is or has a definite length or a definite serial arrangement of parts. I mean only that there are questions which direct us to the fundamental spreading- out character of the sensation, and that there are other questions which direct us to its fundamental going-on character; and that we are able to attend, by abstraction, to these attributes and to neglect the rest. I have never believed, in particular, that locality and order, place in space and position in time, are attributes of sensation. A discussion with which I am in essential agreement will be found in H. Ebbinghaus, Grundziige der Psychologies i., 1905, 445 ff., 480 ff. 326 NOTES TO LECTURE I ^^ Kiilpe says, in Outlines, 30, that " extension belongs only to the visual and cutaneous sensations (Hautsinn) " ; and, in Outlines, 335, that it belongs to " the visual and ' tactual,* — the latter term embracing both cutaneous sensations proper and the articular sensations set up in the motile parts of the body (sowohl die Hautempfindungen als auch die Gelenkempfindungen)." Ebbinghaus (Grundzuge, i., 445 f.) predicates extension only of visual and cutaneous sensations. " Cf . my Postulates of a Structural Psychology, Philo- soph. Review, vii., 1898, 461 f . ; L M. Bentley, The Psychological Meaning of Clearness, Mind, N. S., xiii., 1904, 242 ff. ^' C. Stumpf, Tonpsijchologie, i., 1883, 202 f . ; ii., 1890, 524 ff. According to the footnote, ^6^VZ., ii., 525, the attri- bute of tone-tint was recognised independently in 1885 by Stumpf and by G. Engel ; I have not seen Engel's paper. J. Passy distinguishes between the 'pouvoir odorant' and the intensity of odours. The former is inversely pro- portional to the RL. "Tout le monde sent," he says, "que le camphre, le citron, la benzine sont des odeurs fortes, la vanille, I'iris des odeurs faibles," although the 'pouvoir odorant' of vanilla is at least a thousand times as great as that of camphor : Comptes rendus de la Societe de Biologie, [19 Mars] 1892, 240. It is clear that the * pouvoir odorant' belongs to psychophysics, not to psy- chology; but it is clear also that we must distinguish, psychologically, between the intensity and the penetrat- ingness of an olfactory sensation. G. E. Mliller insists on Eindringlichkeit, as distinct from Intensitdt, in his Psychophysik der Gesichtsempfindungen, Zeits. /. Psychol, x., 1896, 26 ff. Cf. also Die Gesichts- I NOTES TO LECTURE I 327 punkte und die Tatsachen der psychophysischen Methodiky 1904, 123; J. Frobes, Zeits. f. Psychol, xxxvi., 1904, 368 ff. ; and see Lecture V., note 33. On the Eindringlichkeit of certain pains, see W. James, Psychol. Review, i., 1894, 523 note; M. von Frey, Die Gefiihle und ihr Verhdltnis zu den Empjindungen, 1894, 15. ^* Ebbinghaus, Grundzilge, i., 1905, 195 ; H. Aubert, Grundzilge der physiologischen Optik, 1876, 532. Ebbing- haus is speaking in general terms ; it is diflScult to see the basis of Aubert's statement. Cf. also Kiilpe, Outlines, 106 f., 122, 127; Helmholtz, Physiol. Optik, 1896, 325. I owe to Professor Bentley the suggestion that tonal volume may vary without variation of pitch; the point is well worth investigation. Preliminary experiments of my own have, so far, yielded a positive result. ^'R. H. Lotze, Metaphysik, 1879, § 258; 1884, 511 ff. ; Outlines of Psych., (1881) 1886, 17; cf. Medicinische Psychologie, 1852, 208, and my Exper. Psychol., II., ii., 1905, xlviii. if. M. W. Calkins recognises * sensational elements' of brightness or visual intensity, of loudness, etc.: An Introduction to Psychology, 1901, 42, 53, 59, 61, 67, 75, 77. Ebbinghaus writes {Grundzilge, i., 444) : " Gesehene und getastete Ausdehnung sind ohne weiteres miteinander vergleichbar, ebenso die Dauer eines Tones mit der eines Schmerzes. Dagegen hell, laut und heiss, . . . oder . . . violett, sauer, hart, haben schlechterdings gar nichts miteinander gemeinsam.'* NOTES TO LECTURE II ^ Brentano is vouched for by Stumpf, Zeits. f. Psychol.^ xliv., 1906, 4. , ^ The doctrine of continuity is pithily expressed by J. Rehmke in the sentence: "es lasst sich doch garnicht leugnen, dass der sogenannte ' Ton der Sinnesempfindung ' im 'physischen Schmerze' oder in der 'Wollust' und das durch einen Todesfall oder eine Siegesnachricht bedingte * Geflihl ' wesentlich gleiche Bewusstseinsbestimmtheiten sind." Lehrhuch der allgemeinen Psychologie, 1894, 317. For my own view, I can refer only to chap. ix. of my Outline of Psychology, which unfortunately is both sche- matic and, to some extent, out of date. Stumpf's view is given in his paper, Ueber den Begriff der Gemiithsbewe- gung, Zeits. /. Psychol., xxi., 1899, 47 ff., which has full references to the literature. I add only G. F. Stout, A Manual of Psychology, 1899, 63. The phrase quoted from Stumpf will be found in Zeits., xliv., 7. — A pathological case, in which the same conditions were apparently re- sponsible for the loss both of feeling and of emotion, is reported by G. R. d'Allonnes, Rev. Philos., Dec, 1905, 592 ff. ; I confess, however, that I attach no great weight to observations of this sort. Cf. P. Sollier, Le mecanisme des emotions, 1905, 126 ff. ^ Tonpsychologie, i., vi. ; ii., vii. ; Zeits., xliv., 1 ff. In the article of 1899 (Zeits., xxi., 63) Stumpf writes: "Macht man bei diesen Organempfindungen noch einen Unterschied z^qschen der Empfindung selbst und ihrem 328 NOTES TO LECTURE II 329 *Gefuhlston,' z. B. der Hungerempfindung und der Unannehmlichkeit dieser Empfindung, so versteht es sich wohl von selbst und ist von James zuletzt auch noch besonders hervorgehoben, dass fiir die Natur des Affects der GefUhlston das Ausschlaggebende ist." The * zuletzt' refers to James' discussion of the Physical Basis of Emo- tion, Psychol. Review^ i., 1894, 516 ff. I am altogether unable to read Stumpf's interpretation into this paper. James says, when discussing the ' tone of feeling,' * pleasant- ness or unpleasantness of the sensible quality,' that "in addition to this pleasantness or painfulness of the con- tent, we may also feel a general seizure of excitement, . . . which is what I have all along meant by an emotion. Now whenever I myself have sought to discover the mind-stuff of which such seizures consist, it has always seemed to me to be additional sensations . . . localized in divers por- tions of my organism" (523). That is, the GefUhlston \s precisely not the important or decisive feature of the emo- tion. Again (524) : " I am even willing to admit that the primary GefUhlston may vary enormously in distinctness in different men. But speaking for myself, I am compelled to say that the only feelings which I cannot more or less well localize in my body are very mild and, so to speak, platonic affairs. I allow them hypothetically to exist, however, . . . where no obvious organic excitement is aroused." This is very different from making them das Ausschlaggebende where organic excitement, the 'emo- tional seizure,' is the very thing to be explained. ^ With the foregoing paragraphs (criterion of subjec- tivity) cf . M. F. Washburn, Some Examples of the Use of Psychological Analysis in System-Making, Philos. Review, xi., 1902, 445 ff. ; E. H. Hollands, Wundt's Doctrine of 330 NOTES TO LECTURE II Psychical Analysis and the Psychical Elements, and Some Recent Criticism: i. The Criteria of the Elements and Attributes, Avier. Joiirn. Psychol., xvi., 1905, 499 ff. ; ii. Feeling and Feeling-x\nalysis, ibid., xvii., 1906, 206 ff. (esp. 221, 226) ; J. Orth, Gefiihl und Bewusstseinslage, eine kritisch-ex'perimentelle Stiidie, 1903, 20 ff. ; Stumpf, Zeits.y xliv., 8 ff., 34; R. Saxinger, Dispositionspsycholo- gisches liber GefUhlskomplexionen, Zeits., xxx., 1902, 399; Ktilpe, Outlines, 227 f . ; Wundt, Physiol. Psychol, iii., 1903, 110 ff., 514 f., 552 ff. ; G. T. Ladd, Psychology, Descriptive and Explanatory: a Treatise of the Phe- nomena, Laws, and Development of Human Mental Life 1894, 181 ; J. Ward, Psychology, Encyc. Britan., xx., 1886, 67; W. B. Pillsbury, Attention, 1908, 191. I say, on p. 38, that nobody confuses organic sensations with properties of external things. This seems to be true of all the more specific organic sensations, — hunger, thirst, nausea, lust, etc. I am not sure, however, that certain organic sensations or organic complexes, of a diffuse and general character, are not projected along with the accompanying affection into the outer world. TOiat do we mean when we speak of ' a pleasant day,' ' very un- pleasant weather,' 'a comfortable chair,' 'an uncomfort- able waiting room '? I do not find the analysis easy ; but I think that these adjectives are applied as objectively, at least in many instances, as the adjectives 'green' or 'hot.' Von Frey points out (Die Gefiihle und ihr Verhdltnis zu den Empfindungen, 1894, 14) that cutting and stabbing weapons, instruments of torture, etc. are directly appre- hended as ' schmerzhaft ' * ; we speak in English of a *Cf. M. Dessoir, Arch. f. [Anat. u.] Physiol., 1892, 230; W. Nagel, Handbuch d. Physiol, d. Menschen, iii., 1905, 731. NOTES TO LECTURE II 331 'painful-looking* instrument. Against the illustration of p. 37 we might cite such expressions as : " How pleasant your wood fire is ! '* The appeal to language is always dangerous, because a given phrase may mean very different things. Unless I am mistaken, however, we do at times objectify our feel- ings (diffuse organic sensations and affection) * just as we objectify the 'secondary qualities.' It is needless to say that this amendment of the text does not at all invalidate, but rather supplements, the argument of the paragraph. Some sensations, I there say, are subjective. Affection, I here add, is sometimes objective. ^ On the second criterion, of non-localisableness, see M. von Frey, Die Gefiihle, 1894, 12; W. Nagel, Handhuch der Physiologie des Menschen, iii., 1905, 617; J. R. Angell and W. Fite, Psychol. Review, viii., 1901, 245, 451, 455, 458; J. R. Angell, ibid., x., 1903, 5, 14; A. H. Pierce, Studies in Auditory and Visual Space Perception, 1901, 191 f. ; Orth, Gefiihl und Bewusstseinslage, 1903, 29 ff., f 117 ff.; Stumpf, Zeits., xliv., 1906, 12 ff. ; R. Lagerborg, Zur Abgrenzung des Geflihlsbegriffs, Arch. f. d. ges. Psijchol, ix., 1907, 460; Kulpe, Outlines, 1895, 264 f., 274; Ebbinghaus, Grundzuge, i., 1905, 564 f . ; Wundt, Physiol. Psychol, ii., 1902, 341; J. Sully, The Human Mind: a Text-hook of Psychology, ii., 1892, 43; W. McDougall, Physiological Psychology, 1905, 80; Ladd, Psychology, Descriptive and Explanatory, 1894, 182, 201, 536 f., 554; Rehmke, Lehrhuch, 1894, 323 ff. ; T. Lipps, * In the same way, we objectify the pleasantness and unpleasant- ness of tastes and smells. Cf. the discussion of Gefiihlsbetonung by E. Freiherr von Gebsattel, Arch. f. d. ges. Psychol., x., 1907, 145 ff. t Cf. E. Meumann, Arch. f. d. ges. Psychol., ix., 1907, 57 f. 332 NOTES TO LECTURE II Komik und Humor: eine psychologisch-dsthetische Unter- suchung, 1898, 114 f . ; E. Kraepelin, Zur Psychologic des Komischen, Philos. Studien, ii., 1885, 329, 351; A. Lehmann, Die Hauptgesetze des menschlichen Gefuhls- lehens, 1892, 177, 201, 214, 216, 258, 267; G. Storring, Arch.f. d. ges. Psychol, vi., 1905, 318 f . ; P. Sollier, Le mecanisme des emotions, 1905, 81 ff. (cf. 75 ff.) ; N. Alech- sieff. Die Grundformen der GefUhle, Psychol. Studioiy iii., 1907, 259 ff. ; C. H. Johnston, The Combination of FeeHngs, Harvard Psychological Studies, ii., 1906, 159 ff. (esp. 175-179) ; cf. Journ. Phil. Psychol. Sci. Meth., iv., 1907, 215; Psychol. Bulletin, ii., 1905, 163, 166; iv., 1907, 363 ff. On methodical difficulties and the attitude of observation, see F. E. O. Schultze, Arch. f. d. ges. Psychol., viii., 1906, 372 ff . ; xi., 1908, 151 ff. M. Geiger's Bemerkungen zur Psychologic der Gefulils- clemcnte und Gefuhlsvcrbindungen (Arch. f. d. ges. Psychol., iv., 1904, esp. 262 ff.), and the paper by Saxinger quoted in the previous Note, published under the auspices of Lipps and of Meinong respectively, rest upon elaborate theoretical foundations, and arc beyond the range of the present discussion. ® With this discussion of the third criterion cf . Wundt, Grundriss der Psychologic, 1896, 40; 1905, 40 (Engl., 1897, 33 ; 1907, 36) ; Vorlesungen Uber die Menschen- und Thierseele, 1897, 240; Physiol. Psychol, i., 1902, 353; and many other passages. Rehmkc, Lehrhuch, 1894, 295 ff. Stumpf, Zeits., xliv., 1906, 7 note, 17, 22. Orth, GefUhl und Bewusstseinslage, 1903, 28 f. Kiilpe, Outlines, 1895, 93, 242. Ebbinghaus, Grundziige, I, 1905, 564. T. Lipps, Grundtatsachen des Seelenlehens, 1883, 273 ff. W. Wirth, Vorstellungs- und Gefuhlscontrast, Zeits. f. NOTES TO LECTURE II 333 Psychol., xviii., 1898, 49 ff. P. Sollier, Le mecanisme des emotions, 1905, 244 ff., esp. 253. Lipps, if I understand him aright, has recently changed his opinion with regard to ' mixed feehngs ' ; see Leitfaden der Psychologic, 1906, 297 f. ' On Kulpe's criterion see KUlpe, Outlines, 1895, 185 f., 225 f., 238; Ladd, Psychology, Descriptive and Explana- tory, 1894, 196, 199; Stumpf, Zeits., xHv., 1906, 23 ff . ; Pillsbury, Attention, 1908, 190 f. On the difference between *zufalHge innere Wahrneh- mung' and 'planmassige Selbstbeobachtung,' see Wundt, Essaijs, 1885, 127 ff . ; 1906, 187 S.; Philos. Studien, iv., 1888, 292 ff. (esp. 301) ; etc., etc. On the phrase * centrally excited sensations,' see E. Meumann, Vorlesungen zur Einfiihrung in die experi- mentelle Pddagogik und ihre psychologischen Griindlagen, i., 1907, 205. It might be objected to Kiilpe that the ex- periments of H. Munsterberg (Beitr. z. experiment. Psy- chol, iv., 1892, 17 ff.), A. Goldscheider and R. F. Muller {Zeits. f. Uin. Medizin, xxiii., 1893, 156 ff.), and W. B. Pillsbury {Amer. Journ. Psychol., viii., 1897, 355 ff.) indicate, under certain conditions, an intensive equivalence of peripherally excited and centrally excited sensations. KUlpe has, however, forestalled the objection in Outlines, 183. ^ On habituation, see James, Principles, ii., 475 f. ; Stumpf, Zeits., xliv., 1906, 7 note; Bericht uher d. II. Kongress f. expcr. Psychologic, 1907, 213; Kiilpe, Out- lines, 261; Ebbinghaus, Grundziige, i., 1905, 574 ff . ; A. Lehmann, Die Hauptgesetze des menschlichen Gefiihls- lehens, 1892, 182 ff. ; Wundt, Physiol. Psychol, ii., 1902, 332 (with continuous stimulation, initial pleasantness may 334 NOTES TO LECTURE II pass directly, through indifference, into unpleasantness) ; Sollier, Le mecanisme des emotions, 1905, 97 ff. "Use blunts feeling and favours intellection," says Ward: Encyc. Britan., xx., 1886, 40. ^ On the relation of affection to attention, see Kiilpe, Outlines, 1895, 258 ff., 430; Titchener, Philos. Review, iii., 1894, 429 ff. (the systematic setting of this paper is crude, but I think that the observations are reliable) ; Psychol. Review, ix., 1902, 481 ff. ; P. Zoneff and E. Meu- mann, Philos. Studieri, xviii., 1903, 4 f., 67 ff. (cf. Lec- ture III., note 43) ; W. B. Pillsbury, Psychol. Review, ix., 1902, 405; Meumann, Experimentelle Pddagogik, i., 1907, 82; W. Hamilton, Lectures on Metaphysics, 1859, i., 236; ii., 432; J. Ward, Psychology, Encyc. Britan., xx., 1886, 40 ff. ; Hollands, Amer. Journ. Psychol., xvii., 1906, 211; Wundt, Phijsiol. Psychol, ii., 1902, 357; iii., 1903, 114, 348; Saxinger, Zeits., xxx., 1902, 400, 412; A. Lehmann, Die kdrperlichen Aeusserungen psychischer Zustdnde, i., 1899, 140 ff. ; W. H. Burnham, A7ner. Journ. Psychol., xix., 1908, 16; F. E. O. Schultze, Arch. /. d. ges. Psychol., viii., 1906, 373. Sully says definitely (Human Mind, i., 1892, 77) that "we can intensify a pain or pleasure by attending to it as such," — as definitely as he says (ibid., 143) that "objects of attention are either sensations, and their combinations, sensation-complexes, or what we call ideas or representa- tions." But he counts 'bodily pain,' 'the pain of indiges- tion' as an 'affective state,' and admits that "in attending to the feeling we necessarily embrace [the presentative element] to some extent." It is true, as he remarks, that "to listen to a musical sound so as to note its pitch, etc., and to listen to it solely for the sake of enjoying it, illustrate NOTES TO LECTURE II 335 two different directions of the attention"; but there is here no evidence of direction of the attention