If we agree to call the motor centre for speech {inp of Fig. XVII. b, p. 413) the 'intrinsic ' seat of stimulation to the organs of speech, and, on the other hand, to call the other centres pointed out ' extrinsic,' the question now current runs: Are these extrinsic centres capable, each for itself, of arousing the speech centre; or does one of them, the centre for sensations and memories of actual move- ment, the ' kinaesthetic ' word centre (;;/<:, of the same figure), always stand between the motor seat and the other sensory centres.?
Or, put psychologically, do we, when we remember words and speak them, always recall them in terms of the sensations of movement involved in speaking or writing them; or is it possible to speak simply from remembering the visual form of the word, or its sound t Is the kinaes- thetic centre, with the memories of movement to which its processes correspond, intrinsic or extrinsic } The view that verbal memories are always motor, or kinaesthetic, is associated with the name of Strieker.^ Recent results have refuted Strieker. A variety of facts have been adduced to show that the function of speech is not dependent in all cases upon the possibility of reinstating 1 Strieker, Ueber die Bewegiingsvorstelliingen, Ueber die Association der Vorsiellungen, Ueber die Sprachvorstellungen, Langage et Ahisiqiie. See also (j. E. Miiller, Griindlegttiig der Psychophysik.
434 ^^^^ Mechanism of Revival, motor experiences; although in some cases it is, for patients are reported who could not speak unless they first traced the words with hand or pen.^ Many of these facts are already common property; but a few of the latest points on this side of the discussion are these: (i) Cases are cited of verbal hallucination, in which the patient hears two or more voices, one of which he takes to be his own, the other that of some one else; only the former can be accounted for as due to the incipient stimulation of his own speech centres, the other is probably auditory.^ This interpretation is supported by the interesting fact, estab- lished by Pierre Janet, that some patients can themselves speak during their verbal hallucinations, while others can- not. Again, only of the latter class must we hold that the motor memories are necessary to speech.^ Indeed, there is a characteristic difference between the two classes, — a difference first pointed out, it seems, by Baillarger — i.e.^ with those patients who are able to speak without interrupting the voice which they hear, we have a halluci- nation of objective speech: they hear what they think is a real voice outside them. While the other class have a hallucination of internal speech. They declare that there is some one inside them, speaking to them. Seglas holds, with evident truth, that these latter hallucinations are *psycho-motor ' * in their seat, while the 'objective' kind 1 See Sommer's report on the so-called Grashey case — a patient named Voit — in Zeitsch. fiir Psychologie, II., heft 3, p. 158, and the citations of Pick, same journal, III., heft I, p. 50.
2 See case of Charcot quoted by Ballet, Le langage interietir, p. 64, also cases in Seglas, Les troubles du langage chez les alienes, p. 1 26.
3 Cf. Revue Philosophique, November, 1892, p. 520, and Seglas, loc. cit.^ p. 117 and p. 145. A case is reported of a patient who could stop his internal voice by holding his breath (^Annates Psychol., January, 1893, p. 103).
* Seglas, loc. cit., p. 147; Janet, loc. ciL, who advocates the expression Internal Speech. 435 are auditory. (2) There are cases of motor aphasia due to impairment of hearing, the motor centres being intact, i.e., cases of auditory verbal amnesic aphasia.^ (3) We recognize and understand words which we are unable to pronounce, and which we have never written; this recog- nition must be by aid of visual or auditory images. The part played by the visual and motor memories respectively, in my own case, is seen in the fact that when I wish to speak in any language but English, the German words come first into my mind; but when I sit down to write in a foreign language, French words invariably present them- selves. This means that my German is speech-motor and auditory, having been learned conversationally in Germany, while the French, which was acquired in school by reading and exercise-writing, is visual and hand-motor.^ It is interesting also to note the joyous recognition which young children show, when they speak a new vowel or consonant sound correctly. The memory of the correct sound can- not, in this case evidently, be from the motor centres.^ (4) There is evidence of direct functional connection be- * kinaesthetic verbal ' instead of * psycho-motor,' as applying to this hallu- cination of internal speech.
1 See cases collected by Ballet, loc. cit., pp. 91-92; also Bastian's case, Brain as Organ of Mind, p. 642; cf. also Paulhan, Revue Philosophique, " A similar case, apart from details, is reported by Ballet, loc. cit., p. 62.
