These papers are the result of an attempt to condense into comparatively small space the fruit of some 30 years of study and speculation. The problem with which they deal is for the most part ignored—in some quarters, because the fact with which they deal is taken for granted; in some, because that fact is denied. It is ordinarily taken for granted by those who, deeply moved by music, as they are by appeals from many other sources, ask no more than the gratification they have thus received. It is denied by those who, having inquired into the actuality of the idea supposedly expressed by music, have found that idea indefinable and have thus come to believe it nonexistent.

They who believe that music expresses nothing are few in number. This, considering the vast number opposed to them, should be, to the minority, a disturbing fact. It is evident, indeed, that for that minority the history of music as it is written can have little real meaning; for the true history of our art is the record of its adaptation to the larger, constantly changing field of thought of all humanity. But in so far as the thought which music expresses fails to be understood, its history remains a defective chronicle.

My undertaking is thus highly presumptuous. It assumes a defect—of course, unsuspected—in the work of men far more learned than I. But what I here propose—in spite of its length, too briefly for the true dimension of the problem—is nevertheless a method of observation by which I can work (and which has proved, to a good many of my students, likewise workable) toward a more certain understanding of what music means. This problem, once it is pondered, can hardly appear less than vital.

The Nature and the Process of Musical Expression

A good many years ago, I ran across a definition of physical man, so ingeniously simplified that it has often amused me since. “Man,” it said, “consists of some twelve to fifteen pounds of solid matter mixed up in six pails of water.” That gross understatement is true, as far as it goes; but its chief value is to make us think of all that man is besides. Religion, philosophy, and the sciences have all striven for definitions of that more complete man, but not with complete success. Psychology—which word suggests that science deals with what, to the ancient Greeks, was indisputably the soul—has attempted to describe the behavior of this mixture of solids and water as that of a mechanism designed for reaction to stimuli. Rigidly pursued, this mechanistic psychology presents the human creature as the mere sport of impressions, present and past, that impinge upon him—an automaton in the clutches of a complex which he fondly believes to be his will. And its ultimate product is perhaps the strangest paradox in the history of thought—a thinker who thinks he cannot think. Such a creature could hardly obey the old Greek maxim “Know thyself,” for he would have no self to know.

Yet you and I and, I suspect, our mechanistic psychologist also, have somewhere within that mixture of solids and water which we and others identify as ourselves a spiritual self in which we ineradicably believe. Like that physical mixture, we also find the spiritual self in constant need of nourishment. Science has studied pretty competently the needs of the physical body, and in various intellectual ways the needs of the spirit have also been explored. Two world wars in three decades, however, suggest that these explorations have left a good many essentials undiscovered. Religion is one general type of such nourishment: the arts are another; and humanity has cherished both these contributions accordingly. Aesthetics, in the somewhat narrow view I am now taking, might be called a sort of spiritual dietetics. At any rate, its business is to assess the value of the arts as spiritual foods.

Criticism, in the best sense, is an assessment of artistic values—a practical aspect of aesthetics. It has developed a very elaborate process—so elaborate that its original and proper purpose—the assessment of values in terms of human needs, is often obscured by the working of its process. Musicology has contributed vastly, of late, to the process itself and to that kind of objectivity in judgment which keeps criticism from turning into the mere operation of a formula. But while I hope I understand enough of the achievements of musicology not to undervalue them, it seems to me that even its elaborate appraisals are inadequate, in one very important field, for the valuation of music in terms of those demands of the spirit for nourishment of which I have spoken.

I shall therefore first set before you what seems to me a serious deficiency in the existing apparatus of musical criticism. I shall then propose an addition to that apparatus which, I believe, will fit into the gap of which I complain. I cannot pretend that it will wholly fill the gap, but I shall hope to convince you that the gap needs to be filled and that the earnest effort of musicians like yourselves will be helpful toward that end.

The deficiency I complain of is the lack of any tangible definition, first, of what music expresses (that is, of the nature of the nourishment it offers to the spirit), and secondly, of the process by which expression is achieved. I am glad that I speak to an audience of church musicians, among whom I am sure there are few who are so absorbed in the interest of musical structures as to believe that music is incapable of expressing any other than a purely musical idea. The musical heritage of Lutheranism was not accumulated out of any such belief. I shall be able to refer to that heritage only incidentally, but you will see, when I have finished, that I could find in it thousands of other examples than the single one I shall take for illustration; and I believe you will see in this heritage alone—if the vaster literature of music were not available—an impregnable defense of the thesis I shall propose.

My problem, as I have already hinted, can be set forth in two simple questions: (1) What does music express? (2) How does it achieve expression? These questions sound simple, but you will recognize that they are not so. Precise answers, even to the first of them, are difficult. I shall therefore deal with it at first in somewhat general terms by describing an example in which both the purpose and the fact of expression are indisputable, and then, by presenting the music to you, allowing you to judge the accuracy of my statements about it.