upon the enjoyment, and the latter's consequent intensification. Cf. also i., 67; ii., 12. In his Attention, 1908, 187, Pillsbury writes: "as mat- ters stand, the introspective evidence is universally favour- able to the assertion that attention is antagonistic to the pleasantness-unpleasantness process as well as to the vague unanalysed processes of consciousness." The first part of this sentence, at any rate, confirms my own position. On another point, however, Pillsbury seems to disagree. " Of Wundt's three pairs," he says, " strain and relaxation w^ould not be opposed to attending. ... Of depression and exaltation it is difficult to speak, but it is by no means certain that attention to these processes would either oppose or favour their presence" (187 f.). I think that the disagreement is only apparent. Pillsbury is con- sidering the Wundtian processes, so to speak, on their merits, as they occur in his experience ; I am setting forth Wundt's own doctrine. I believe with Hollands that feeling, in Wundt's system, cannot be made the object of attention; and I find this teaching in his tridimensional theory as in the theory of affective tone. Nevertheless, I point out in Lecture VIII. that Wundt's present view of the relation of affection to attention is, in my judgment, transitional, and I therefore regard it as possible that his systematic position may be changed. ^^ The idea of this paragraph is that the criterion of move- ment between opposites may be coupled with that of coextension with consciousness, — opposition meaning, in fact, conscious incompatibility; and that the criterion of lack of clearness may be coupled with that of subjec- 336 NOTES TO LECTURE II tivity, — lack of clearness implying a textural difference between sensation and affection, which finds expression in the term 'subjective.' We thus reach a twofold char- acterisation of affection, to be explained and justified by psychophysical theory. I think that this bracketing "considerably strengthens the case for an elementary affection." We are led by it, e.g., to mistrust the instances of localised affection, such as occur in Storring's experi- ments. Storring, it will be remembered, secured Stim- mungslust and Empfindiingslust by the following experi- mental procedure : " wahrend man bei der Erzeugung von Lust, die an eine Geschmacksempfindung gekniipft ist, die Geschmackslosung wahrend der Dauer des Versuchs im Munde behalten lasst, gab ich zum Zweck der Erzeugung von Stimmungslust der Vp. die Anweisung, die Losung zu schlucken und dann von der Empfindung abzusehen, mit dem Schlucken den Geschmacksreiz als eine erledigte Tatsache zu betrachten" {Arch. f. d. ges. Psychol. , vi., 1905, 317). It is clear that the instructions are not paral- lel: so long as the fluid is in the mouth, the observer's attention is upon it, and the affection is localised along with its sensation (cf . Note 4, above) ; when the fluid has been swallowed, and the taste is past and done with, the affection is not localised. So much depends upon the conditions ! But let the instructions be made parallel : let the observer be told, in the experiments on Empjiridungs- lust, to consider the taste as past and done with so soon as it has come clearly to consciousness; let the retention in the mouth be merely a matter of convenience, of not interrupting the experiment. In this case, if I may trust my own introspection and that of three other obser\^ers, there is no localisation of the pleasantness; it is of pre- NOTES TO LECTURE II 337 cisely the same character as the pleasantness after swallow- ing. However, the subject needs renewed investigation of a systematic kind. We have an analogy to the argument of the text in the position of those psychologists who make the image a dis- tinct mental element, coordinate with sensation. 'You cannot distinguish sensation and image on the ground of quality alone, or of intensity, or of duration, or of exten- sion, or of clearness. Can you not distinguish them in terms of the consensus of these attributes ? Is there not a total textural difference between the two processes ? ' This, it seems to me, is the gist of the separatist argument, when it is couched in terms of content. No doubt, there is, in the case of the image, a further appeal to characteristic differences of context or background or setting. NOTES TO LECTURE III ^ B. Bourdon, La sensation de plaisir, Rev. philos.y Sept., 1893, 226 f. 'M. von Frey, Die Gefilhle, 1894, 14 f., 17. — In this connection, mention should also be made, perhaps, of Sollier's recent theory of 'cenesthesie cerebrale': Le mecaiiisme des emotionSy 1905, esp. 192 ff., 257 f. ' Op. city 227 f. ^ See, e.g.y Rep., ix., 583 D; Phaedo, 60 A; Phil.y 51. ^ Bericht iiber den II. Kongress filr experimentelle Psy- chologiey 1907, 209 if.; Zeits., xliv., 1906, 1 ff. On p. 15, Stumpf makes the terminological suggestion that Gefuhls- empfindung be rendered by 'emotional sensation.' This translation seems to me hardly possible; the English equivalent would be, I think, either 'affective sensation' or *algedonic sensation.' The adjective 'algedonic' was coined by H. R. Marshall {Pain, Pleasure and J^sthetics, 1894, 9), not — as Stumpf and Lagerborg (Arch. f. d. ges. Psychol. y ix., 1907, 454) say — by Baldwin. In a systematic connection I should prefer the phrase ' algedonic sensation'; for the purposes of this Lecture its introduc- tion appeared unnecessary. «Pp. 1 ff. ' Pp. 2 ff . ' Outlines, 1895, 227 f. ' i., 282. ^" T. Ziehen, Leitfaden der physiologischen Psychologic in 15 Vorlesungen, 1906, 162 note. 338 NOTES TO LECTURE III 339 " P. 5 note. '' P. 4. 13 H. R. Marshall, President's Address, American Psy- chological Association, Chicago Meeting, December, 1907: The Methods of the Naturalist and Psychologist, Psychol. Rev., XV., 1908, 16 f. "If we could isolate psychic ele- ments, ... we would [sic] discover in connection with them elemental qualities ... of the nature of pain and pleasure." " T. Ziegler, Das Gefiihl : eine psychologische Unter- suchung, 1893, 100. Cf. Stumpf, p. 43 note. ^^ P. 6. — Under the theory which makes the affective processes "eine neue Gattung psychischer Elemente, Zustande oder Funktionen, die weder Empfindungen noch Eigenschaften von Empfindungen sind" (p. 3), falls the view which considers them as *Gestaltqualitaten.' In Bericht, 213, Stumpf definitely rejects this view. I must say, however, that he seems to me to come very near it in his doctrine of the 'ReinheitsgefUhl,' the feeling for the purity of consonant intervals: see F. Krueger, Bericht, 212; Psychol. Studien, ii., 1906, 371 f., 375 ff. (where full references are given). At any rate, we are here far re- moved from the positive rejection of mental chemistry, from the "Aus Nichts wird Nichts," of the Towpsy- chologie (ii., 1890, 209, 525, etc.). '' Pp. 6 ff. ^' Pp. 15 ff. ^« P. 18. " P. 21. '« P. 19. '^ Pp. 16 f. ^^A. Goldscheider, Gesam. AbhandL, i., 1898, 411 f . ; ^Stumpf, pp. 19 note, 21. 340 NOTES TO LECTURE III 2^ P. 19. ^' Pp. 16 f. ^ Pp. 19 f. 2« Pp. 2, 22. 2^ P. 21. "^ See, ^.^r., Kulpe, Outlines, 124. 2^H. Ebbinghaus, Grundziige, i., 1905, 581; cf. R. Lagerborg, Z)a5 Gefilhlsproblem, 1905, 95; P. Sollier, Ze mccanisme des emotions, 1905, 244, 254 f. '« M. Kelchner, ^rc/i./ d. ges. Psychol, v., 1905, 86. 3^ Pp. 21 f. ^2 P. 22. ^^ Pp. 22, 38 f . In the latter passage Stumpf, while still leaving the question open, indicates his own belief in a plurality of affective qualities. ^^ Let it not be objected that these Lttstempjindungen are sensations of pleasure for the reason that they "ein in- stinktives Annehmen und Begehren mit sich zu fUhren pflegen" (p. 22) ! For on p. 16 it is written: "die An- nehmlichkeit ist nicht das Annehmen und der Schmerz nicht das Ablehnen." ^' Pp. 26 ff. '« Pp. 27 f . " P. 28. '' P. 29. ^^ P. 31 ; Tonpsychologie, ii., 1890, 209. '' Pp. 29 f . " E.g., pp. 18 note, 21, 22, 23, 26, 27, 29 f., 36, 39 f. ''P. 2; cf. p. 41. '^E. W. Scripture, Vorstellung und Gefuhl, Philos. Studien, vi., 1891, 536 ff . ; O. Vogt, Die directe psycholo- gische Experimentalmethode in hypnotischen Bewusst- NOTES TO LECTURE III 341 seinszustanden, Zeits. f. Hypnotismus, v., 1897, 180 ff. ; F. Kiesow, Sul metodo di studiare i sentimenti semplici, Rendiconti delta r. Accademia dei Lincei, Classe di scienze Jis.y mat. e natur., (5) viii., 1, 1899, 469 ff. Kiesow*s experiments were made as follows. He first determined two areas of the tongue that were equally sensitive to sweet, and platted the curve of sensible dis- crimination. He then gave his observers preliminary practice "a distrarre la loro attenzione dalla sensazione e a concentrarla esclusivamente sul tono sentimentale (Gefuhlston) che accompagna ogni grado di sensazione.'* The instruction is, evidently, ambiguous (cf. reference to Zoneff and Meumann, Lecture II., note 9). Kiesow naturally found, in consequence, that "in sulle prime queste esperienze sono diflBcili e affaticanti : in alcune persone mi pare di non essere potuto giungere a una sufficiente concentrazione delF attenzione sul tono senti- mentale; esse erano sempre distratte passivamente dalla sensazione. ... In altri soggetti coll' esercizio si puo giungere al punto da poter astrarre dalla sensazione in modo sufficiente." After practice, he began systematic work upon pleasantness-unpleasantness, and platted an affective curve, starting with the RL and employing the DL as unit of abscissas. The curve shows, first, a stage of indifference ; next, a stage of slowly increasing pleasant- ness; thirdly, a second stage of indifference; and lastly a stage of somewhat rapidly increasing unpleasantness. Kiesow remarks that " nella curva cosi ottenuta le ordinate non sono stabilite numericamente con una precisione eguale a quella delle ascisse." He gives no further details, and does not figure the curves. ^* Affective Memory, Philos. Rev., iv., 1895, Q5 ff. See 342 NOTES TO LECTURE III T. Ribot, The Psychology of the Emotions, 1897, 140 ff. A reference to the recent annual bibliographies will show that the subject is still in debate. I find it discussed, e.g., by W. Heinrich, La psych ologie des sentiments {Bull, de VAcad. des Sciences de Cracovicy Jan., 1908, 36), which reaches me as these pages are passing through the press. ^^ Organic Images, Journ. Phil. Psychol. Sci. Meth., i., 1904, 36 ff. '« P. 23. ^^ P. 25. — The illustration is the more striking since G. H. Meyer, a highly practised observer, declares his inability to reproduce cutaneous sensations of intrinsi- cally brief duration. " Auf der Haut gelingt es mir leicht, an welcher Stelle ich will, subjective Empfindungen her- vorzubringen. Weil aber langere Unterhaltung der Ans- chauung dazu nothwendig ist, kann ich nur solche Emp- findungen wecken, welche langere Zeit andauern, wie Warme, Kuhle, Druck; schnell vorubergehende dagegen, wie von einem Stich, Schnitt, Schlag, etc., vermag ich nicht hervorzurufen, weil es mir nicht gelingt^ die entsprech- enden Anschauungen so ex abrupto in der gehorigen In- tensitdt zu wecken'' {Untersuchungen uher die Physiologic der Nervenfasery 1843, 238: italics mine).* Personally, * Meyer's work is not in the possession of any one of the four university Hbraries to which I am accustomed to appeal. It may, however, be procured from the Librarian of the Surgeon General's Office, Washington, D.C. I may mention here — since I find that the fact is less generally known than it deserves to be — that the Surgeon General's Library is admirably supplied with the older and scarcer books that bear upon experimental psychology. A postcard from the librarian of any college or university library will bring the required volumes, usually by return, and they may be held for a fortnight. NOTES TO LECTURE III 343 I can image pressure and, I think, warmth; I cannot image pain, and I am very dubious as regards cold. See my Organic Images, 38 f., and cf. F. E. O. Schultze, Arch. f. d. ges. Psychol., xi., 1908, 157 f., 185 f. ^« P. 26. ^^ See, however, pp. 35 note, 47. I am not sure that I understand Stumpf s doctrine as regards rhythmical, formal, and harmonic feelings. '' Pp. 31 f. '^ P. 32. ^^ G. T. Fechner, Elemeiite der Psychophysik, i., 1889, 75 (see also refs. given in my Exper. Psychol., II., ii., 1905, Ixviii.) ; Ebbinghaus, Grundziige, i., 1905, 91. ^^ W. Nagel, Handhiich d. Physiol, d. Menschen, iii., 1905, 620. ^' P. 36. =' Pp. 32 f . ^« Pp. 33 fif. " P. 36. "' Pp. 33, 37. '' Pp. 37 f. ^^ By the admission of instances from the field of emo- tion and of aesthetic and intellectual sentiment. See, how- ever. Note 49 above; and cf. Stumpf, pp. 33 ff. ^' Pp. 38 f . I may add that, so far as my experience goes, American libra- ries contain practically everything that is needed for historical research in experimental psychology. In the preparation of my Exper. Psychol. I had to read a great many out-of-the-way things ; but there were very few instances in which I was obliged to have final recourse to European collections. As the great majority of the large libraries — there are a few bad exceptions ! — are courtesy itself in the matter of lending, there is no excuse for 'ignorance of the literature' on the part of the American student. 344 NOTES TO LECTURE III «' Pp. 39 f. «' GrundzUge, i., 1905, 5m. '' Pp. 41 ff. «' Pp. 42, 48 f. «« Pp. 42 f., 44. «^ Pp. 43 f . ®^ See, ^.^., R. H. Lotze, Medicinische Psychologie oder Physiologie der Seele, 1852, 254 f. ; Stumpf, p. 32. *'P. 44; Ebbinghaiis, GrundzUge, i., 1905, 582 f. '"^ Outlines, 1895, 228 ff. '' Op. ciL, 577. '' Pp. 44 f . '' P. 45. '' Pp. 45 ff . '' P. 46. ^« P. 47. ^^ Tonpsychologie, ii., 1890, vii. In Note 15 above I have expressed a certain misgiving as regards Stumpf s doctrine of the Tongefiihle. '« P. 48. NOTES TO LECTURE IV ' Physiol. Psychol, i., 1893, 555 f., 561, 570 f. ""Grundriss der Psychologie, 1896, 33 f., 36, 39 ff., 97 f., 100 (Engl., 1897, 28 f., 30, 33 ff., 82 f., 84 f.) ; cf. 1905, 34 f., 36, 39 ff., 99 f., 101 (Engl., 1907, 31 f., 32 f., 35 ff., 91 f., 93). ^ This is, I am convinced, the true version of a state of affairs which James has unwittingly misrepresented in Psychol. Review, i., 1894, 72 f. ^Physiol. Psychol., 1874, 445. The passage which re- lates to the aesthetic value of the higher senses is retained, with some modification of context, in the edition of 1893 (i., 571) ; but the important thing is the insertion, in that edition, of the new paragraph, i., 561. ^Ibid., 1874, 426; i., 1893, 555. ® Gefilhl U7id Bewusstseinslage, 1903, 49; Wundt, Grund- riss, 1896, 97 ff. (Engl., 82 ff.) ; cf. 1905, 99 ff. (Engl., 1907, 91 ff.). 'Physiol. Psychol., 1874, 441, 721. Ubid., u., 1902, 284 ff. ""Grundriss, 1896, 100, 213 (Engl., 85, 181); cf. 1905, 101, 217 (Engl., 1907, 93, 202). ^"^ Zeits., xliv., 1906, 7 f. note. " Grundriss, 1896, 103 (Engl., 88). Cf. 1905, 103 ff. (Engl., 1907, 95 ff.). ^^ Vorlesungen iiher die Menschen- und Thierseele, 1897, 224 ff. *^ Philos. Studien, xv., 1900, 151. 345 346 NOTES TO LECTURE IV " This Lecture was not printed ; but the material upon which it was based may be found in H. C. Stevens' A Plethysmographic Study of Attention, Amer. Journ. Psychol, xvi., 1905, 48!2. " Grundriss, 1896, 35 (Engl., 29) ; cf. 1905, 35 (Engl., 1907, 32). '' Cf. my note in Psychol Bulletin, iv., 1907, 367 f. ^' Grundriss, 1896, 99 f. (Engl., 84 f.). This § 9 is omitted in 1905. ^^ Philos. Studien, xv., 1900, 177. ^^ Vorlesungen, 1897, 238. "'^ Ibid., 239. ^^Ihid., 239 f. ^''Physiol Psychol, ii., 1902, 311 ff., 318 f., 326, 333, 336 f. The intensive reference of Lust-Unlust indicates a return to the teaching of 1874 ; see Note 4 above. '^ Ihid., 374. ^^ Cf. Wundt's own statement in Grundriss, 1896, 96 (Engl., 81); cf. 1905, 98 (Engl., 1907, 90). ^^ Philos. Shidien, xv., 1900, 175. '^ Grundriss, 1896, 98 (cf. 1905, 99) ; Physiol Psychol, ii., 1902, 286, 295. ^'Physiol Psychol, 1874, 724; Vorlesungen, 1897, 238. In Physiol Psychol, iii., 1903, 253 f., 306 f. (cf. Grundriss, 1896, 256; 1905, 265), the Erfiillungsgefiihl appears as a total feeling, based essentially upon relaxation and tran- quillisation : cf., however, 347. Befriedigung (iii., 221) is a Lustaffect: the Totalgefiihl will then be based upon pleasantness and relaxation. ^^Vorlesungen, 1897, 228; Grundriss, 1905, 105. ^®A. Gurewitsch, Zur Geschichte des Achtungsbegriffes und zur Theorie der sittlichen Gefilhle, Wurzburg dissert., 1897. NOTES TO LECTURE IV 347 ^^ O. Vogt, Normalpsychologische Einleitiing in die Psychopathologie der Hysteric, Zeits. f. Hypnotismus, viii., 1899, 212. ^^Phijsiol Ps7jchoL, iii., 1903, 249. ^2 J. Royce, Outlines of Psychologij, 1903, 176 ff. ^^Philos. Studien, xv., 1900, 172 f. ^' J. Cohn, Philos. Studien, x., 1894, 562 ff . ; xv., 1899, 279 ff. D. R. Major, Amer. Journ. Psychol., vii., 1895, 57 ff. ^'"Physiol Psychol, ii., 1902, 285. ^^ Locc. citt. ^' Phijsiol. Psychol, ii., 1902, 333. 