^ At the risk of too much personality (of which, however, the literature of this topic is necessarily full), I may quote the following about my two-year-old child H., written in a letter by her aunt, who was far from intending it as a psychological observation or for publication: " She rejoices greatly when she succeeds in sounding a new letter. The other day she achieved /, and went about telling everybody, * Baby can say sleep and slipper.' This morning I am informed that she can say 'save' and 'give ' (letter v). She notices at once herself, when she can pronounce the word as the rest of us do — no one tells her."
43^ The Mechanism of Revival.
tvveen the visual and auditory seats and the centre of motor discharge. Here I may best give the words of Janet, who writes in view of the pathological evidence: "This hypothesis is confirmed by investigations on anaes- thetic hysterics. In my opinion, it is impossible to explain the fact that these persons preserve their power of move- ment intact, in spite of the absolute loss of kinaesthetic sensations and images, unless we admit that movement may be directly stimulated by visual and auditory pictures. There are individuals with whom the auditory image of a word suffices for its pronunciation." ^ (5) The law of 'dynamogenesis,' in accordance with which every sensory stimulation tends to bring about a motor discharge, indi- cates such a direct connection in cases of closely associated function. Fere demonstrates that speaking makes the hand-grasp stronger, that seeing colours and hearing sounds influence the motor centres; so it is altogether probable that stimulations of sight and hearing react directly to stimulate the motor speech centres.^ (6) Cases may be cited of direct antagonism between memories of words and the sensations produced by the speech movements which they stimulate. The pathological state called para- phasia^ is duplicated sometimes temporarily in cases of severe head-ache; one intends to mention one object (chair) and really speaks another (spoon), without detecting 1 Pierre Janet, Automatisme Psychologique, p. 60. The common cases of patients who can copy, when they cannot initiate writing and speech, are in evidence.
■■2 Fere cites his results in support of Strieker's contention; see his Sensa- tion et Alouvetnent. He fails, however, to distinguish between the direct motor effect of a sensation, and the round-about motor effect — i.e., through the kinaesthetic centre, or via the motor correlations which the attention re- quires— which latter is required by Strieker's view.
2 Cf. Bastian's cases of * incoordinate amnesia,' Brain as Organ of Mind, Internal Speech. 437 the mistake. I have myself had this experience; being quite unable to name correctly an object seen, until some one else has spoken the word with emphasis — yet all the while allowing my own incorrect word to pass, and feeling astonishment that others have not understood my meaning. Similar are those cases in which patients take their own words for those of some one else, declaring, when questioned, that they themselves did not speak them.^ Reflection leads us to the view that in these cases there is a direct flow from the auditory or visual centre to the motor speech centre, the kinaesthetic speech centre being, perhaps, temporarily inhibited. The same kind of antagonism is also seen, from the other side, when there is ' exaltation ' of the kinaesthetic centre, or what is called uncontrollable Verbal impulse.' The patient speaks certain words or phrases in spite of himself — against his utmost effort to speak something else.^ This conception of the case — not to dwell upon other points of evidence^ — seems to harmonize well with the 1 See Seglas' very interesting cases, loc. cit., pp. 150 f.
2 See Seglas on * hysterical mutism,' loc. ciL, pp. 97 f. In dreams this is probably the case: the kinesthetic centres are no longer inhibited, and we talk meaningless sounds, which in our dream consciousness are interpreted as rational discourse. In view of all such cases of antagonism, I suggested in an earlier statement of the main considerations on this point {Philos. Reviezv^ II., 1893, P- 389)5 that a distinction was legitimate between psychic and cortical dumbness, corresponding to the current distinction on the sensory side. Just as there is a distinction between being unable to hear words (cortical deafness), and being unable to understand the meanitigs of words we hear (psychic deaf- ness), so there is a distinction, shown pathologically, between being unable to speak words, and being unable to speak the words we mean. Put in different terminology, the former case would be due to a lesion of the motor elements at the ' second level,' and the latter case to a lesion of the motor connections between the second and the cortical or ' third level.' Compare the allusions made to these differences above. Chap. XIII., § 3, p. 408.