3« Ihid., 332. ""^Ihid., 335. '' Ibid., 336 f. '^ Grundriss, 1896, 96 (Engl., 81) ; cf. 1905, 98 (Engl., 1907, 90). ^''Ihid., 99 (Engl., 84); cf. 1905, 101 (Engl., 1907, 93). '' Vorlesungen, 1897, 235 ff. ^^ Physiol. Psychol, ii., 1902, 290. ^^Ihid., 290 f. ^® G. T. Ladd, Psychology, Descr. and Explan., 1894, 167 ff. *' Ibid., 537. *^ Psychol Review, I, 1894, 525. ^* T. Lipps, Vo7n Fuhlen, Wollen und Denken, 1902. ^"^ Zeits., xliv., 1906, 38 f. ^^ H. Hoffding, Psijchologie in Umrissen, 1887, 279; 1893, 305; Eng. tr., 1891, 222. Klilpc, Outlines, 1895, 232 f. F. Jodl, Lehrbuch d. Psijchol, 1896, 378 f . ; ii., 1903, 1 ff. Ebbinghaus, Gnindziige, l, 1905, 564 ff. A. Lehmann, Hauptgesetze d. menschl. Gefiihlslebens, 1892, 32. J. Rchmkc, Zur Lehre vom Gemiit, 1898, 47 ff. 348 NOTES TO LECTURE IV ^2 Grundziige, i., 1905, 566. ^ Gefiihl und Bewusstseinslage, 1903, 39. ^* Psychol. y Descr. and Explan.y 183. "Nor is he who has felt that joy of scientific discovery which Niebuhr compared to the divine feehng in view of a new-made universe, Hkely to confuse it, as respects distinctive quahty, with the sensuous thrill of gratified bodily appetite," etc. ^^ Das Selbstbewusstsein, Empfindung mid Gefiihl, 1901, 14. '« Grundrissy 1896, 188 f. (Engl., 160) ; cf. 190.5, 192 f. (Engl., 1907, 178 f.); Physiol Psychol, ii., 1902, 344 f. ^^ Gnmdriss, 187 f. (Engl., 159 f.) ; cf. 1905, 191 f. (Engl., 1907, 177 f.) ; Vorlesungen, 234 ff . ; Physiol Psychol, ii., 341 ff. The doctrine varies a little, from time to time, but is in principle as I state it in the text. It goes back as far as the essay on Gefiihl mid Vorstellung {Essays, 1885, 213; Vjs. f wiss. Philos., iii., 1879, 143), but appears clearly for the first time in the Vorlesungen of 1892. ^* Wundt speaks, both in Grundriss and in Physiol Psychol, of the 'musical' tone, the 'musical' triad: my demonstration w^as made with tuning forks. I do not think that objection can be taken to the change, since musical reference, aesthetic association, must in any event be ruled out. Personally, I get the same result with har- monica! or piano chords, except that the musical reference is, with them, much more difficult to exclude. The ob- servation, to be strictly valid, should be varied as Wundt suggests. ^^ Physiol Psychol, ii., 1902, 42 ff. It is very interest- ing, in this connection, — and, indeed, in connection with the general subject of the present Lecture, — to read NOTES TO LECTURE IV 349 Wundt*s account of GemeingefUhly in Beitrdge zur Theorie der Sinneswahrnehmung, 1862, 376-400. *^ Arch.f. d. ges. Psychol., ix., 1907, Literaturbericht, 94. Meumann himself has recently published an extended article on the subject: Arch., ix., 1907, 26 ff. «^ Amer. Journ. Psychol, xvi., 1905, 212. Cf. T. Lipps, Das Selhsthewusstsein : Empfindung und Gefilhl, 1901, 24; M. Kelchner, Arch.f. d. g. Psychol., v., 1905, 124. *2 Grundzuge, l, 1905, 567. ^Zeits., xliv., 1906, 2 note. ^ Cf. Wundt's doctrine in Physiol Psychol, l, 1893, 561. ®^ O. Vogt, Zur Kenntnis des Wesens und der psy- chologischen Bedeutung des Hypnotismus, Zeits. f. Hypno- tisnius, iv., 1896, 127. Cf. Grundriss, 1905, 102 (Engl, 1907, 94). ^® Gefuhl und Bewusstseinslage, 1903, 129. ®^ G. Storring, Arch.f. d. ges. Psychol, vi., 1905, 320 f. I am not even sure that the first of the phrases quoted re- fers to quality at all ; the complete sentence runs : " Stim- mungslust ist gleichartiger, die Lust erfullt mehr das Bewusstsein.'* It may be that these two clauses express the same fact in different terms. Storring himself sums up in the words : " uber Qualitat der beiden Lustzustande machen alle drei Vp. die Angabe, es liege deutliche quali- tative Differenz vor." Why does he not quote their words .^ He goes on: "ich lege aber auf diese Ueber- einstimmung kein Gewicht, weil diese Aussage von . . . der Annahme der Realitat qualitativer Differenzen . . . (die ich iibrigens selbst akzeptiere), abhiingig sein kann." ®® Ein Versuch, die Methode der paarweisen Verglei- chung auf die verschiedenen Gefuhlsrichtungen anzu- wenden, Philos. Studien, xx., 1902, 382 ff. ; S. P. Hayes, 350 NOTES TO LECTURE IV A Study of the Affective Qualities, i. The Tridimen- sional Theory of Feeling, Amer. Journ. Psychol., xvii., 1906, 358 ff. I must refer the reader for details to these two articles, in both of which the ' curves ' of the affective judgments are figured. Criticisms are met in the latter article (361 note), and also in my note on N. Alechsieff's Die Grundformen der Gefuhle {Psychol. Studien, iii., 1907, 156 ff.), Amer. Journ. Psychol., xix., 1908, 138 ff. ^^ Physiol. Psychol., ii., 1902, 287. "So viel man auch mit der Eindrucksmethode hin und her experimentiren oder die unten zu erorternden Ergebnisse der Ausdrucks- methode zu Hulfe nehmen mag, immer kommt man bei der Analyse der concreten Gefuhlszustande oder der zusammengesetzteren Gemiithsbewegungen wieder auf diese [drei Gegensatzpaaren] zuriick." I read this posi- tive statement with surprise when it appeared ; but, what- ever grounds Wundt may have had for it in 1902, there is small evidence of it in 1908. Possibly for this reason Wundt, in his latest exposition (ibid., i., 1908, 23 ff.), speaks very disparagingly of the Reizmethode as an affective method. ^^ M. Brahn, Experimentelle Beitrage zur Geflihlslehre, Philos. Studien, xviii., 1903, 127 ff. ; W. Gent, Volumpuls- curven bei Gefuhlen und Affecten, ibid., 715 ff. Wundt, Physiol. Psychol, ii., 1902, 274, 291 ff. Orth, Gefiihl und Bewusstseinslage, 1903, 58 ff. ^^ A first critique of Wundt's theory, under the title Zur Kritik der Wundfschcn Geflihlslehre, will be found in Zeits., xix., 1899, 321 ff. A detailed Kritik der modernen Gefuhlslehre (Lipps and Wundt) is given by Orth, op. cit., 20 ff. Lipps' doctrine of feeling may be studied in Grundtat- NOTES TO LECTURE IV 351 sachen des Seelenlebens, 1883, 15 ff., 177 ff. ; Bemerkungen zur Theorie der Gefiihle, Vjs.f. wiss. Philos., xiii., 1889, 160 fF. ; Gottingische gelchrte Anzeigen, 1894, 85 ff. ; Komik und Humor, 1898; Das Selbstbewusstsein : Emp- Jindung und Gefiihly 1901 ; Vom Filhlen, Wollen und Denken, 1902; Aesthetik: Psychologie des Schonen und der Kunst. I. Grundlegung der Aesthetik, 1903, Abschn. i., vi. ; Psychologische Studien, 1905 ; Leitfaden der Psy- chologie, 1906, 281 ff. ; and numerous articles in psy- chological journals. NOTES TO LECTURE V * Philos. Studien, x., 1894, 124. Wundt gives illustra- tions, 123 f. ^ As seen, e.g., in Kiilpe, Outlines, 1895, 169 fF. ; Wundt, Physiol. Psychol., iii., 1903, 518 ff. ; Ebbinghaus, Grund- zuge, i., 1905, 633 fF. ^ Meumann says (Exper. PddagogiJc, i., 1907, 326 note) : "wer den ganzen Fortschritt der experimentellen Psy- chologie gegeniiber der friiheren Psychologie der inneren Wahrnehmung deutlich vor Augen haben will, der ver- gleiche die in den folgenden Ausfuhrungen dargestellten Methoden und Ergebnisse der experimentellen Forschung individueller Unterschiede mit dem, was ein so geistvoller Vertreter der alteren Psychologie wie Sigv\'^art liber unser Problem zu sagen wusste. Vgl. Sigwart, Die Unterschiede der Individualitaten. Kleine Schriften, Bd. ii. [1889] S. 212 ff." ' Cf . my Experimental Psychology, I., ii., 1901, 186; W. James, Pri7ic. of Psychol., i., 1890, 402. ^ W. Hamilton, Lect. on Metaphysics, i., 1859, 237 ff. ^ J. Mill, Analysis of the Phenomena of the Human Mind, ii., 1869, 369 ff., with the notes by J. S. Mill and A. Bain. ^ A. Bain, The Emotions and the Will, 1880, 370 ff., 540; The Senses and the Intellect, 1868, 558. * D. Braunschweiger, Die Lehre von der Aufmerksam- keit in der Psychologie des 18. Jahrhmiderts, 1899, 2. ® Lectures on Human and Animal Psychology, 1896, 270. 352 NOTES TO LECTURE V 353 ^^ Unfortunately, the knowledge often acts as a deterrent, — James, here as elsewhere, serving as excuse {Psychol. Rev., i., 1894, 516 note). I had hoped a good deal from the publication of Wundt's Grundriss: but that, even in English translation, is too difficult for the average under- graduate. We sorely need a clear discussion, historical and critical, at the text-book level. — I do not think that my statements with regard to atten- tion are too strong, even in the light of what I said in Lecture I. of the threefold root of the psychological system. It seems to me that the doctrine of attention is of funda- mental importance. And I believe that the strength of Wundt's system lies — and will lie, historically — in the fact of its being an attentional system, whether its special teaching is right or wrong. A system which makes little of attention is, in my judgment, foredoomed to failure. I do not think, either, that I have entered too strong a claim for the modernity of the psychology of attention. Braunschweiger makes a great deal of the eighteenth cen- tury doctrine {Die Lehre von der Aufmerhsamkeit in der Psijchologie des 18. Jahrhunderts, 1899, 2 f., 38, 69, 95 f., 124, 150 ff.). But he is a special student within a special period, and the judgment of the special student is likely to lack perspective. A more impartial witness is M. Des- soir {Geschichte der neueren deutsclien Psycliologiey i., 1897- 1902), and a cursory glance through Dessoir's index will show the approximate place that attention held in eigh- teenth century systems. No doubt, the older psychologists were acute observers. Let me give an instance. I was looking up my Light of Nature, to verify the reference given in Note 20 of the following Lecture, — and I naturally read on, for a few 354 NOTES TO LECTURE V pages, as one is apt to do. I came upon the following: " It has been generally remarked by schoolboys, that after having laboured the whole evening before a repetition day to get their lesson by heart, but to very little purpose, when they rise in the morning, they shall have it current at their tongue's end without any further trouble " (i., 1805, 248 f.). Here is a direct anticipation of the modern psychophysics of association ! And a few pages further : " in a language we are masters of what we read seems wholly to occupy the imagination, yet, for all that, the mind can find room for something of her own : how quick soever the eye may pass along, the thought flies still quicker, and will make little excursions between one word and the next, or pur- sue reflections of its own, at the same time it attends to the reading" {ibid., 253), — the very illustration that I had myself chosen for my discussion of introspection ! — It will hardly be argued, now, that Tucker can compare, as a student of association and memory, with Ebbinghaus and Miiller: and what holds here holds, so far as my reading has extended, of attention as well. Everything depends upon the context, upon the way the problem is seen, upon the suggestion of method, upon the fruitfulness of the idea for scientific purposes. ^^ Grundziigey i., 1905, 611. So also Klilpe, Outlines, 1895, 423. ^^ I. Kant, Metaphysische Anfangsgrilnde der Naturwis- senschaft, 1786, x. f . ; Sdmmtliche Werke, ed. Rosenkranz and Schubert, v., 1838-1842, 310. Cf. my Exper. Psychol., IL, ii., 1905, cxlv. " L. M. Solomons and G. Stein, Normal Motor Autom- atism, Psychol. Review, iii., 1896, esp. 503 ff. ^* The preceding paragraphs are taken, with some com- NOTES TO LECTURE V 355 pression, from the forthcoming edition of my Outline of Psychology. The position agrees, in the main, with that of W. B. Pillsbury, A Suggestion toward a Reinterpreta- tion of Introspection, Journ. Philos. Psychol. Sci. Meth.y i., 1904, 225 ff. See also KUlpe, Outlines, 1895, 8 ff.; Ebbinghaus, Grmidzilge, i., 1905, QQ ff. ; Wundt, Physiol. Psychol, i., 1902, 4 ff. (Engl. 1904, 4 ff.), or i., 1908, 4 if., 23 ff., with references there given. It need hardly be said that the essential similarity of the methods of psychology and the natural sciences does not necessa^-ily carry with it a corresponding similarity of subject-matter and problem. ^^ Kiilpe, The Problem of Attention, The Monist, xiii., 1902, 42. ^® Cf. the remarks in Amer. Journ. Psychol., xvi., 1905, 218 f. ^' Physiol. Psychol, iii., 1903, 341 ; 1874, 717 f. ^^ Vattention, 1906, 2. ; Attention, 1908, 2. ^^ Diet of Philos. and Psychol, i., 1901, 86. 2^ T. Ribot, Psychologic de Vattention, 1889, 6, 9, 36, 95. 2* G. T. Ladd, Psychol, Descr. and Explan., 1894, 61, 66. ^'Stumpf, Tonpsychol, ii., 1890, 276 S.; cf. i., 1883, 67 ff. 2' G. F. Stout, A Manual of Psychology, 1899, 65 f. Cf. Analytic Psychology, l, 1896, 125 ff., 180 ff. ^* J. M. Baldwin, Handbook of Psychol: Senses and Intellect, 1890, 69 ff. Cf. Feeling and Will, 1891, 280 ff., 351 ff . ; Mental Devel in the Child and the Race: Methods and Processes, 1906, 428 ff. ^^ F. H. Bradley, Is there any Special Activity of Atten- tion? Mind, O. S., xi., 1886, 306. 2®D. Ferrier, The Functions of the Brain, 1886, 463. 356 NOTES TO LECTURE V " Grundzuge, I, 1905, 600, 612. ^* J. R. Angell, Psychology: an Introductory Study of the Structure and Function of Human Consciousness, 1904, 65. ^^ C. H. Judd, Psychology: General Introduction^ 1907, 191, 193. ^""Exper, Pddagogik, i., 1907, 78 f. 31 Cf. C. S. Squire, A Genetic Study of Rhythm, Amer. Journ. Psychol., xii., 1901, 541 f. 3^ On intensity of stimulus, see Klilpe, Outlines, 1895, 438; A. Pilzecker, Die Lehre von der sinnlichen Auf- merksamkeit, 1889, 19; Ebbinghaus, GrundzUge, I, 1905, 602; James, Princ. of Psychol., l, 1890, 416 f . ; Pillsbury, Fatteniion, 1906, 38 ff. ; Attention, 1908, 28 ff. ; Wundt, Physiol. Psijchol, iii., 1903, 336; G. E. Muller, Zur Theorie der sinnlichen Aufmerksamkeit, [1873] 110 ff. Muller refers to duration only as a condition of Abstump- fang of the attention : 126 ff. 33 See James, Princ. of Psychol., I, 1890, 417; ii., 383 ff. G. E. Muller, Zur Psychophysik der Gesichts- empfindungen, Zeits., x., 1896, 27 f. Muller describes Eindringlichkeit as follows : " die Eindringlichkeit betrifft die mehr psychologische Seite der Empfindungen, sie scheint sich hauptsachlich nach der Macht zu bestimmen, mit welcher die Sinneseindrlicke unsere Aufmerksamkeit auf sich Ziehen, und konnte daher in sachlicher Hinsicht nicht unpassend auch als die Aufdringlichkeit der Sinnes- eindrlicke bezeichnet werden. . . . [Sie] ist, wie es scheint, nicht bloss von der Intensitat des psychophysis- chen Prozesses abhangig, sondern bestimmt sich zugleich auch nach der Haufigkeit der betreffenden Empfindung in unserer Erfahrung, nach dem Gefuhlswerte derselben und nach anderen derartigen fur die Erweckung unserer NOTES TO LECTURE V 357 Aufmerksamkeit wichtigen Faktoren." Ibid., 26 f. Fre- quency is mentioned incidentally by Mliller, in the Sinn- liche Aufmerksamkeit y 135, as a condition of involuntary attention. — Ebbinghaus, Grundzilge, i., 1905, 602 f. ^^Pillsbury, Vattention, 1906, 39 f . ; Attention, 1908, 29 f.; Ebbinghaus, Grundzilge, i., 1905, 603 ff. The criticism of the text applies also to Ebbinghaus' treatment of expert disregard of irrelevant details (604 f.) and of habituation (712 ff.), in so far as these are made to depend upon mere repetition of stimulus. — Wundt, Physiol. Psychol, iii., 1903, 340. In Lecture VIII. I express my personal opinion that habit always implies foregone attention. ^^ Mliller, Sinnliche Aufmerksamkeit, [1873] 125 f . ; Pilzecker, Sinnliche Atfmerksamkeit, 1889, 20; Pillsbury, U attention, 1906, 40; Attention, 1908, 30; James, Princ. of Psychol., \., 1890, 416 f . ; L. W. Stern, Psychologie der Verdnderungsauffassung, 1898, 211 ff. ^^ Mliller, op. cit., 135; Pilzecker, op. cit., 20; Klilpe, Outlines, 1895, 300 f . ; Stumpf, TonpsijchoL, ii., 1890, 337 ff . (with refs. to S. Exner) ; T. Heller, Philos. Studien, xi., 1895, 249; Stern, Verdnderungsauffassung, 