^ For instance, cf. Stumpf, Tonpsychologie, I., pp. 160 ff. Further evidence accrues, also, from the consideration of tune memories, which seem to be 438 The Mrxhanism of Revival.
doctrine of nervous function now becoming more and more current. According to this doctrine, the brain is a series of centres of relatively stable dynamic tension; the various associative connections among these centres are paths of less and morcy rather than of least and most, resistance; that the range of alternative adjustments is excessively wide; and, consequently, that any individual has his 'per- sonal equation ' in all functions as complex as that of speech. One man is a ' motor,' another a ' visual,' a third an * auditive,' according as one or another of the extrinsic sources of stimulation suffices to release the necessary energy into his motor speech centre. No one doubts Strieker, therefore, when he says that he remembers words only by means of sensations of incipient movement; but for the same reason we cannot dispute the claim of Stumpf, and Wernicke, and Kussmaul, and Lichtheim, that audi- tory and visual images may, in other cases, play an equally leading role.
§ 2. Internal Song: Hoiv do We think of Times?
The question of 'internal song' is a new one. What do we mean when we say that a ' tune is running in our head ' } What sort of images are really in consciousness then } The factors involved are evidently less complex than those already shown to be involved in speech memory, in the discussion in the preceding paragraph,^ at the same time that the entire phenomenon is more obscure. Eviindependent, in many adults, and generally in children, of the singing or playing of the tunes. Cf. above, Chap. VI., § 5, and the next section of this chapter.
1 See also Chap. XV., § 3.
Internal Song. 439 dence goes to show that the internal tune is almost entirely auditory: that is, that the auditory centre is intrinsic to musical reproduction.
An adequate discussion of the nature of tune reproduc- tion should provide a theory of tune perception which takes account of three factors — pitch, time or rhythm, timbre — and possibly of a fourth character, ordinarily designated by the phrase ' musical expression ' or, more properly, emotional tone.^ There are certain interesting points of relationship between the process of internal speech and that of ' inter- nal' or remembered music. For example, many persons find internal tunes generally fuller, more real, and some- times only tunes at all when vocal movements are involved; either, that is, when they remember the appropriate words, when they have sung the words to the tune, or when they have hummed the refrain aloud. Here there is clearly a motor type of music performers. But this motor require- ment is extremely variable. In some cases the tune must be associated with a particular instrument, and this is done only by the reproduction of the proper sensations in the finger tips, lips, etc., used in playing that instrument. On 1 There is not a great deal of literature on this topic; see the following titles: Egger, La parole interieure; Strieker, Langage et Ditisiqiie; Stumpf, Tonpsychologie, I., pp. 135 ff.; Wallaschek, Vierteljahrschrift fur Mtisik- wissenschaft, 189 1, heft I, and die Bedeutung der Aphasia fiir die Mtisik- vorstellung, Zeitsch fiir Psychol.^ VI., heft I, and his review of my theory in the same journal, VII., heft i; Wallaschek has a popular article also in the Contemporary Review, September, 1894; Lotze, Medicinische Psychologic^ p. 480; G. E. Miiller, Grundlegung der Psychophysik, p. 288; v. Franckl- Hochwart, Ueber den Verlust des musikalischen Ausdrucksvermogens in Deutsche Zeitschrift fiir Nerveiiheilkunde, 189 1, I., pp. 283-29 1; Oppenheim, Charite Annalen, XIII., 1888, 345-383; besides the voluminous literature of aphasia. An interesting late article, full of bibliographical references, is by Brazier, Revue Philosophiqtie, October, 1892, p. 337.
440 TJic Mechanism of Revival.
the other hand, there are facts which show that the motor type is only a type, and that even in these cases auditory tune memories are necessary. Musical recognition in childhood often precedes verbal recognition. Musical expression usually precedes verbal expression, both when there is a clearly inherited musical tendency,^ and in ordinary imitative reactions.^ In cases of ' absolute hear- ing,' discussed below, we have apparently recognition of pitch without any motor speech or song images. Further, there is the critical fact that motor aphasia, and even verbal deafness, may exist with no impairment of the musical faculty — no aimcsia, as defects of musical faculty are called by Brazier. This is true both for musical recog- nition (case of Wernicke), and for musical expression.^ Cases show, however, that the latter, musical expression, is never lost, without involving speech; although musical recognition seems sometimes, as in Carpenter's case and in Brazier's cases of musical amnesia, to be lost without impairing speech.* The conclusion that musical repro- duction is auditory is supported also by such facts as the following: that we often recognize an air after hearing it once, even when we have never tried to sing it, and could not if we tried; that in singing or humming a tune, we know that we are wrong even when we are unable to correct it; tune hallucinations are reported without words 1 Interesting cases are cited by Ballet, loc. cit.^ p. 24.