1898, 143, 181 ff., 201; James, Princ. of Psychol, i., 1890, 417; ii., 173 f. ; Pillsbury, op. cit, 62 ff. (Engl., 48 f.). ^^ Miiller, op. cit., 135; Kiilpe, oj). cit., 438; James, op. cit., i., 417; Ebbinghaus, op. cit., 715; Pillsbury, op. cit, 42, 64 f. (Engl., 31 f., 49) ; Wundt, Physiol Psy- chol, 336. Cf. the passage in T. Lipps, Suggestion und Hypnose, Sitzungsher. der philos. -philol u. der histor. Classe der k. bayer. Akad. d. Wiss., ii., 1897, 424: "der Reiz des Neuen und Ungewohnten ist nichts anderes als der Reiz d. h. die Fahigkeit der Inanspruchnahme 358 NOTES TO LECTURE V und der Festhaltung psychischer Kraft, die einem Vor- stellungsinhalte oder Komplex von solchen zukommt, ehe diese Fiihigkeit durch die auf Erfahrungsassocia- tionen beruhende Tendenz der Ausgleichung und des Abflusses sicli vermindert hat. Der Reiz des Neuen ist nichts als der unverminderte Reiz des Objektes." At bottom, this doctrine agrees with that of Muller; the stimulus of novelty is the stimulus of the object as such, the claim that it has to attention in virtue of its inten- sity, quality, duration, etc. While, however, I accept Lipps' analysis, I still think that novelty has a special place in our empirical classification. '' Muller, op. cit, 40 ff., 123 ff. ; Pilzecker, op. cit., 19, 34 ff. ; Pillsbury, op. cit, 44 ff. (Engl., 32 ff.) ; Kulpe, Out- lineSy 439 f. ; Monist, xiii., 1902, 46 ff. ; Ebbinghaus, op. cit, 603, 605 f.; Stumpf, TonpsychoL, ii., 1890, 339; H. Helmholtz, Physiol. Optik, 1896, 890 f. ; Popular Lectures cm Scientific Subjects, [First Series] 1885, 294 f. ; On the Sensations of Tone as a Physiological Basis for the Theory of Music, 1895, 50 f . ; J. Jastrow, Fact and Fable in Psychol., 1900, 89; James, op. cit., i., 437 f. Relevant observations are reported by B. B. Breese : On Lnhibition, 1899, 18 ff.; W. McDougall, Mind, N. S., xii., 1903, 473 ff.; A. Bruckner, Zeits.f Psijchol, xxvi., 1901, 45, 53. ^^Wundt, Physiol. Psychol, iii., 1903, 336; Kulpe, Outlines, 1895, 437 f . ; Monist, xiii., 1902, 44. ^•'Muller, op. cit., 132 ff. ; Ebbinghaus, op. cit., 622, 714; Fechner, Elemente d. Psychophysik, ii., 1889, 446; J. Delboeuf, Examen critique de la loi psychophysique, sa base et sa signification, 1883, 166; Lehmann, Haupt- gesetze, 1892, 194 ff . ; Pillsbury, op. cit., 38 f. (Engl., 29); MUnsterberg, Grundziige, I, 1900, 228 f. NOTES TO LECTURE V 359 " This paragraph follows Muller, Sinnl. Aufmerksam- keit, 110 ff. ''Lotze, Med. Psychol, 1852, 507 ff . ; Wundt, Physiol Psychol, iii., 1903, 336 ff . ; Kulpe, Outlines, 436 ff . ; Monist, xiii., 1902, 46 ff. ; Pillsbury, op. ciL, 35 ff. (Engl., 26 ff.) ; James, Princ. of Psychol, I, 1890, 434 ff. ; J. von Kv'iQs, Zeits. f. Psychol, vili., 1895, 12 ff. Whence Lotze derived his list it would be difficult to say. The topic was a favourite one with the eighteenth century psychologists: see Braunschweiger, Die Lehre von der Aufmerksamkeit, etc., 1899, 50 ff. ; Dessoir, Geschichte, I, 1902, 418. In i., 1894, 238, Dessoir ex- claims, apropos of E. Platner (1744-1818): "ware es nicht vielleicht besser gewesen, Platner hatte uns gesagt, wodurch die Aufmerksamkeit nicht gereizt wird ? " — In Die Lehre von der Aufmerksamkeit, 1907, E. Diirr takes a view of attention (11 f.) which is practically the same as my own ; he also reaches a like conclusion upon various special problems, though in certain cases (e.g., on the subject of fluctuation, 131 ff.) his position is different. Diirr, however, is writing of attention "mit besonderer Beriicksichtigung padagogischer Interessen" (4), so that the course and contents of his exposition are widely diver- gent from those of the present Lectures. NOTES TO LECTURE VI ^Kulpe, Outlines, 1895, 336 f., 379. 2 Wundt, Phijsiol. Psychol, iii., 1903, 339. ^ Pillsbury, V attention, 1906, 3 ff. ; Attention, 1908, 2 if. 'Kulpe, oj). cit., 429, 441 f. MVundt, op. cit., 339 f. ® Ebbinghaus, Grundziige, I, 1905, 612 ff. 'Stumpf, Tonpsychologie, l, 1883, 72, 374; ii., 1890, 293. * H. Mlinsterberg, Grundziige der Psychologie, i., 1900, 227. — The mention of the lengthened Hne may, at first sight, appear gratuitous. The reference is, however, to experiments upon "the distances between visible points, the distances serving as measures for the intensity of the sensations produced by the movement of the eyes." See Psych. Rev., \., 1894, 39; and cf. Amer. Journ. Psych., viii., 1896, 50, 53. ^ See Stumpf, op. cit., ii., 293 f. Stumpf does not specify the instrument : he speaks only of a ' Zungenpfeifenaccord.' It is, perhaps, worth while to add that, even when direct observations agree, their interpretation may be extremely difficult. Thus, Krueger writes of Stumpf's Reinheits- gefuhl: "sicherlich ist die Theorie ausgegangen von wichtigen und genau festgestellten Tatsachen. Hoch- musikalische Beobachter haben mit uberraschender Fein- heit und Konstanz kleine Verstimmungen der ihnen gelaufigsten konsonanten Tonschritte als scharf, spannend, uberreizt, beziehungsweise (die subjektiv verkleinerten 360 NOTES TO LECTURE VI 361 Intervalle) als matt, schal, stumpf bezeichnet. Dass es sich dabei, wie weit und in welchem Sinne, um Gefiihle handle, ist naturlich nicht mehr zweifellos." — Psychol. Studien, i., 1906, 381. ^° See, e.g., H. MUnsterberg and N. Kozaki, The In- tensifying Effect of Attention, Psychol. Review, i., 1894, 39 ff. ; A. J. HamHn, Attention and Distraction, Amer. Journ. Psychol., viii., 1896, 3 ff. ; O. Kulpe, Ueber den Einfluss der Aufmerksamkeit auf die Empfindungsin- tensitat, ///. Internat. Congress/. Psychol., 1897, 180 ff . ; M. Tsukahara, Problem of the Relation of Intensity of Sensation to Attention, 1907. — Cf., further, G, E. Miiller, Zur Theorie d. sinnlichen Aufmerksamkeit, [1873] 2 ff. ; Fechner, Elemente d. Psychophysik, ii., 1889, 452 f . ; Re- vision, 1882, 271 ; T. Lipps, Suggestion u. Hypnose, 1898, 398 ff. ; A. Lehmann, Die Hypnose, 1890, 22 ; J. Geyser, Ueber den Einfluss der Aufmerksamkeit auf die Intensitat der Empfindung, 1897. " Reported briefly in Psychol. Bulletin, iv., 1907, 212 f. Professor Bentley allows me to quote here the full text of his paper. "I have to report, at this time, only a single group of experiments, which deal with the intensity of noise; and I shall reserve for some future occasion a full discussion and interpretation of the results. " Both my apparatus and my method are familiar. The Leipsic type of gravity phonometer was used, and the sound stimuli were given in pairs. To the one stimulus the observer was attentive; from the other he was dis- tracted. The distraction was made effective both by the brevity of the sound and by the character of the distracting stimuli (odours). The success or failure of distraction 362 NOTES TO LECTURE VI was always checked (as were also the state and degree of attention and the conscious filling of the silent interval) by introspective control. Care was taken to eliminate constant and variable errors; and especially to keep the physical and organic conditions as unvarying as pos- sible. " The pairs of stimuli were given usually in series of ten, each pair occupying 13 sec, with an interv^al for rest between successive pairs. An equal number of Distrac- tion-Attention (D-A) and Attention-Distraction (A-D) pairs were introduced in haphazard order in every series. The difference in height of fall Avithin each pair was (with one exception noted below) 5 cm., and the absolute heights varied between 24.4 and 89.6 cm. "Table I. gives the results for Set I. (100 pairs for each one of three observers) and Set II. (120 pairs for each observer). Set I. covers all intensities between 24.4 and 79.4 cm., while Set II. contains only two pairs of stimulus intensity, 24.4-29.4 and 74.4-79.4 cm. As a check upon these observations, a group of 'attention' experiments was added to Set II., where both sensations were received in maximal attention (A- A order). "As regards arrangement in the Tables, note the fol- lowing points. The first horizontal line of figures con- tains the number of correct or ' true ' judgments ; the second line, the times that the ' attention ' stimulus (Rq) was over- estimated, i.e. judged too great ; the third line, the under- estimations of the 'attention' stimulus; and the fourth line, the cases thrown out on the basis of introspection (failure to attend or to distract). In the lower half of the Table, at the right, are given the A- A series, which are self-explanatory. NOTES TO LECTURE VI 363 No. OF Set "TABLE I (10 A-D AND 10 D-A Series) II. (8 A-D, 8 D-A AND 4 A-A Series) Height of Fall 24.i-79.4 cm. 24.4-29.4 cm. 74.4-79.4 cm. >-3 _ ^^ Observee B G M o B G M B G M 1 True 35 U 40 109 17 22 23 62 13 17 13 43 105 -d Ra over- est'd 52 42 42 136 11 11 12 34 19 17 13 49 83 s§ Ra under- 91^ est'd 7 15 18 40 7 4 3 14 8 6 9 23 37 Thrown < out 6 9 15 5 3 2 10 5 5 15 Total 100 100 100 300 40 40 40 120 40 40 40 120 240 Right X 2 12 28 24 64 14 12 12 38 102 a Wrong •^ X 2 22 12 16 50 24 28 26 78 128 0? Doubtful < X 2 Total 6 6 2 2 4 10 X 2 X . 40 40 40 120 40 40 40 120 240 "It is to be observed that, in nearly half (136) of the 300 experiments of Set I., the sound attended to is, for a wide range of intensities, overestimated. The number is somewhat greater than the number of 'true' cases (109) and is about three and one half times as great as the num- ber of underestimated cases (40). Thus far, the results indicate, then, that a noise attended to is sensibly louder than the same objective sound received in distraction. "But the problem demands more specific treatment. It demands, in the first place, the distribution of over- estimated cases throughout the scale of intensity. Set I. furnishes too few judgments at any single intensity to meet this demand. But Set II. contains results from a 364 NOTES TO LECTURE VI single weak (24.4-29.4 cm.) and a single intensive pair (74.4-79.4 cm.). A comparison of the two pairs shows that the number of overestimated cases is uniformly in- creased, for all observers, with increase in physical in- tensity. That this relation does not obtain with weak and strong stimuli whose differences are relatively (not absolutely) the same, is shown by Table II. "TABLE II No. OF Set III and IV.* (14 A-D, 14 D-A, and 8 A-A Series) Height of Fall 24.4-29.4 cm. 74.4-79.4 cm. Observer B G M Total B G M Total Total 1§ ll < True Ra overestimated . Ra underestimated . Thrown out .... Total 30 31 37 37 25 18 7 13 3 7 2 70 70 70 98 80 2(J 12 210 29 38 41 33 26 19 4 4 7 4 2 3 70 70 70 108 78 15 9 210 206 158 35 21 420 1 < Eight X 2 . . . Wrong X 2 . . . Doubtful X 2 . . . Total X 2 . . . 32 35 41 30 32 27 8 3 2 70 70 70 108 89 13 210 27 52 36 33 18 34 10 70 70 70 115 85 10 210 223 174 23 420 "Table II. presents results from two pairs of stimulus intensities which may be supposed to measure (under Weber's Law) like sense-distances. Under the given con- ditions, the overestimations for weak and loud sounds are found to be almost identical (80 and 78) ; or, viewed from the negative side, distraction may be said to weaken, by a like amount, loud and weak auditory sensations. "A comparison of the upper and lower halves of the Tables reveals the curious fact that the number of 'true' * In Set IV. (4 distraction and 2 attention series), the position of O's head was controlled by means of a biting board. The re- sults were consistent with those of Set III. NOTES TO LECTURE VI 365 cases with distraction is almost as great as the number of 'right' cases with continuous attention (A-A series), — namely, 206 and 223. It would seem, at first sight, as if the large constant error introduced by distraction must have materially damaged the function of judgment. But a moment's reflection will make it plain that this error would tend as often toward the increase as toward the decrease of the difference between sensations. Its effect appears, therefore, rather in the distribution (to under- estimations and overestimations) than in the number of incorrect judgments. The small difference in number obtained (17) is probably due to the more unfavourable conditions for judgment afforded by distraction from the one of the sounds compared. — "It is plainly impracticable, at the present state of the problem, to attempt an explanation or even a full inter- pretation of the bare results. Granted that distraction lowers the intensity of certain sensations, we have still to ask what factor in the distracted consciousness is responsi- ble for the effect. Is loss of intensity due to loss of clear- ness ? or to the affective colouring of the distracting odours ? or to the impairment of memory through dis- traction ? or, finally, are the conditions purely physio- logical, i.e. without conscious representation ? We shall hope, by further work, to find satisfactory answers to these questions. At present we can make only preliminary observations. (1) Whatever relation obtains between clearness and intensity, the two things were distinct in the minds of the observers. Not only were the latter familiar with the difference between strength and clear- ness; they were also warned, during the experiments, against confusion of the terms. (2) As regards the pos- 366 NOTES TO LECTURE VI sible influence of feeling, it may be noted that striking individual differences in depth and range of feeling did not, in our three observers, seem to run parallel with the overestimations in question. Finally, (3) against the in- direct effect of memory upon intensity, we may bring the fact that overestimation through attention w^as independent of the interval separating the ' attention ' and the ' distrac- tion' stimuli. It obtained whatever the order: whether distraction came before or after the interval, and therefore whether judgment followed upon the heels of the dis- traction or only after the interpolation of another, attentive consciousness. — "In conclusion: certain strong and weak sounds suffer an intensive reduction under distraction. Whether this reduction represents a general dependency of intensity upon attention, and whether the reduction rests upon physiological or psychophysical grounds, remain questions which demand further investigation." This reference to the * intensifying effect of attention' naturally brought out instances and opinions of a contrary tenor. So far as I see at present, the cases of the 'weak- ening' of an impression by direction of the attention upon it may be classified under the following heads. (1) We prepare for the reception of a very intensive stimulus by protective adjustment of the sense-organ and by inhibition of the start of surprise. Cf. Mliller in Pilzecker, Sinnliche Aufmerhsamkeit, 1889, 80 f. (2) Foregone accommodation of attention, sensory and motor Euistellung, may give a like result. If we are habituated to very heavy weights, a moderately intensive weight will seem light. — Cf. G. E. Mliller and F. Schu- NOTES TO LECTURE VI 367 mann, Pfliiger's Archiv, xlv., 1889, 42 ff. ; and references in my Exj>er. Psychol, II., ii., 1905, 366. (3) Expectation may be 'worse than the reality.' The expected impression may be weakened (a) by the fatigue that follows from the strain of expectation itself, or (6) by the conscious Einstellung, by expectation's overshooting its mark and predisposing us for something more intensive than we actually experience. (4) Intensity may be affected in two ways by the con- currence of other stimuli, (a) An associated whole may be stronger than any one of its constituents. Thus the pain of a dental operation is enhanced by the odour of the room, the sight of instruments, the uncomfortable posi- tion of the jaws and lips, etc., etc. : a resolute fixation of attention on the pain itself will sometimes reveal a sur- prisingly low degree of intensity. Cf. Kiilpe, Outlines^ 398. No doubt, a part is here played by (3) (h). — (6) A sensation may be weakened by its fusion with other sensations; thus an overtone, singled out by anticipatory attention, may appear surprisingly weak. Cf. Muller, Sinnliche Aufmerksamkeit, [1873] 21, 38 f., 71 ff. ; Stumpf, TonpsychoL, ii., 1890, 231. (5) A simple case is that of the weakening of a sensation by peripheral adaptation. The most sustained attention cannot prevent adaptation; and, if sustained attention is there, the adaptive weakening may appear, at first sight, to be the direct result of attention itself. — Where the stimulus is too strong for noticeable adaptation, or where the phenomenon of adaptation is absent, sustained atten- tion may, I think, result in a sort of hypnotic anaesthesia ; but I am not sure upon this point. ^2 J. M. Baldwin, Senses and Intellect, 1890, 63 ff., 68. 