^ My child E. imitated a run of three notes, vocally, before she showed any verbal imitations.
4 Wallaschek, Zt. f. Psych., VII., March, 1893, P- 671, in criticising this statement of mine, cites cases of musical inability through stage-fright, while speech remains, as possible exceptions. I think, however, that stage-fright is such an emotional and interested thing that the inability is not really musical at all, but is rather due to general nervous inhibition.
Internal Song. 441 or vocal quality, and illusions of tunes may be started by accidental sounds; ^ many persons are able to remember and recall musical chords and combinations which it is impossible for the human voice to reproduce, i.e., we can mentally depict harmony; further, there are cases of per- sons who can recognize the pitch of tones from instru- ments, but not that of the tones of their own voice/'^ It seems clear, indeed, on the surface, that of the elements distinguished above as essential to musical reproduction — pitch, rhythm, timbre, and emotional tone — the most essential, pitch, finds no adequate basis in motor speech or song memories. The range of intonation in speaking and singing is too narrow to supply the material for musical reproduction, although there are, no doubt, indi- viduals whose musical capacity — especially of expression — is confined to these limits.
It is probable, accordingly, that there is a brain-centre for tune memories — a centre whose impairment produces so-called notal aimisia — that this centre is a part, in func- tion, at least, if not anatomically, of the auditory centre, and that cases will occur of partial amusia in different per- sons, due to the degree in which this function involves others.^ This general conclusion is confirmed, I think, 1 Ordinary internal tunes are usually stimulated in this way, as I have said above, Chap. VI., § 5.
2 Cases of v. Kries cited below.
3 For example, musical deafness without verbal deafness; case of Grant Allen in Mind, III., p. 157, and that of Brazier, he. cit., p. 359. Bastian, loc. cit., p. 664, quotes a case from Lasegue of an aphasic musician, who could write nothing but passages of music which he \\z.A just heard. A recent case of Pick's {Arch. fUr Psych., 1892, p. 910) seems at first sight to give trouble, i.e.., a case of loss of musical recognitioii with no impairment of musical expression. Yet Pick's location of the lesion as subcortical sufficiently accords with the view in my text. The seat of auditory attention was not injured. Cf. note on Pick's position, and the theory of ' muscular control,' below 442 The Mechanis7n of Revival.
by what follows on pitch memory, the only one of the four elements of musical reproduction which is in order here.
§ 3. PitcJi Recognition.
The recognition of the pitch of notes gives two cases apparently distinct from each other, i.e., 'relative' and ' absolute ' pitch recognition. In relative recognition the musical interval seems to supply the real locics of the recog- nition. Given the initial note and the proper rhythm — and the rest of the tune comes up by reason of the asso- ciated tone intervals, note by note. It is the case of objective recognition by assimilation of content, as already described. 1 Comparatively few persons lack the ability to carry through a familiar tune mentally. Absolute recog- nition, on the other hand, is a different accomplishment; even among competent musicians it is often ^ conspicu- ously absent. It is the power of reproducing a note of any desired pitch absolutely from memory.
The auditory character of all relative pitch recognition is shown by the following facts — in addition to the general considerations already adduced: (i) Brazier^ cites cases of aphasic patients who could speak words only by singing them: that is, they must first recognize an air, and then arouse the motor speech function from that cue. The motor centre not being available in these cases, it is diffi- cult to see on what but auditory grounds the tune recogni- tion could proceed. It often occurs, in my own case, that I cannot recall the words of a song until I get the tune 2 In the case of some of those who carry tuning-forks in their pockets.
Pitch Rccog-iiitioii.
cb started. Another case of this kind is cited immediately below. (2) I find it possible, with Paulhan,i to think dif- ferent notes very clearly while the vocal organs are held rio;id. I am able to think one note while I am utterins: aloud a long-drawn-out vocal sound, say a, in a different pitch. And lest it may be said that it is the overtones which are heard internally in this case, I may add, that I am able with the greatest ease to hold aloud an a sound at c^ sa}^, and at the same time to cause a whole tune — say Yankee-doodle — to run its course 'in my ear.' Strieker's inability to think one consonant while speaking another is due, probably, to the fact that, in uttering labials, etc., pronounced and explosive muscular combinations are necessary, and that they have no clear auditory character, being usually merged in accompanying vowel sounds. (3) My internal tunes have very decided pitch — deter- mined upon an instrument in a number of cases. Yet, as I have said above,^ it is not always the normal pitch of the tune as written and learned, nor is it constant for recur- rences of the same tune.