368 NOTES TO LECTURE VI '' J. R. Angell, Psychology, 1904, 65 f. - ^'Monist, xiii., 1902, 38 f., 57. ^^Zeits.f. Phil. u. philos. Kritik, ex., 1896, 31 f., 35. Cf. H. E. Kohn, Zur Theorie der AufmerksamkeiU 1894; F. Schumann, Zeits., xxiii., 1900, 24. ^® Art. Psychology, Encyc. Britan., xx., 1886, 47. ^^Fechner, Elem. d. Psychophysik, ii., 1889, 39. Cf. my Exper. Psychol., II., ii., 1905, clxiii. ^* H. R. Marshall, Instinct and Reason, 1898, 38 f. ^^ H. Helmholtz, Zur Lehre von den Tonempjindungen, 1877, 107; Sensations of Tone, 1895, 62. ^"K. Fortlage, System der Psychologie, i., 1855, 102 ff . ; Lotze, Med. Psychol, 1852, 505; A. Tucker, The Light of Nature Pursued, 2d ed., i., 1805, 225. I have not found the metaphor, as I had expected to do, in I. H. Fichte's Psychologie, though the author indicates his famiharity with it in i., 1864, 161. 'MVundt, Physiol. Psychol, 1874, 717; iii., 1903, 33 f. The change was made in the fourth edition of 1893. ^^Ihid., iii., 1903, 117 if., 552, 557. ^' W. Hamilton, Lectures on Metaphysics, i., 1859, 352 f. ; Wundt, ibid., 554 ff. with references. ""^Ihid., 595 f., 600 f. ^^ C. L. Morgan, Introd. to Comparative Psychol , 1894, 14, 19. ^""Princ. of Psychol, i., 1890, 224 ff. " Op. cit, 13 f. ^'Op. cil, 119. ^^ G. Dietze, Untersuchungen liber den Umfang des Bewusstseins bei regelmassig auf einander folgenden Schalleindrucken, Philos. Studien, ii., 1885, 362 ff. ; Wundt, op. cit., 351, 353. NOTES TO LECTURE VI 369 '« Dietze, op. ciL, 391. ^' Op. ciL, 353, 356. ^^F. Schumann, Zeits.y i., 1890, 77 f., 80; ii., 1891, 115 ff.; xvii., 1898, 121; Wundt, Philos. Studien, vi., 1891, 250 ff. ; vii., 1892, 222 ff. ; W. Wirth, ibid., xx., 1902, 561 ff.; J. Quandt, Psychol. Studien, i., 1906, 137 ff. Kiilpe leaves the introspective question open: Outlines y 394. Miinsterberg seems to accept Wundt's view, since he identifies the question of the range of consciousness with the question "wieviel Schallnachbilder bei regel- massig succedierenden Schalleindriicken gleichzeiting in unserem Bewusstsein bleiben": Grundzilge, i., 1900, 214. — Cf . also the discussion of the Psychische Prasenzzeit by L. W. Stern, Zeits., xiii., 1897, 325 ff., and the refer- ences there given. '^ Wundt, Physiol. Psychol, iii., 1903, 353. These are the observations which I believe Wundt has in mind ibid., 119. ^* On cognition, see Wundt, ibid., 535 ff. This explana- tion, which I have given in lectures since 1904, is also offered by K. Mittenzwey, Psychol. Studien, ii., 1907, 386. ^^ See W. Wirth, Die Klarheitsgrade der Regionen des Sehfeldes bei verschiedenen Verteilungen der Aufmerk- samkeit, Psychol. Studien, ii., 1906, 30 ff. ; K. Mittenzwey, Ueber abstrahierende Apperzeption, ibid., 1907, 358 ff . ; A. Kastner and W. Wirth, Die Bestimmung der Auf- merksamkeitsverteilung innerhalb des Sehfeldes mit Hilfe von Reaktionsversuchen, ibid., iii., 1907, 361 ff. ^« Angell, Psychology, 1904, 65. "James, Princ. of Psychol, i., 1890, 237 ff., 284 ff.; esp. 255, 258 f. In his last section James is dealing with clearness and obscurity. "We actually ignore most of 370 NOTES TO LECTURE VI the things about us." "Attention, . . . out of all the sensations yielded, picks out certain ones as worthy of its notice and suppresses all the rest." Here, however, is no mention of 'fringes/ ^^ Certain of my hearers objected to this argument that I had worked out the ' law of the two levels ' on the basis of the observation with the puzzle picture. This, they contended, represents an exceptional case : a consciousness may, in fact, show only a single level or may show a great variety of levels at a given moment. The first part of the objection does not hold. I had worked out the 'law' long before I thought of the use of the puzzle-picture; it was in the course of an extended search for a suitable illustration of the law that the puzzle- picture occurred to me. I give the illustration because it seems to me to present, in clear and striking form, what is the normal state of affairs ; but it is this latter, the nor- mal conformation of consciousness, that I am concerned with. Continued observation of the conscious levels under very different circumstances, in the laboratory and in everyday life, led me to the law. The second part of the objection raises the question of fact. So far, it is both legitimate and welcome. When, however, I pressed for instances, I found the following sources of error: (1) confusion of peripheral with atten- tional clearness ; (2) confusion of attentional clearness with cognition; (3) confusion of a single consciousness with a series of consciousnesses, and therefore of a single 'act of attention ' w^ith several successive acts ; and ^ven (4) con- fusion of the question of the conscious 'levels' with the question of the 'range' of attention. I got no clear case either of a single-levelled or of a many-levelled consciousness. NOTES TO LECTURE VI 371 In the abstract, the objection usually took this form. *An idea or a presentation may pass very slowly from the background to the focus of consciousness; it does not always, does not ordinarily, jump from the one to the other. Hence a cross-section must show ideas or presenta- tions in all sorts of intermediate positions between ob- scurity and maximal clearness; there are many levels.' I grant, of course, that we do not perceive anything of the nature of a jump, — but I cannot either find anything of the nature of the slow passage. Now a certain idea is obscure; now, without conscious transition, it is clear. I admit, also, that our response to an intruding stimulus, when the attention is already engaged, varies widely with variation of conditions; I discuss some instances in Lecture VII., under the Maw of temporal instability.' But I do not find, in this variation of response, any evidence of new levels. For the rest, observation of consciousness in the rough is ahvays unsatisfactory; the appeal lies to experiment. I may add — what my critics did not suggest — that in abnormal circumstances consciousness may conceiv- ably show but one level; or, at any rate, that the normal relation of the two levels may be radically changed. The narrowed consciousness of profound hypnosis may, at moments, be wholly clear; the idiotic consciousness may be wholly obscure. Cf. my Primer of Psychology, 1902, 273. ^' ToiipsychoL, i., 1883, 309. '" Wundt, Physiol Psychol, iii., 1903, 96 f., 337 f., 434, 439. Here we are already encroaching upon the field of inertia. — On accommodation-time in reaction experi- ments, cf. G. della Valle, Psychol Studien, iii., 1907, 294 fF. 372 NOTES TO LECTURE VI "See G. Kafka, Psychol. Studien., ii., 1906, 'HBQ ff.; B. Berliner, ibid., iii., 1907, 91 ff., with references. In order to bring out my point clearly and sharply, I have spoken in the text almost as if the ^n^^i^^-determina- tions might be transferred bodily from their present in- tensive context to that of clearness. I have, it is true, safeguarded the proposal by insisting on the necessity of 'interpretation.' However, my idea may be expressed more accurately — and more cautiously — as follows. Clearness and intensity are both involved in the determi- nations; it is evidently wrong to ascribe everything to intensity and nothing to clearness. Let us, then, take up the Anstieg-G^uQsiion from the side of clearness, varying our method in such a way as to secure varying degrees of clearness. We shall then be able to give the earlier results a setting in which due regard is paid to each one of the two concurrent factors. ^"^ r attention, 1906, 17; Attention, 1908, 13. ^^ G. T. Fechner, Revision der Hauptpuncte der Psy- chophysiky 1882, 283; cf. Ueber einige Verhaltnisse des binocularen Sehens, Ahhandl. d. kgl. s. Ges. d. Wiss., vii., 1860, 395: "bei der willkuhrlichen Richtung der Auf- merksamkeit selbst ist der bewusste Willensact, durch den wir die Aufmerksamkeit richten, von dem Erfolge, d. i. der gerichteten und fixirten Aufmerksamkeit, wohl zu unterscheiden. Jener Act erfolgt ein- flir allemal, und dann bleibt die Aufmerksamkeit gerichtet, ohne dass wir einen fortgesetzten oder neuen bewussten Willensact nothig haben, sie in dieser Richtung zu erhalten." — Stumpf, TonpsychoL, i., 1883, 386; cf. 244, 391; ii., 1890, 318, 358. F. Auerbach, in Wiedemann's Annalen, iv., 1878, 509 f. F. Schumann, Nadir, d. Ges. d. Wiss. zu NOTES TO LECTURE VI 373 Gottingen, 1889, 536 ff. ; Zeits., iv., 1893, 1 ff • xxiii 1900, 9; McDougall, Uind, N. S.. xv., 1906, 349 Here belong, at least in part, the experiments on word- exposure with previous suggestion; certain phenomena of sensor £z».^e««„^,- absolute impression, etc.; and perhaps also certain optical illusions. " On Perseveralicnstendenz see, provisionally, Ebbing- haus. Gruruizuge, i., 691; Wundt, Physiol. Psychol., l, '^ See Wundt, op. cit, 45 flf. NOTES TO LECTURE VII ^ James, Princ. of Psychol, i., 1890, 427 ff . ; Wundt, Physiol. Psychol. y iii., 1903, 410 fF. ; Ebbinghaus, Grund- zuge, i., 1905, 614 ff. 2 Wundt, op. cit, 64 ff. ^ H. C. Stevens, A Simple Complication Pendulum for Qualitative Work, Am. Journ. Psychol., xv., 1904, 581. ^ M. Geiger, Neue Complicationsversuche, Philos. Stu- dien, xviii., 1903, 347 ff. See also the references there given. ' Wundt, op. cit, 67. ® W. von Tchisch, Ueber die Zeitverhaltnisse der Apper- ception einfacher und zusammengesetzter Vorstellungen, untersucht mit Hulfe der Complicationsmethode, Philos. Studien, ii., 1885, 603 ff., esp. 621 f. ^ Physiol. Psychol., ii., 1880, 272 ff. Geiger (op. cit., 400) says that Wundt's and von Tchisch's explanations are 'grundverschieden.' So they are! But I think that von Tchisch, if he believed that he was repeating Wundt, had some excuse for his mistake. He says (622) : " die Apperception reproducirt unmittelbar den Eindruck." Wundt says (273) : " es wird auch von der Apperception der Eindruck unmittelbar reproducirt." (Cf. also the passage in Lehrhuch der Physiologic des Menschen, 1878, 793: "das Centralorgan scheint auf einen erwarteten Eindruck so sich vorzubereiten, dass der Vorbereitungsact selbst, wenn er eine gewisse Intensitat erreicht, zur Erregung wird.") All through the exposition of 1880 374 NOTES TO LECTURE VII 375 Wundt makes very incautious use of the ' Erinnerungsbild ' : it is not difficult to read 'hallucination' into his pages. In the edition of 1887 (ii., 339 f.) the* reproduction' and the ' memory image ' have disappeared, and Wundt's theory stands out in sharp contrast to von Tchisch*s. Since James had the edition of 1887 in his hands, when preparing his chapter on Attention for the press, Geiger is justified in charging him with an unwarranted confusion of the two views. ' Op. cit, 415 f. * Geiger, op. cit., 399 ff. It is to be noted that the ex- planation offered by Ebbinghaus in Grundziige, i., 1902, 593, does not appear ibid., 1905, 615, save in the reference to "allerlei Ueberlegungen." ^^ Physiol. Psychologies 1874, 767. The phrase appears in all subsequent editions. ^^ Geiger, op. cit., 409 ff. I have, of course, taken the simplest possible case. ^''Phijsiol. Psychol, iii., 1903, 75 f., 86. " Ibid., 352. " Grundziige, {., 618 ff. Ebbinghaus is, perhaps, follow- ing James: Princ, i., 409. *^ Op. cit., 334, 352. The italics are in the original. ^^Op. ciY., 621. — Both Ebbinghaus (620) and James (408) refer here to the experiments of F. Paulhan (ReviLe scientifique, 3 S., xiii., 1887, 684). Neither these nor the kindred experiments of A. Binet {Rev. philos., xxix., 1890, 138) appear to me, however, to bear out the conclusions derived from them. I have myself repeated and extended PauDian's work, and have attained, with practice, to mark- edly greater time-differences. But the results belong to the domain of habit, of * normal motor automatisms,' — 376 NOTES TO LECTURE VII where analysis is not difficult, but where we gain no know- ledge of the range of attention. Simultaneity of two psy- chologically disparate 'attentions' is, in my experience, altogether impossible. — Cf . S. E. Sharp, Amer. Journ. Psychol, X., 1899, 356, 381. It is noteworthy that, in the case reported by R. d'Allonnes {Rev. philos.. Deer. 1905, 592 ff.), there was, apparently, no loss of attention with loss of feehng: see pp. 611 f. Cf., however, Lect. II., note 2. " Op, cit, 404. '« Op. cit, 634. '' Op. cit., 366. 2' Op. cit, 520; cf. 518 ff., 544 ff., 558 ff. ^^ Principles, i., 237. The reader may be reminded that the distinction between ' substantive ' and ' transitive ' parts of the stream of thought, in James' exposition, is a distinc- tion in terms of time alone ; it does not imply discontinuity. *' The successive psychoses shade gradually into each other, although their rate of change may be much faster at one moment than at the next" : ibid., 243. 2' Op. cit., 624. ^^ References to the experiments on touch are given by L. R. Geissler, Fluctuations of Attention to Cutaneous Stimuli, Amer. Journ. Psychol., xviii., 1907, 309 ff. Cf. the general account in my Exper. Psychol., I., ii., 1901, 194 ff. ; and, for the *Tastzuckungen,' J. Czermak, Sitzungsher. d. mathem.-naturw. Classe d. kais. Akademie d. Wissen- schaften zu Wien, xv., 1855, 486 f. {Physiol. Studien, ii., 64 f.). ^ References are given by K. Dunlap, The Fluctuation of Diapason and Gas Flame Tones, Psychol. Rev., xi., 1904, 314 ff. See also W. Heinrich, Zeits. f. Sinnesphysiologie, xli., 1906, 57. NOTES TO LECTURE VII 377 ^^ See my Exper. Psychol., loc. cit. ; C. E. Ferree, An Experimental Examination of the Phenomena usually attributed to Fluctuation of Attention, Amer. Journ. Psychol, xvii., 1906, 84, 94 f. ; J. W. Slaughter, ibid., xii., 1901, 331; W. Heinrich, Sur la fonction de la membrane du tympan. Bull, de V Academie des Sciences de Cracovie, July, 1903, 536 ff. ; Ueber die Intensitatsanderungen schwacher Gerausche, Zeits. f. Sinnesphysiologie, xli., 1906, 57 f. ; Ueber das periodische Verschwinden kleiner Punkte, ibid., 59 ff. Heinrich's view of auditory accom- modation is that the drum-skin reacts to a given tone in very different states of tension, so that the ^pulsations of the ten- sor tympani have no effect upon tonal hearing. On the other hand, the adjustment of the membrane to noise is extremely delicate ("das Trommelfell ist ausserst fein auf Gerausche gestimmt ; . . . man ist erstaunt zu sehen, wie indifferent das Trommelfell gegen Gerausche ist, bis man zu der richtigen fiir das Gerausch entsprechenden Span- nungkommt"), so that after accommodation is effected the slight changes of tension due to the pulsating muscle make themselves apparent in sensation. 2« C. E. Ferree, Amer. Journ. Psychol., xvii., 1906, 81 ff., esp. 83; xix., 1908, 58 ff., esp. 129; C. Hess, Arch. f. Ophthalmol., xl., Abth. 2, 1894, 274 ff. ; B. Hammer, Zeits. f. Psychol, u. Physiol, d. Shmesorgane, xxxvii., 1905,363 ff., esp. 365, 375. Hammer has the priority of extended pub- lication ; but I have given first place to Ferree in my text because his work was begun, and his theory already out- lined, in 1903. A brief report will be found in Journ. Philos. Psychol. Sci. Meth., i., 1904, 240; Science, N. S., xix., 1904, 659. Hammer is criticised by C. E. Seashore in Zeits., xxxix., 378 NOTES TO LECTURE VII 1905, 448 ff. With the critique of the auditory experi- ments I am in general agreement. I do not understand, however, how Seashore can say of adaptation and eye- movement : " die wichtige Rolle der genannten und anderer physiologischer Momente ist wohlbekannt.'* I suppose that he refers to Pace, who had written in 1902: "for the eye, the ' peripheral ' includes the retina ; and, so far as I am aware, the retinal conditions as affected by the fluctua- tions have not been investigated'* (Philos. Studieriy xx., 234). Pace himself works in terms of retinal * fatigue' (242), i.e. of local adaptation: but he brings peripheral fatigue into speculative connection with 'central changes' and the process of accommodation, and says nothing what- ever of eye-movement (244). I know of no further refer- ence before 1904, when G. E. Miiller sets Hess' observations in the perspective of the experiments on attentional fluc- tuation (Gesichtspunkte und Tatsachen, 110), and the two notes on Ferree's work appear in Woodbridge's Journal and in Science. — In a paper entitled The Fluctuation of Visual Stimuli of Point Area, read by title at the 5th Annual Meeting of Experimental Psychologists (Cambridge, Mass., April 15-17, 1908) and to be published in the Amer. Journ. Psychol., Mr. Ferree reports a repetition and extension of the experiments of Heinrich, referred to in the fore- going'^Note, and concludes that " in so far as adaptation- tests can be applied, these stimuli follow the laws of adaptation and recovery, in the phase-relations of their fluctuations, as closely as do the areas commonly em- ployed." Positive evidence is also adduced against Heinrich's theory of lenticular pulsation. Mr. Ferree informs me, further, of the following re- NOTES TO LECTURE VII 379 suits of unpublished experiments with auditory stimuli. (1) The tone of an electrically driven tuning-fork does not fluctuate at the limen, and objective interruptions of the sound are at once remarked. (2) Of three trained observers, tested at the same time with the watch-tick, two reported no fluctuations (90 sec.-2 min.), while the third gave fluctuations of the orthodox sort. Unfor- tunately, the positions of the observers were not inter- changed. Hammer has done good service in calling attention to the objective inconstancy of the watch-tick (op. cif., 371 ff.). I am sure, however, that this observation must be supplemented, for explanatory purposes, by reference to sound-reflections or similar phenomena. We were accustomed, in the early nineties, to perform Sanford's expenment (5Sb, A7ner. J ourn. Psychol., iv., 1891, 307; 61 6, Course in Exper. Psychol., 1898, 55) in the Cornell Laboratory with a number of students simultaneously. We found, as I remember, occasional instances of absence of fluctuation, and a good many cases of approximately coincident fluctuation ; but we also found many cases of non-coincidence. " See Stumpf, Tonpsychol., i., 40 f., 360. ^^Helmholtz, Physiol. Opiik, 1896, 242, 510; so J. Muller, Ueber die phantastischen Gesichtserscheinungen, 1826, 15 f . (" diese Lichterscheinung war mit dem Ausath- men synchronisch"). Fechner, curiously enough, found no such oscillation : Revision, 1882, 32. — A. Lehmann, Philos. Studien, ix., 1894, 66 fF. ; J. W. Slaughter, Amer, Journ. Psychol, xii., 1901, 329 ff. ^^ The whole series of related articles is as follows : Slaughter, The Fluctuations of the Attention in Some of 380 NOTES TO LECTURE VII their Psychological Relations, Amer. Journ. Psychol., xii., 1901, 313 fF. ; R. W. Taylor, The Effect of Certain Stimuli upon the Attention Wave, ibid.y 335 ff. ; H. C. Stevens, The Relation of the Fluctuations of Judgments in the Estima- tion of Time Intervals to Vaso-motor Waves, ibid., xiii., 1902, 1 ff. ; W. B. Pillsbury, Attention Waves as a Means of Measuring Fatigue, ibid., xiv., 1903, 541 ff. ; C. E. Gallo- way, The Effect of Stimuli upon the Length of Traube- Hering Waves, ibid., xv., 1904, 499 ff . ; B. Killen, The Effects of Closing the Eyes upon the Fluctuations of the Attention, ibid., 512 ff. ; G. L. Jackson, The Telephone and Attention Waves, Journ. Phil. Psychol. Sci. Meth., iii., 1906, 602 ff. Cf. also Pillsbury's general account in Vattention, 1906, 90 ff. ; Attention, 1908, 69 ff. ; F. G. Bonser, A Study of the Relations between Mental Activity and the Circulation of the Blood, Psychol. Rev., x., 1903, 120 ff. Ct. W. Mc- Dougall, Mind, N. S., xv., 1906, 356 f. ; H. Berger, Ueber die korp. Aeusserungen psych. Zustdnde, ii., 1907, 153 ff. ^^ S. Exner, Entwurf zu einer physiologischen Erkldrung der psychischen Erscheinungen, i., 1894, 302 f. ; Pillsbury, Amer. Journ. Psych., xiv., 552. Exner writes (302) : " nach meinen Selbstbeobachtungen diirfte die Dauer der gleichmassigen Lebhaftigkeit einer Vorstellung kaum eine Secunde sein." Even if we put the very strictest interpretation upon ' gleichmassig,' the state- ment seems curiously exaggerated. ^^ Philos. Studien, xx., 1902, 234. ^^ Lehmann, op. cit. ; other references in Ferree, Amer. Journ. Psychol., xix., 1908, 58 ff. ^^ Cf. The Problems of Experimental Psychology, Amer. Journ. Psychol., xvi., 1905, 218, NOTES TO LECTURE VII 381 In the first edition of his Grundzuge (i., 1897, 263) Ebbinghaus wrote as follows: "Nun sind aber doch die Licht- und Farbenempfindungen, wie wichtig sie auch immer als Material flir weitere Verarbeitungen sein mogen, an und fur sick noch relativ niedere und elementare Be- thatigungen der Seele, das sie vermittelnde Organ ist ein Aussenwerk des eigentlichen Seelenorgans. Wenn also schon das vergleichsweise Einfache sich der eindringenden und intensiven Beschaftigung mit ihm als ein ungeahnt und fast verwirrend Reichhaltiges enthiillt, -wie mag es erst mit dem hoheren Seelenleben, das doch zweifellos etwas be- trachtlich Verwickelteres ist, in Wahrheit bestellt sein?" I fear that there is here a * trace,' as the analysts say, of an untenable genetic psychology. That apart, I am — as the above quotation shows — unable to see the force of Ebbing- haus' argument. The passage is not reprinted in 1905. What are we to say, however, to the results of H. Berger (Korp. Aeusserungen, ii., 1907, 118 ff., 181 ff.), who was able, in Fechner's words, to look into the brain of another person, and who there saw the apperception waves with all desirable plainness ? This, surely : that the introspective control which Berger finds lacking in his first series of experiments (139 f.) is equally necessary for the series made by * Zoneff's method.' There is no evidence that the ' inattention ' of his observers was not an * attention to something else.' The same criticism holds of the experiments of Zoneff and Meumann, so far as their report has been published ; we are not told any- thing in detail of the * Nachlassen der Aufmerksamkeit ' {Philos. Studien, xviii., 1903, 46). But, apart from this, it was a paradox of the older investigations that fluctua- tion of attention occurred without any subjective remis- sion of attention. 382 NOTES TO LECTURE VII ^* L. R. Geissler, Amer. Journ. Psychol., xviii., 1907, 310 f. ; E. A. Pace, Philos. Studien, xx., 1902, 244. The fluctuation of two simultaneously presented stimuli offers no new difficulty. I have myself made prehminary experiments upon memory-images, without observing fluctuation. See, however, N. Lange, Philos. Studien, iv., 1888,408 ff.; H. Eckener, ihid., viii., 1893, 370, 379; H. Munsterberg, Beitr. zur exper. Psychol., ii., 1889, 119 ff. The illusions of reversible perspective (Lange, 406), which still figure in Pillsbury's account {U attentioriy 93 ; Attention^ 1908, 71), have been ruled out of court by Wundt himself, who finds their primary conditions in the physiological pro- cesses of fixation and eye-movement {Die geometrisch- optischen Tduschungen, 1898, 23 [Abh. d. mathem.-phys. CI. d. kgl. sdchs. Ges. d. Wiss., xxiv., 75]; Phys. Psych., ii. 1902, 545 ff.). ^^ Klilpe, Outlines, 429 ; Ebbinghaus, Grundziige, 623 f. ^' See A. J. Hamlin, Amer. Journ. Psychol., viii., 1896, 3 ff. ; F. E. Moyer, ihid., 1897, 405 ; L. G. Birch, ibid., ix., 1897, 45; L. Darlington and E. B. Talbot, ibid., 1898, 332 ; E. B. Titchener, ibid., 343. My outline of method is (as I say later in the text) entirely schematic ; but I think that with time and patience and technical skill the method itself can be carried through. And I know of no other that will serve the same purpose. It has often been proposed that the method of distraction should be applied objectively, without introspective control. See, e.g., A. Bertels, Versuche liber die Ablenkung der Aufmerksamkeit, 1889 ; E. J. Swift, Amer. Journ. Psychol., v., 1892, 1 ff. ; KUlpe, op. cit., 428 f. ; E. Krapelin, Psychol Arbeiten, I, 1895, 57 ff . ; R. Vogt, ibid., iii., 1899, 62 ff . ; W. McDougall, Brit. Journ. Psijchol., l, 1905, 435 ff. NOTES TO LECTURE VII 383 Stumpf sees the difficulty, but does not suggest a way out : TonpsychoL, i., 74 f. Pillsbury, in his new chapter in Attention, 89 fF., treats the method in this objective way, and thus naturally — but mistakenly — regards the work published from the Cornell Laboratory not as preliminary, but as done for its own sake. I have, however, never be- lieved that the method of distraction, taken objectively, could furnish any psychological result, and I can therefore subscribe to Pillsbury's criticism. In 1898 I wrote as fol- lows : " The three Studies . . . were undertaken with the view of discovering a means of distraction that should be capable of gradation, uniform in its working and applicable to normal subjects. With such a distraction it would be possible, on the qualitative side, to describe the attributes of mental processes given in the state of inattention, and, on the quantitative, to measure the magnitude and delicacy of sensitivity and sensible discrimination in the same state " (343 f. : the italics are not in the original). The 'state of inattention' is a clumsy expression, but it is evident that my psychological appeal was to lie to introspection. ^^ Stumpf , TonpsychoL, i., 73 f . ; H. Munsterberg, Die Willenshandlung, 1888, 72; Beitr. z. exper. Psychol., ii., 1889, 24. ^* See, e.g., B. Bourdon, Observations comparatives sur la reconnaissance, la discrimination et Tassociation, Rev. philos., xl., 1895, 166 ff . ; E. Toulouse et N. Vaschide, Atten- tion et distraction sensorielles, Compt. rend, de la soc. de bioL, 1899, 964 ff. ; A. Binet, Attention et adaptation, Annee psychoL, vi., 1900, 248 ff. ; F. Consoni, La mesure de Tattention chez les enfants faibles d'esprit (phrenas- theniques). Arch, de Psychol, ii., 1903, 209 fF. ; W. Peters, Aufmerksamkeit und Reizschwelle : Versuche zur Mes- 384 NOTES TO LECTURE VII sung der Aufmerksamkeitskonzentration, Arch. f. d. ges. Psychol, viii., 1906, 385 ff. ; P. Janet, The Mental State of Hysiericals, 1901, 70 ff. ; Munsterberg, Die Association successiver Vorstellungen, Zeits.f. Psychol, i., 1890, 99 ff. ; W. G. Smith, The Relation of Attention to Memory, Mind, N. S., iv., 1895, 47 ff. ; T. Ziehen, Ein einfacher Apparat zur Messung der Aufmerksamkeit, Monatsschr. f. Psy- chiat. u. Neurologic, xiv., 1903, 231. On the use of the MV, see A. Oehrn, Experimentelle Studien zur Individualpsychologie, Psychol Arheiten, i., 1895,92 ff., esp. 128, 138; V. Henri, yl?m^e psy clwl, il, 1897, 245; J. J. van Bierv^Het, Journ. de Psychol, l, 1904, 230; A. Binet, A7in. psychol, xi., 1905, 71. On the use of the fluctuation-values, see E. Wiersma, Zeits.f. Psychol, xxviii., 1902, 180 ff . ; Pillsbury, Amer. Journ. Psychol, xiv., 1903, 541 ff. ^^ For a list of the investigations of attention by the ex- pressive method, see H. C. Stevens, Amer. Journ. Psychol, xvi., 1905, table facing 469; and add E. A. McC. Gamble, ibid., xvi., 261; M. Kelchner, Arch, f d. ges. Psychol, v., 1905, 7 ff.; H. Berger, Korp. Aeusserungeyi, i., 1904, 77 ff.; ii., 1907, 40 ff., 118 ff., 167 ff. Noteworthy is Binet's suggestion of immobility: Aniiee psychol, vi., 1900, 279. NOTES TO LECTURE VIII ^ With these paragraphs, cf. Wundt^ Physiol. Psychol., ii., 1902, 362 ff. ; also the insertion from the fourth edition in Princ. of Physiol. Psychol., i., 1904, 21 ff. ; Orth, GefUhl u. Bewusstseinslage, 1903, 6 ff. (esp.the remarks onTetens). ^ Amer. Journ. Psychol., xvi., 1905, 213. ^ See the discussion in Psychol. Bulletin, iii., 1906, 52 ff. * J. Merkel, Philos. Studien, iv., 1888, 594; v., 1889, 245. ^ Phijsiol. Psychol, ii., 1902, 369. ^ The first idea of the theory here outHned came to me some years ago — in 1901 or 1902 — in the course of con- versation with my then assistant. Professor G. M. Whipple. How much of it belongs to Dr. Whipple and how much to myself I cannot now say, and I imagine that Dr. Whipple is in the same case. In its general features, the theory seems to resemble that put forward by M. F. Washburn {Journ. Philos. Psychol. Sci. Meth., iii., 1906, 62 f.). I do not agree, however, that mental processes may appear, in alternation, as feelings and as organic sensations or ideas. Another similar view is that of R. Lagerborg, Das Gefiihls- 'problem, 1905, 36 ; Arch.f. d. ges. Psychol, ix., 1907, 455 f. ^ Cf. the terminological note in Orth, Gefilhlu. Bewusst- seinslage, 1903, 5; and H. N. Gardiner, Journ. Philos. Psychol Sci. Meth., iii., 1906, 57 f. * Ebbinghaus, Grundziige, 568 ff. ® J. J. Thomson, The Corpuscular Theory of Matter, 1907, 1. " From the point of view of the physicist, a theory of matter is a policy rather than a creed ; its object is to 2 c 385 386 NOTES TO LECTURE VIII connect or coordinate apparently diverse phenomena, and above all to suggest, stimulate, and direct experiment." '^^ Op. cit, i., 1902, 577 f. ; i., 1905, 602 f. "G. F. Stout, Analytic Psychol, i., 1896, 224 ff.; cf. Manual of Psychol, 1899, 232 ff. '2 Kulpe, Outlines, 272. ^^ U attention, 1906, 72; Attention, 1908, 55. "Things are interesting because we attend to them, or because we are likely to attend to them ; we do not attend because they are interesting." — Cf. with this discussion F. Arnold, The Psychology of Interest, Psychol Rev., xiii., 1906, 221 ff., 291 ff. ; Interest and Attention, Psychol Bulletin, ii., 1905, 361 ff. ; W. H. Burnham, iVttention and Interest, Amer. Journ. Psychol, xix., 1908, 14 ff. ^^ Physiol. Psychol, iii., 1903, 342: the following pages give the distinction between active and passive appercep- tion. Cf. Grundriss, 1905, 266 (Engl, 1907, 246). ^^ Tonpsychol, ii., 1890, 283. Voluntary attention is " nichts Anderes als der Wille, sofern er auf ein Bemerken gerichtet ist." Involuntary attention may pass into volun- tary : " sie ist nicht mehr davon verschieden, als der Wille uberhaupt von Lusgefuhlen verschieden ist. Fassen wir * Geflihl ' im weiteren Sinne, so kann der Wille ja selbst zu den GefUhlen, und zwar natlirlich zu den positiven Ge- fUhlen, gerechnet werden." The whole passage, 277 ff., is interesting, though there are parts that I do not find very clear. ^' Grundziige, I, 1905, 588, 607, 610 f . — On the other side, cf. A. Marty, Vjs.f. wiss. Philos., xiii., 1889, 195 ff. ^' Op. cit, 603. ^^ Physiol Psychol, iii., 1903, 279; cf. Grundriss, 1905, 230 ff. (Engl., 1907, 213 ff.). The doctrine appears first NOTES TO LECTURE VIII 387 in the Physiol. Psychol, of 1880 (ii., 410), and is worked out in the essay on Die Entwicklung des Willens, Essays, 1885, 286 ff. In Essays, 1906, 344, Wundt ascribes his " Bekehrung zu einem psychologisehen Voluntarismus " to two influences: the positive indications of his own experiments on reactions, and the negative effect of J. Baumann's intellectuahsm. Now the reaction experiments were done ten years before the second edition of the P. P. appeared, whereas Baumann's Handbuch der Moral was pubHshed in 1879. Here, then, is another instance of the movement of Wundt's thought, as I have characterised it in Lecture IV. : the voluntaristic idea had been ' incubated ' for a decade ; it was gradually maturing in Wundt's mind ; and Baumann furnished the external stimulus that brought it to clear expression. ^® Art. Psychology, Encyc. Brit, xx., 1886, 43. 2« E. D. Cope, The Origin of the Fittest, 1887, 395, 413, 447. Cope's essays are the more interesting as he seems to have worked out his ideas independently, without know- ledge of contemporary psychology ; he makes at most only a casual reference to Carpenter or Bain. ^^ Cf. my paper in the Pop. Sci. Monthly, Ix., 1902, 458 ff. I should now replace the Wundtian argument, 467 f., by pointing out that there does not appear to be any reflex movement — heart-beat, widening and narrowing of the pupil, etc. — that may not be brought, to a certain de- gree, under 'conscious control'; and I should urge that this state of affairs probably indicates a resumption, not an usurpation of sovereignty. Cf . G. H. Lewes, The Physical Basis of Mind, 1877, 367 ff. I add only, to avoid possible misunderstanding, that my own position is that of parallelism, not of interactionism ; 388 NOTES TO LECTURE VIII and that there is no reason to be scared by the bogey of *the inheritance of acquired characters.' There are more ways than one of speculating oneself out of a biological difficulty ! "^^ Physiol Psychol, iii., 1903, 348; cf. 116. 