In explaining pitch recognition the question of relative pitch comes first. The very fact that it is relative, means that it may be brought under the law of objective conscious recognition in general. If recognition be due to assim- ilation, relationship, 'fringe,' in the representation recog- nized, and vary with the degree of this associative or apperceptive element, then recognition of each note would occur, like the recognition of any other presented content, according as it have or have not a train or fringe of asso- ciated elements. A tune is then recognized, because it is such a train. The degree of precision in its recognition 444 '^^^^ Mechanism of Revival.
depends upon the fineness of discrimination at the original hearino: of it. So also the fact that notes are better reco^:- nized after the musical notation has been learned, simply means that additional elements are brought into the com- plex by the notation — elements which support the claim of the whole. With persons of the motor type, further, the motor speech and song images are prominent in this complex, and so essential, in some cases, that recognition does not occur without them. It seems likely, therefore, that if we grant differences of pitch in tone sensations, the recosrnition of the associated trains which we call 'tunes' is but an instance of a broader mental phenomenon.
Absolute recognition, on the other hand, or ' absolute hearing,' as it is called, presents anomalies which make it difficult to explain it as an ordinary case of recognition by presented association. Either we must find elements of complexity in such tones or confess that here is an excep- tion to the accepted theory. I have already given the gen- eral principles by which this case is to be explained: but it may be well to apply them now to a concrete instance.^ The question which may be asked, is this: Can any one identify a note of any pitch simply and only from the tone-quality of the note itself t One of the latest contributions to this question is from V. Kries,^ who is himself a musician. He possesses the so-called absolute hearing. He also publishes details supplied from other similar cases. He argues that the ability to identify a single isolated note cannot be due to musical practice, i.e., cannot be a refinement of in- 2 Das absohite Gehor, in Zeitschrift fiir Psychologic und Physiologie der Sin- Pitch Recognition, 445 terval recognition,^ because (i) he has had this power from early boyhood, as also have others whom he cites; (2) some of the most celebrated musicians have not been able to acquire it at all, although their sense of inter- val became wonderfully acute; and (3) the power in him- self and others varies with the instrument which sounds the note, and is not best with the instruments used most. He recognizes notes from the piano best, from string and wind instruments, especially the violin, but not those from tuning-forks, or steam and other whistles, or notes sung or whistled with the lips — a state of things shown with some variations also in several of his correspondents. Now the violin is with v. Kries a late accomplishment, while he has, of course, been hearing singing all his life, accompanying singers on the piano from his twelfth year, and whistling habitually. Indeed, these last facts — showing the influ- ence of timbre on pitch recognition — lead him to deny that there are any revived images of any kind belonging intrinsically to musical recognition. He finds it to be a case of the ' association by naming ' as established by Leh- mann; that is, v. Kries was not able to recognize notes until after, in boyhood, he had learned their names and written signs. The case is analogous, therefore, he holds, to the recognitions which Lehmann found to follow from the simple lettering and naming of shades of wool not before separately recognized.
This conclusion of v. Kries is lame, I think. It does not account for the differences due to timbre mentioned above; for the notation is the same practically for all the instruments and for the voice, v. Kries admits this, and says it remains for the future to provide a theory of 1 So Stumpf, loc. cit., I., p. 2S0.
44^ The Mechanism of Revival.
this influence due to timbre — leaning, however, as he does, to the overtone theory. Further, he agrees with other observers in finding that chords are better recognized than single notes; this would indicate that recognition is lue in some way to the complexity and variety of the tone content, rather than to the accident of naming. It is pos- sible, perhaps, to give due weight to the influence of the name association in a theory which does more justice to the essential facts. This and other cases of the recognition of apparently isolated sense qualities can be brought, I think, under the law of * sensori-motor association' already formulated, according to which the recognition is due simply to the modification of the a element in my formula ^ of attention, i.e., to the relative ease of adjustment of the attention, to one particular tone-pitch as such.