2^ Grundriss, 1905, 262 (Engl., 1907, 243). ^^ See Lecture VI., note 38, sub fin. ^ I have avoided any detailed reference to the central conditions of attention. The most recent accounts are those of W. McDougall {Mind, N. S., xi., 1902, 316; xii., 1903, 289, 473; xv., 1906, 329; cf. Physiol Psijchol, 1905, 90 ff.) and Ebbinghaus (Grundzuge, 1905, 628 ff.). Pills- bury gives a general review of theories in his Attention^ 1908, chs. xiv. ff. As for the central conditions of affec- tion, I do not see that we need travel beyond the Korper- fUhlsphdre ; but this is mere guesswork. 2' Physiol Psychol, i., 1893, 588, 590. ^' Ibid., ii., 1902, 357. 2« Grundriss, 1905, 263 (Engl., 1907, 244). " Jeder In- halt des Bewusstseins libt eine Wirkung auf die Aufmerk- samkeit aus, infolge deren er sich teils durch seine eigene Gefuhlsfarbung, teils durch die an die Funktion der Auf- merksamkeit gebundene Gefuhle verrat. Die gesamte Ruckwirkung dieser dunkel bewussten Inhalte auf die Aufmerksamkeit verschmilzt dann aber, gemass den all- gemeinen Gesetzen der Verbindung der Gefuhlskom- ponenten, mit den an die klar bewussten Inhalte gebun- denen Gefuhlen zu einem einzigen Totalgefuhl." Here it is the obscure contents that react upon the attention ! We must surely conclude that the doctrine has not settled down to final form. Indeed, I am disposed to think that the section on 'Die Gefuhle als psychophysische Vorgange' is NOTES TO LECTURE VIII 389 intended to convey that idea {Physiol. Psychol. y ii., 1902, 358 ff., esp. 362). Pillsbury, in his Attention (1908, 189 ff.), appears to refer to the Wundtian doctrine of 1893, and not to that of the current edition of the Physiologische Psychologie. ^^ Physiol. Psychol, iii., 1903, 341, 342 ff . ; Grundriss, 1905, 264 f. (Engl., 1907, 244 f.). ^^ See, e.g., Orth, Gefiihl u. Bewusstseinslage, 1903, 50. ^^Princ. of Psychol, i., 1890, 300. ^^ H. E. Kohn, Zur Theorie der AufmerJcsamkeity 1894, 48; cf. my Exjper. Psychol, I., ii., 1901, 210 f. ^ Kohn says (op. cit., 36) : " wenn Wundt statt wir * ich ' gesagt hatte, so konnte man ihn den Satz [wir nehmen in uns in wechselnder Weise mehr oder weniger deuthch eine Thatigkeit wahr] nicht bestreiten. Ich muss jedoch dem gegeniiber wiederholen, dass ich bei der sorgfaltigsten Prii- fung meiner Bewusstseinslage nur selten ein solches Ge- fiihl gefunden habe." ^ Grundriss, loc. cit. ^^ Physiol Psychol, iii., 1903, 342 ; Grundriss, 265 (Engl., 245). ^^ Physiol Psychol, 341. " Mind, N. S., xi., 1902, 342 f. "The complexity of the upper levels [of the nervous system], their numerous inter- connections, the extreme variability of the resistances pre- sented by them, and the number of alternative paths that may be opened in turn to the excitation-process, are the physiological basis of the ' Lebhaftigkeit ' of the presenta- tion." '« Grundzuge, i., 1905, 628 ff. The effect of Ebbinghaus' cortical Hemmungen and Bahnungen is " die HerbeifUhrung diffuser und sich verlaufender Erregungen einerseits, kon- zentrierter und differenzierter Erregungen andererseits." 390 NOTES TO LECTURE VIII ^T attention, 1906, 194; Attention, 1908, 284. ^" It is, of course, always possible to fall back upon blood- pressure and rate of pulse and respiratory changes, and so to save the motor character of the organism. I do not doubt that these internal reactions occur. But we are talking attention: and to make them available for the theory of attention, it must be shown, first, that a concomitant varia- tion actually obtains, and then, secondly, that it is relevant, — that the two series of correlated phenomena are not referable to a common set of conditions. As things are, the observations upon the first point are comparatively rough, and the evidence available upon the second does not favour the motor hypothesis. See Pillsbury, Attention, 282 f . On * motor' psychology in general, see I. M. Bentley, Amer. Journ. Psychol., xvii., 1906, 293 ff., and the refer- ences there given; as well as my Exp. Psychol., II., ii., 1905, 364 ff. *^ r attention, 1906, 280 f . ; Attention, 1908, 311 ff. *2 Grundzuge, i., 1905, 606 f. There are, Ebbinghaus declares, "gewisse reflektorisch ausgeloste Bewegungen," that we sense "als mannigfache Spannungen oder Betati- gungen, ohne sie doch zumeist bestimmt zu lokalisieren, d. h. : man empfindet ganz allgemein sich als angespannt oder tatig, indem man aufmerksam ist." ^^ Physiol. Psychol, iii., 1903, 254 ff., 342 ff. ; Grundriss, 223 ff., 264 f., 266 (Engl., 207 ff., 244 f., 246 f.). ^* We have mentioned this law above, p. 34 of the text. Wundt uses it, in connection with action, Physiol. Psychol., iii., 1903, 277 ff., 471 ff . ; Grundriss, 230 ff., 239 f. (Engl., 213 ff., 221 f.); and esp. Die Sprache, l, 1900, 31 ff . ; i., 1904, 37 ff. It is also employed by G. H. Lewes, passim; see, e.g.. Problems of Life and Mind, l, 1874, 134 ff., 226 ff ; NOTES TO LECTURE VIII 391 iii., 1879, 93 ff., 143 ff., 397 ff., 432 f . ; Physical Basis of Mind, 1877, 322 ff ., 367 ff. ; Study of Psijchology, 1879, 19 ff., etc. '^ Cf. my Primer of Psychol, 1902, 76 f. ; Outline, 1902, 135 ff., 139 f. ^^ Grundzuge, 607, 610 f. "Bei der unwillkUrlichen Auf- merksamkelt ist weiter nichts vorhanden als [ein energisch hervortretender interessierender Eindruck und Spannungs- oder Tatigkeitsempfindungen], bei der willkiirlichen kommt noch hinzu eine unablassig den Eindruck als bevorstehend oder als fortdauernd vorwegnehmende Vorstelliing. Sie verhalten sich also zueinander wie Trieb und Wille." ^^ Ebbinghaus writes (op. cit., 611): "dass ich endlich sachlich die Beschreibung des Unterschiedes zwischen pas- siver und aktiver Apperception, als eines einfachen, nur durch ein Motiv hestimmten Wollens und eines zwischen mehreren Motiven wdhlenden Wollens, nicht zutreffend finden kann, gelit aus der oben gegebenen abweichenden Darstellung dieses Unterschiedes hervor." This formula- tion is not quite fair to Wundt, since the Wahlhandlung is, for him, just as much ' bestimmt ' as is the Triehhand- lung. As for the 'nicht zutreffend,' I have shown in the text that Wundt's distinction includes that of Ebbinghaus, and simply adds a causal to the common descriptive ac- count. The terminological issue, of the definition of ' will, ' is, as Ebbinghaus says, a ' Zweckmassigkeitsfrage ' ; and here, again, I am obliged to side with Wundt. — The reader of Ebbinghaus' GrundzUge must, I think, feel that the author is not particularly interested in attention, and has not made the most of the available material. The chapter is, surely, far below the level of those on sensation and memory. I am sorry, nevertheless, to end these Notes with adverse criticism of a work which I greatly admire. INDEX OF NAMES References to the notes begin with "page SSI. Alechsieff, N., 47, 51, 55, 332, 350. Allonnes, G. R. d', 328, 376. Angell, J. R., 44, 187, 221 f., 228, 230, 240, 331, 356, 368 f. Aristotle, 291. Arnold, F., 386. Aubert, H., 27, 327. Auerbach, F., 372. Bain, A., 82, 172, 352, 387. Baldwin, J. M., 184 f., 220 ff., 228, 230, 338, 355, 367. Baumann, J., 387. Bentley, I. M., 217 f., 323, 326 f., 361, 390. Berger, H., 380 f., 384. Berliner, B., 372. Bertels, A., 382. Biervliet, J. J. van, 384. Binet, A., 280. 375, 383 ff. Birch, L. G., 382. Bonser, F. G., 380. Bourdon, B., 81 f., 288, 338, 383 Bradley, F. H., 185, 355. Brahn, M., 167, 350. Braunschweiger, D., 172, 352 f., 359 Breese, B. B., 358. Brentano, F., 33, 328. Bruckner, A., 358. Bumham, W. H., 334, 386. Calkins, M. W., 325, 327. Carpenter, W. B., 387. Cohn, J., 161, 347. Consoni, F., 383. Cope, E. D., 300, 387. Czermak, J. N., 268, 376. Darlington, L., 382. Delbceuf, J. R. L., 201 f., 358. Descartes, R., 82. Dessoir, M., 330, 353, 359. Dietze, G., 233, 237, 368 f. Diirr, E., 359. Dunlap, K., 269, 376. Ebbinghaus, H., 4, 12, 27, 46, 57, 65 f., 68 f., 92, 106, 112, 117, 153 f., 160, 174, 187, 191 f., 196, 213 f., 260 ff., 265, 294 ff., 303, 310 ff., 315, 321, 324 ff., 331 ff., 340, 343 f., 347, 352, 354 ff., 358, 360, 373ff., 381f.,385, 388f., 391. Eckener, H., 269, 382. Engel, G., 326. Exner, S., 193, 273, 357, 380. Fechner, G. T., 13, 20, 106, 136, 173, 200 f., 224, 245, 274, 323, 343, 358, 361, 368, 372, 379, 381 Ferree, C. E., 267 ff., 274, 377 ff. Ferrier, D., 186. 355. Fichte, I. H., 368. Fick, A. E., 274. Fite, W., 44, 331. Fortlage, K., 226, 368. Frey, M. von, 43, 82, 288, 327, 330 f.. 338. Frobes, J., 327. Galloway, C. E., 380. Gamble, E. A. McC, 384. Gardiner, H. N., 385. Gebsattel, E. von, 331. Geiger, M., 254, 258, 262, 332, 374 f. 393 394 INDEX Gcissler, L. R., 267 f,, 275, 376, 382. Gent, W., 167, 350. Geyser, J., 361. Goethe, J. W. von, 135. Golclscheider, A., 17, 88, 333, 339. Golgi, C, 18. Giirber, A., 274. Gurewitsch, A., 147, 346. Hamilton, W., 75, 172, 334, 352, 368. Hamlin, A. J., 361, 382. Hammer, B., 272, 377, 379. Hayes, S. P., 47, 55, 165, 349. Heinrich, W., 269 f., 342, 376 ff. Heller, T., 193, 357. Helmholtz, H. L. F. von, 8, 173, 196, 225, 273, 327, 358, 368, 379. Henri, V., 384. Hcrbart, J. F., 227, 255, 286 f. Hering, E., 7, 20, 324. Hess, C, 377 f. Hillebrand, F., 20, 324 f. Hoffding, H., 153, 347. Hollands, E. H., 38 f., 76, 329, 334. Irons, D., 35. Jackson, G. L., 380. James, W., 14, 34 ff., 153 f., 160, 190, 196, 210, 228, 239 ff., 256, 265, 306 f., 323, 327. 329, 333, 345, 352 f., 356 ff., 369, 374 ff. Janet, P., 280, 384. Jastrow, J., 198, 358. Jodl, F., 153 f., 347. Johnston, C. H., 48 f., 51 ff., 55, 332. Judd, C. H., 187, 356. Kastner, A., 369. Kafka, G., 372. Kant, I., 174 f., 286, 354. Kelchner, M., 92, 340, 349, 384. Kiesow, F., 100, 341. Killen, B., 380. Kohn, H. E., 223, 306 f., 368, 389. Kozaki, N., 361. Kraepelin, E., 332, 382. Kries, J. von, 359. Krueger, F., 339, 360. Kiilpe, O., 20 f., 41 f., 46 f., 57 f., 61 ff., 66, 68 ff., 75, 84 f.. Ill, 117, 129 f., 153, 196, 199,209, 212, 221 ff., 276, 321, 324, 326 f., 330 ff., 340, 347, 352, 354 ff., 358 f., 360 f., 367, 369, 382, 386. Ladd, G. T., 42, 62, 65, 152, 154, 184 f., 330 f., 333, 347, 355. Lagerborg, R., 45, 331, 338, 385. Lange, C., 34 f., 160. Lange, N., 267, 382. Lehmann, A., 66 ff., 76, 153, 203, 267, 273 f., 332 ff., 347, 358, 379 f. Leibniz, G. W. von, 225. Lewes, G. H., 387, 390. Lipps, T., 33,61, 153 f.. 331 ff., 347, 349 f., 357 f., 361. Lotze, R. H., 28, 116 f., 206, 226, 327, 344, 359, 368. McDougall, W., 310, 331, 358, 373, 380, 382, 388. Mach, E., 8, 215 f., 273, 322. Major, D. R., 347. Marshall, H. R., 85, 224 f., 228, 338 f., 368. Marty, A., 386. Meinong, A., 322, 332. Merkel, J.. 289, 385. Memnann, E., 71, 73 ff., 159, 187, 331, 333 f., 349, 352, 381. Meyer, G. H., 342. Meyer, M., 323. Mill, J., 172, 352. Mill, J. S., 352. Mittenzwey, K., 369. Morgan, C. L., 227 f., 230, 240, 368. Moyer, F. E., 382. Muller, G. E., 10, 20 f., 191, 195 f ., 210, 236, 323 ff., 354, 356 ff., 361, 366 f.. 378. Muller, J., 379. Muller, R. F., 333. Miinsterberg, H., 215, 323, 333. 358, 360 f., 369, 382 ff. INDEX 395 Nagel, W., 44, 107 f., 159, 330 f., 343. Oehm, A., 280 f., 384. Orth, J., 44, 46 f., ol, 55, 57, 135, 154, 160, 330 fif., 350, 385, 389. Pace, E. A., 269, 273, 275, 378, 382. Passy, J., 326. Paulhan, F., 375. Peters, W., 383. Pierce, A. H., 44, 331. Pillsbury, W. B., 5, 174, 183, 187, 191, 193, 196, 206, 212. 214, 218, 243, 273, 281, 296, 298, 311 f., 322, 330, 333 ff., 355 ff., 360, 380, 382 f., 388 ff. Pilzecker, A., 356 ff., 366. Plainer, E., 359. Plato, 82. Preyer, W., 273. Quandt, J., 369. Rehmke, J., 56, 153, 329, 331 f., 347. Ribot, T., 101, 184, 342, 355. Royce, J., 147, 149, 347. Sanford, E. C, 379. Saxinger, R., 39, 76, 330, 332, 334. Schultze, F. E. O., 332, 334, 313. Schumann, F., 236, 366 ff., 369, 372 Scripture, E. W., 100, 340. Seashore, C. E., 377 f. Sharp, S. E., 376. Sigwart, C, 352. Slaughter, J. W., 269, 273 f., 377, 379 Smith", W. G., 384. Sollier, P., 328, 332 ff., 338, 340. Solomons, L. M., 180, 354. Squire, C. S., 356. Stein, G., 180, 354. Stem, L. W., 192 f., 357, 369. Stevens, H. C, 252, 346, 374, 380 384 Storring, G. W., 45, 160, 332, 336, 349. Stout, G. S.. 35, 185, 295 f., 298, 328, 355, 386. Stumpf, C, 26, 33, 35 f., 40, 45, 57, 63 ff., 82 ff., 136, 153 f., 160, 185, 193 f., 210, 214 ff., 228, 242, 245, 273, 278, 297, 323 f., 326, 328 ff., 332 f., 338 ff., 355, 357 f., 360, 367, 372, 379, 383. Sully, J., 46, 331, 334. Swift, E. J., 382. Talbot, E. B., 322, 382. Taylor, R. W., 380. Tchisch, W. von, 255 f., 374 f. Tetens, J. N., 385. Thomson, J. J., 385. Torok, L., 324. Toulouse, E., 383. Tsukahara, M., 361. Tucker, A., 226, 354, 368. Urbantschitsch, V., 269. Valle, G. della, 371. Vaschidc, N., 383. Vogt, O., 100, 147, 160, 340, 347, 349. Vogt, R., 382. Ward, J., 75, 223 ff., 299 f., 330, 334 Washburn, M. F., 323, 325. 329, 385. Webor, E. H., 193, 214, 276, 364. Whipple, G. M., 385. Wiersma, E., 267 f., 281, 384. Wirth, W., 332, 369. W^undt, W., 5, 33, 35 f., 38 f., 46 f., 56. 63, 75 f., 105 f., 125 ff., 171, 173 f., 182, 199, 206, 210 f., 213, 225 f., 230 f., 233 ff., 239, 257, 261, 263 ff., 276, 286, 290. 295, 297, 299, 301 ff., 305 ff., 312, 314 ff., 321 ff., 330 ff.. 345 f., 348 ff., 352f., 355ff., 360, 368 f., 371, 373 ff., 382, 385, 387, 389 ff. Ziegler, T., 85, 339. Ziehen, T., 84, 338, 384. Zoneff, P., 71, 73 f., 334, 381. INDEX OF SUBJECTS References to the notes begin with page 321. Accentuation, subjective, 234 f., 239. Accommodation, attentional, 242 ff., 251, 371. Accommodation, peripheral, as condition of clearness, 199, 205, 243 f., 246, 266; as concerned in fluctuation, 268 fiF.; Heinrich's theory of auditory, 377. Action, psychology of, 297 f., 299 f . ; see Reaction. Action, reflex, 299 f., 387; see Reflex arc. Activity, feeling of, 147, 155, 306 ff., 389 f. Adaptation, sensory, 40, 65 ff., 69, 265 f., 367; range of, 265 f. ; ^^sual, in phenomena of fluctuation, 270 f., 378; affective, 40, 65 ff., 100; see Habituation. ^Esthetics, 103, 126 f., 1.30, 343, 348. Affection, definition of, 33; criteria of, 33 ff., 77 f., 127 f., 153, 289 f., 292 f., 335 f. (see Antagonism; Clearness, lack of; Habituation; In- tensity, central ; Non-local- isableness; Subjectivity) ; attributes of, 84, 128; condi- tions of, in Wundt's system, 140 ff. ; manifold qualities of, in Wundt's system, 128, 150 ff., 290 f.; as relation of sensation to consciousness, 131, 154; as undifferentiated conscious process, 291 f. ; and sensation, 33 ff. ; and attention, 294 ff., 296 ff., 334 f.; see Feeling. Affective judgment, analysis of, 163 ff. (pleasantness-un- pleasantness, 165; excite- ment-depression, 165 f.; ten.sion-relaxation, 166). After-image, negative, 69, 271 f . ; of feeling, 290. Algedonic quality, 85; sensa- tion, 338; see Gefuhlsemp- findungen. Alimentary sensations, 18 f., 57 f., 329. Anaesthesia, 116. Analgesia, 115. Antagonism, qualitative, as cri- terion of affection, 56 ff., 77 f., 128, 289, 293, 332 f., 335. Apperception, 75 f., 1.50, 233 f., 2.37 f., 2.55 ff., 263, 295, 297, 303 ff., 312 f.; and cognition, 237; and reproduction, 256, 374 f. ; see Attention. Apperception waves, 265, 381; see Fluctuation. Arousal, feeling of, 147. Associability, as test of atten- tion, 279. Association, general law of, 262; indissoluble, 96; medi- ate, 227; experiments on, 100, 172, 354. Associative consciousness, 264. Attention, problem of, 5, 172 ff., 209 f., 353; popular psy- chology of, 181 f., 301 f., 311; Wundt's analj'sis of, 182 f.; distribution of, 26 (see Levels, Range) ; practi- cal importance of, 181 f. ; conditions of, 294 ff., 312 f. (see Clearness, conditions of); 397 398 INDEX fluctuation of, 199 (see Fluc- tuation); genesis of, 313 f.; theories of, 168, 184 ff., 388 f. (aflfectional, 184; of psychical energy, 184 f . ; cona- tive or motor, 185, 309 ff., 390 ; of reenforcement, 185 f . ; of inhibition, 186); laws of, 207 ff., 251 ff. (clearness as attribute, 211 ff. ; two levels, 220 ff. ; accommodation and inertia, 242 ff.; prior entry, 251 ff. ; limited range, 259 ff. ; temporal instability, 263 ff.; degrees of clearness, 276 ff.); forms of, 311 ff. (active or voluntary, 311 ff., 386, 391; passive or involuntary, 311 ff., 386, 391; secondary passive, 309, 313 f.; reflex or mech- anised, 298 f., 308); and apperception, 150, 168; and intensity of sensation, 212 ff., 219 f., 361 ff., 366 f.; and affection, 294 ff., 296 ff., 334 f.; and will, 297 ff., 306, 387, 391; and interest. 294 ff., 386; in experimental psychology, 172 ff., 353; as sensory clearness, 69, 182 ff. (see Clearness) ; as motor reaction, 309 ff., 390; as total consciousness, 181, 296 ff.,301 f., 313 f.,315; not a sporadic formation, 301 f. (see Inattention) ; directed upon feehng, 69 ff. (see Clearness, lack of). Attitudes, organic (receptive, elaborative, executive), 310. Attributes of sensation, 4, 8 ff., 57 f., 127, 321, 322 f., 337; definition of, 8, 84 f.; in- separable, 8, 23, 85; in- dependentlv variable, 9 f., 16, 20, 2^1 f.; intensive, 10 f., 19 ff., 183; qualitative, 10 ff. ; of the second order, 26 f.; of affection, 84, 127 f.; of image, 337. Auditor}'- sensations, 12 ff., 16, 20, 25 f., 44, 94; pain, 94; intensity and clearness of, 217 f.; fluctuation of, 267, 269 f., 272, 376, 379; see Noise; Tone, sensations of. Ausfragemethode, 50 f., 332. Automatism, in introspection, 179 f. ; normal motor, 180, 375; of attention, 298 f., 308. Biology, 300, 387 f. Blickfeld and Blickpunkt, meta- phor of, 73, 225 f., 234, 236, 368. Blood-pressure (Traube-Hering waves), 273 f. Cessation of stimulus, as condi- tion of clearness, 199 ff. Chroma, 12, 25. Clearness, as intensive attri- bute, 11, 24 ff., 28 f., 173, 183 f., 209 f., 211 ff., 219 f., 285, 361 ff., 372; as starting- point of a psychology of attention, 182 ff., 209 f., 211 ff., 285, 300, 305, 369 f.; phj^siological and psy- chological, 222 f., 370; con- ditions of, 188 ff., 231, 241 f., 243 f., 246, 266, 359 (in- tensity, 188 ff.; quality, 190 f. ; temporal relations of stimulus, 191 ff.; move- ment, 193 ff.; novelty, 195 f.; contents of consciousness, 196 ff. ; peripheral accom- modation, 199; cessation of stimulus, 199 ff.; as im- pressing nervous system, 204 f., 206, 220, 298 f., 300; as objective and subjective, 206; see Distraction); dif- ference of, at two leA^els of consciousness, 229; at same level of consciousness, 229 ff. (lower, 230 ff. ; upper, 233 ff.) ; measurement of degrees of, 276 ff., 309; variation of, with degree of pleasantness- unpleasantness, 302 ; and cog- nition, 238 f., 244. INDEX 399 Clearness, lack of, as char- acter of aflcction, 09 ff., 77 f., 180, 289, 292, 334 ff. Cognition, 238 f., 240 f., 244, 323 f., 369. Colour, sensations of, 11 f., 19, 25, 27; and attention, 26, 190. Colour feelings, 105 ff., 126 f., 134, 148 f., 163 ff. Colour pyramid, 12, 325. Complication, process of, 254 f. Complication experiment, 254 ff. ; inversion of, 252 ff. Concentration, tests of, 279 ff. Consciousness, area of, 220 f.; span of, 233 ff., 369. Contents, conscious, as condi- tion of clearness, 196 ff., 205. Contrast, 53 f., 61, 63, 89, 92 f., 290. Curves of affective judgment, in method of impression, 162 ff., 341. Cutaneous sensations, 16 ff., 25, 87 ff., 93 f., 326; af- fective tone of, 91 ff. ; see Pain, Temperature, Touch. Depression, feeling of, 145 ff., 160, 163 ff., 322; and at- tention, 335. Differences, individual, 172, 190, 216, 228, 352. Differences, maximal, as char- acteristic of sensation, 56, 127. Dimensions of feehng, 128, 131, 133 ff., 141; guaranteed by emotive classification, 135 f. ; by method of expression, 136 ff. ; by time-relations, 138 ff.; by conditions of affection, 140 ff. ; question of spatial, 142 ff.; nomen- clature of, 145 ff. ; number of, 147 ff.; experimental investigation of, 161 ff., 350. Discrimination, sensible, as test of attention, 279, 383; and motor theory of attention. 310; and affective discrimi- nation, 341, Displacement, temporal, 251 f., 255 ff. Disposition, conscious, 149. Disposition, psychophysical, see Predisposition. Distraction, as condition of maximal clearness, 203, 278 f. Distraction, method of, 216 ff., 252, 277 f., 281, 361 ff., 382 f.; inverse method of, 232 f Duration, 11, 22 ff., 28 f., 189 f., 204, 325, 327, 380. Economy, principle of scientific, 86 f., 112, 114. Effort, 203, 278 f., 308 ff., 311 f. ; see Kinsesthetic sen- sations. Strain. Eindringlichkeit, 24, 26 f., 191, 326 f., 356 f. Einsiellung, 243, 373. Element, psychological, 4, 7 f., 37 f.; psychophysical, 7 f.; 'sensational,' 327. Emotion, James-Lange theory of, 34 f., 160; Stumpf's theory of, 35, 63, 86; in Wundt's system, 128 f., 135 f. ; classification of, 136, 139; and feeling, 33 ff., 54, 328. Empfindungsiust, 45, 160 f., 336. Epistemology and psychology, 36 f., 81. Exaltation, feeling of, and at- tention, 335. Excitement, feeling of, 128, 138, 140 f., 145 f., 149 f., 156, 160, 163 ff., 322. Experiment, definition of, 175. Expression, method of, 70, 136 ff., 160, 281 f., 384. Extension, 11, 22 ff., 28 f., 63 f., 189 f., 204, 325 ff.; range of attribute, 25, 326. Eye movement, in fluctuation experiments, 270 f., 378. Faculties, psychology of, 181, 286 f. 400 INDEX Feeling, problem of, 4 f., 33 ff., 125 ff., 285 ff.; intellectu- alistic view of, 285 ff. ; af- fectional view of, 286 f. ; tridimensional theory of, 5, 36, 128 f., 291; sensation- alistic theory of, see Ge- fiihlsempjindungen ; author's theory of, 291 ff., 385, 388; always present in conscious- ness? 116 f.; faculty of, 286; dimensions of, 128, 131, 133 ff., 141 (see Dimen- sions) ; objectification of, 330 f. ; as reaction of ap- perception upon sensory con- tents, 75, 294 f., 303 ff., 306, 388 f.; as condition of attention, 191, 294 ff., 365 f.; as condition of action, 297 f . ; as dependent on quality of sensation, 117 f.; and emo- tion, 33 ff., 54, 328; and organic sensation, 293, 328 f., 385; and will, 306. Feelings form a single system, 128; mixed, 45 ff., 61, 293, 333, 335; relational, 179, 240; coexistence of, in con- sciousness, 39, 53 f. ; fusion of, 38 f., 42, .52, 1.56 f.; in- hibition of, 52 f. ; reenforce- ment of, 52 ; summation of, 52. Fluctuation of attention, 199, 263 ff., 281, 376 ff., 381 f. Focal processes, 26, 227 f., 240 f., 307. Fringe, psychical, 179, 239 ff., 370. Fusion, affective, 38 f., 42, 52, 1.56 f.; sensory, 303; of sensation and affective tone. 97; of taste and smell, 97 ff.; tonal, 97 f., 156. Gefiihlsempfindunge?!, 64, 81 ff., 286, 288', 290, 294. Gestaltqualitat, 339. Grey, MuUer's endogenous, 21 f. Habit, introspective, 179, 197; scientific, 198; implies fore- gone attention, 300, 357, 375. Habitual, indispensableness of the, 67 f., 203. Habituation, effect of, 192; as characteristic of feeling, 65 ff., 77, 333 f. Hallucination, affective, 64, 102 f., 104, 108; of pain, 102 f., 104; in comph cation experi- ment, 256, 375. Hue, 12, 25. Hunger, 18 f., 57, 59, 329. Hypnosis, 100, 371. Ideas, intensive, spatial and temporal, 142. Idiocy, 371. Illusions, optical, 373, 382. Image, 3, 61 f., 102, 104, 342 f.; affective, 101 ff., 110, 290, 341 f. ; of weak sensations, 106 f.; of momentary sensa- tions, 102, 342; fluctuation of, 382; and sensation, 63 f., 321, 337. Imagery, organic, 102, 343. Impression, method of, 50 f., 105, 148, 152, 161 ff., 350; twofold control by, 162 f. Inattention, 301 f., 381, 383; field of, 224 f. Independence, movement for affective, 286 ff. Indifference, 67, 85 f., 116 f., 302. Inertia of attention, 242, 245 f., 251, 371 f. Inhibition, feeling of, 128, 138, 145 f. ; in theory of attention, 305, 389; of feelings, 52 f. Instinct, 191, 298. 308. Intellectualism, 286 f. Intensitv, central, as character- istic of feeling, 61 ff., 333. Intensity, definition of, 10, 24; independent variability of, 20; of visual sensations, 20 ff. ; a 'quahtative' attri- bute? 28; doctrine of sensi- ble, 173; as condition of feeling, 131, 141; as condi- INDEX 401 tion of clearaess, 188 flF., 204, 356; in classification of emotions, 136; and clear- ness, 211 ff., 218 ff., 361 ff. Interest and attention, 294 ff., 386. Interests, permanent, 197 f. Intervals, repeated, affective tone of, 163 ff. ; see Purity. Introspection, 63, 77, 132, 144 ff., 151, 162 f., 165 f., 197, 225, 235 f., 254 f., 262, 264 f., 277, 281, 293, 306 ff., 333, 336, 354, 362, 369, 381 ff.; compared with inspection, 175 ff., 355; interpretation of, 332, 360 f. Itch, 17, 90 f., 324. Kinaesthetic sensations, 18, 25, 58 f., 309; see Effort, Strain, Touch. Lability of attention, 242, 264, 276. Law of continuity, 224 f.; of indispensableness of the habitual, 67 f., 203; of reduction and expansion of conscious processes, 34, 313 f., 390; of tedimn, 68; Weber's, 214, 218, 276, 364. Laws of attention, 211 ff., 251 ff. ; see Attention. Levels of consciousness, 220 ff., 301 f., 305, 370 f. Limen, temporal, 246, 251 f. Local sign, 43 f., 55. Localisation, as characteristic of sensations, 43 ff., 303. Marginal processes, 26, 227 f., 231, 240 f., 309. Measurement of attention, 276 ff. ; by introspective dis- tinction of degrees of clear- ness, 277 f . ; by measure- ment of effort, 278 f., 309; see Tests. Memory, 172, 365 f.; see Image. 2d Movement, as condition of clearness, 193 ff., 205; sensa- tions of, 193. MV, as measure of attention, 280 f., 384. Nausea, 18 f. Need, organic, 68, 203. Noise, sensation of, 16, 20, 270, 379. Non-localisableness, as char- acteristic of feeling, 43 ff., 77 f., 331 f., 335 f.; external, 43 ff.; internal, 45 ff. Novelty, as condition of clear- ness, 195, 205; as non- associatedness, 195, 357 f. Nuancirung of pleasantness- unpleasantness, 160. Objectivity, as characteristic of sensation, 36 ff.; of feel- ing, 330 f. Observation, definition of, 175. Odours, as distracting stimuli, 278, 361; see Smell. Opposites, movement between, see Antagonism. Organic sensations, 18 f., 25 f., 38 f., 44, 54, 57, 93 ff., 158 ff., 349; locahsation of, 44, 330 f. ; affective tone of, 126; and attention 190; and the Wundtian dimensions, 160 f., 163, 301 f.; and feeling, 293, 328 f., 385. Organs, peripheral, of feeling, 292. Pain, sensation of, 17 ff., 25 f., 43, 87 ff., 92 ff., 292, 324; quality of, 17 f., 88 f.; Ein- dringlichkeit of, 26, 190,327; affective tone of, 89, 92 f.; as unpleasantness, 82 f., 87 ff., 334; from intensive stim- ulation of pressure, tempera- ture, sight, and hearing, 94. Paired comparisons, method of, 161 f. Parallelism, psychophysical, 225, 387. 402 INDEX Partial feelings, 155 ff.; tones, 196 f. Perseverationstendenz, 246, 373. Physiologj^ 4, 206, 256, 310 f.; and psychology. 286. Pitch, tonal, 12 ff., 26 f. Pleasantness, qualitative dif- ferentiation of, 160 f., 293, 336 f., 349. Pleasantness-unpleasantness, 33, 125 ff., 290 ff., 298, 308 f., 329, 341; see Affection, Di- mensions, Feeling. Pleasure, sensation of, 81 f., 83, 93, 96; cutaneous, 93; organic, 83, 93, 96; due to intensive stimulation, 95 f.; as absence of pain, 82. Practice, 192, 252, 277, 375. Predisposition, psychophysical, 196, 199, 202, 205 f., 209, 243, 246, 2.52, 266; and fluctuation of attention, 275. Pressure, sensations of, 17 ff., 25; pain, 94. Prick, sensation of, 17, 88, 90 f., 108; image of, 102, 342. Prior entry, law of, 251 ff. Psychology, descriptive, 83, 99, 111, 113, 266, 312, 314 f.; experimental, 171 f., 206, 210, 266 f., 274, 286, 288 f., 316 f.; explanatory, 288, 313, 315; genetic, 113, 11.5, 118 ff., 126, 206, 381, 387 (of feeling, 291 ff. ; of refiex action, 299 f., 387; of attention, 313 f.); physiological, 288; syste- matic, 3, 73, 129 ff., 142, 154, 158f., 167f., 282, 296ff., 314, 353; of feeling, 4 f., 285 ff., 291 ff. ; of attention, 5, 313 f. ; and epistemology, 36 f., 81. Psychophysics, 7 f., 21 f., 90, 97 ff., 101, 112 ff., 118, 189, 206, 289, 336, 354. Pulse, and fluctuation of at- tention, 273. Purity, feeling of tonal, 119, 339, 360. Puzzle picture, observation of, 228 f., 370 f. Qualities of affection, inWundt's theory, 150 ff.; see Affec- tion, Pleasantness. Quality, definition of, 10, 24, 28 ; of visual sensations, 1 1 f . ; of auditory sensations, 12 ff. ; of pressure, 17; of pain, 17 f., 88 f. (see Pain); of kinsesthetic sensations, 18 f. ; of alimentary sensations, 18 f. ; of smell, 16, 19; of taste, 16, 19; algedonic, 85; as criterion of sensation, 27 ff. ; as condition of feeling, 131, 141; as condition of clear- ness, 190 f., 204; in classi- fication of emotions, 136; doctrine of sensible, 173. Quiescence, feeling of, 147, 150. Range of attention, 259 ff., 370, 376; as test of degree of concentration, 279. Reaction, simple, 242, 252, 280, 308, 371. Reflex arc, 310, 390. Reizrnethode, 160, 350. Relaxation, feehng of, 128, 138, 140 f., 145 ff., 160, 163 f., 166, 322; and at- tention, 335. Repetition, as condition of clearness, 191 f., 204. Respiration, and fluctuation of attention, 273, 379. Restlessness, feeling of, 147. Retrospection, as psychologi- cal method, 178 f. Rise of sensations, 243, 245, 251 f., 372. Sensation, attributes of, 4, 8 ff. (see Attributes) ; character of, as elementary process, 4, 14 f., 323 f.; central concomitant, 96, 112 f., 115 ff.; centrally excited, 333; criterion of, 27 ff.; single, coextensive with conscious- ness, 55; in psychology and psychophysics, 7 f. ; and image, 63 f., 321, 337. INDEX 403 Sensationalism, 286, 288; see Gefuhlsempfindungcn. Sensations, number of discrimi- nable, 27; focal and mar- ginal, 26, 227 f., 231, 240 f., 307, 309; alimentary, etc., see Alimentary sensations, etc. Sense-feelings, 83, 131. Sensitivity, as test of con- centration, 279, 383. Smell, quality of, 16, 19; pene- tratingness of, 26, 190, 326; localisation of, 43 f. ; af- fective tone of, 126, 331; as distraction, 278, 361. Space, psychology of, 209. Stimmungslust, 45, 160 f., 336, 349. Sting, sensation of, 17, 88, 90 f., 108; image of, 102, 342. Strain, sensation of, 18, 278, 307, 312; feeling of, 322, 335 (see Tension). Stream of thought, 228, 376. Streaming phenomenon, 271. Strcben-Widerstrehen, 147. Strcbungsgefilhl, 147. Subconscious, the, 220, 224, 226 f., 230. Subjectivity, as character of feeling, 36 ff., 77 f., 329 ff., 335 f. (as tendency to fusion, 38 ff. ; as individual variabil- ity of experience, 40; as inability to stand alone in consciousness, 41 f., 61, 100; as textural flimsiness, 43, 336). Suddenness, as condition of clearness, 191 ff., 195, 204 f. Summation of stimuli, 192; of feelings, 52. Tachistoscopic experiments, 231, 237 f., 259 ff. Taste, quality of, 16, 19; im- portunity of, 26, 190. Taste feelings, 100, 126, 331, 336, 341. Tastzuckungen, 268. Teleology, 120 f. Temperature, sensations of, 16, 25, 57 f.; pain, 94. Temporal conditions of clear- ness, 191 ff. Temporal course of emotions, 136; of mental processes, 138; as condition of feeling, 141. Tension, feeling of, 128, 138, 140 f., 145 ff., 160, 163 f., 166. Terminology of Wundt's theory of feeling, 145 f.; of feeling, 385. Tests of concentration, 279 ff. Theories, scientific, 48, 198, 293 f., 385 f. Tickling, 17, 81. Time, psychology of, 209; as condition of feeling, 149 f.; as condition of clearness, 189. Tingling, 95. Tint, 12, 25. Tonal change, 194; fusion, see Fusion. Tone, affective, 41, 83 ff., 125 ff., 131, 286 f., 329; as concomitant sensation, 96, 112 f., 115 ff.; as pleas- antness-unpleasantness, 125 ff. Tone, organic, 293, 300. Tone, sensations of, 12 ff., 25 f., 119 ff.; localisation of, 44; affective tone of, 105 ff., 120 f., 126 f., 132, 149, 156 ff., 163 ff., 343 f., 348; and attention, 215 f. Tone-colour, 26, 326. Tones, partial, 196 f. Total feeling, 39, 147, 151, 155 ff., 346. Touch, 25, 326; affective tone of, 126; fluctuation of, 207 ff., 272, 376; direct and in- direct, 193 f. Tranquillisation, feeling of, 128, 140 f., 145 f., 156. Unconscious, the, 220 f., 226 f. Unnoticed stimuli, 199 ff. Unpleasantness, see Affection, Pleasantness-unpleasantness. 404 iNDiEX Vision, stereoscopic, 196 f. Visual sensations, 11 ff., 20 fi., 24 ff., 326; pain, 94; fluc- tuation of, 267, 269 ff., 378. Volume of tones, 13 ff., 26 f., 324, 327. Voluntarism, psychological, 387. Waves, apperception, 265, 381; Traube-Hering, 273 f. Weakening of sensation by attention, 366 f. Will, and attention and action, 297 ff., 306, 387, 391; and feeling, 306. ^^ University of Caiifornia SOUTHERN REGIONAL LIBRARY FACILITY 405 Hilgard Avenue, Los Angeles, CA 90024-1388 Return this material to the library from which it was borrowed. ^.-.. M ^*- Q|^JAN23 1^«5 OCT 1 1 Z004 ncr^ r\ ^mf ■fe^ iJA A 000 943 579 3 ■.^ -^/A c»