You probably all know the D-Minor Piano Concerto of Bach. It was originally written for violin, and that original version is lost; but Bach’s arrangement of it for piano fortunately exists. The first version was written at Cöthen. Some years later, at Leipzig, he chose as text for a church cantata the somber phrase (from Acts 14:22) Wir müssen durch viel Trübsal in das Reich Gottes eingehen (“We must through much tribulation enter the kingdom of God”). The first movement of the concerto, somewhat uncomfortably transcribed for organ and orchestra, forms the introduction to the cantata. The slow movement—a kind of passacaglia on one of the most pregnant themes Bach ever wrote, with potent harmonies and what must have been originally an exalted violin solo above this bass—he kept intact but added, set to the text I have already quoted, four amazingly interwoven choral voices. The music is a miracle of structure, but it is also evidently intended to appear as appropriate to that text; and it is of the sense of those words, and particularly of the word Trübsal, that I should like you to think just now.

Will you ponder therefore, not merely the dictionary’s definition of the word “tribulation” but the whole spiritual condition it implies, as if you were at this moment immersed in that condition. We shall presently attempt to explore more minutely the substance of this verbal idea; but for the moment I shall ask you only to judge whether there is not a palpable kinship between the mental state to which this word refers and that which is evoked by Bach’s theme:

Click here for music example.

You have now for comparison two very different utterances of what is in essence one state of mind. The words, which are more concrete in their suggestion, may for that reason at first appear to “define” the music; but if you ponder the music as you did the words, I think you will agree that, on the contrary, the music defines the words—that it evolves a more illuminated awareness of the condition of Trübsal than that word, even in its Biblical context, is capable of suggesting. For the word, although it is a conventional symbol for that state of mind, had to ferment for some time in our consciousness before it yielded any vivid awareness. But the music—while it was also intelligible and deeply interesting merely as a tonal structure, yielded so kinetic an awareness of that state that it seems to me at least, the very stuff of tribulation.

Now we cannot suppose that this musical idea, conceived originally as an instrumental piece, had for its composer a different meaning than it displays when associated with this text. Like Beethoven, on the occasion of Napoleon’s death, Bach doubtless found, when confronted with the verbal thought, that “he had already written the music for it.” He merely identified the expressive purpose of his theme by associating it with appropriate words. The music, in its original form, possessed for Bach the power of expression. We recognize, as he did, the propriety of its association with the idea of tribulation. Therefore, for us also music possesses that power. I have thus answered, if only in a general way, the first of my crucial questions. Music is capable of expression.

The second question, however, remains, and that one is much harder to answer. If we admit that in some form the idea of tribulation was expressed, just how was this expression achieved? Where, in the substance of tone that you have just heard, does that agency reside which is capable of evoking in our minds that awareness of tribulation whose existence we have just acknowledged? What, precisely, is that agency? It is to this problem that we must now turn.

I shall have to take you through what may appear a labyrinth of reasoning before the answers can be clear. I cannot pretend that the going will always be easy. Our original question, how does music achieve expression, has at once divided itself into questions of what and where; and still more divisions will appear as we go on, each of them imperatively demanding an answer. I could, by ignoring many of these, apparently simplify the problem and perhaps more easily win your assent to the theorem I shall propose. But in that case, I fear your comment would be that most devastating query, “So what?” And I hoped that I might forestall that question. For I have more to do than to set forth a theorem. I have also to convince you that this problem of expression is really a problem in spiritual nourishment. And I shall confine our analytical observations to this single theme from Bach because its nutritive content, thus minutely examined, will illustrate more decisively than a variety of examples would do, the dimension of this problem.

I believe we all grasped Bach’s theme in a similar sense, unmistakably in accord with the purpose of the composer. That is to say, a process of expression worked. The text has a similar expressive purpose to that of the music. That is to say, there is a common ground upon which verbal and musical expression may meet. Obviously that common ground is not discoverable in the substances or the technical processes of music and language. They meet in the world of experience; and in this case in that region of that world which we call the experience of tribulation. We evoked an idea of tribulation in two very different forms—a verbal and a musical. The actual substance of the ideas was, of course, very different in detail, but in both cases there was reference to the experience of tribulation. How was this done?

The substance of verbally expressed ideas is familiar enough so that we can to some extent explore the process by which this verbal idea arose in our minds. To do this may also reveal something of high importance in the process of musical thought.

What, then, was actually in our minds as we pondered the meaning of the word Trübsal? I think we shall find there two interrelated awarenesses—one, of the impact upon us of some bitter and ineluctable circumstance—a bereavement, a deep injustice, a frustration of cherished hope, or whatever. (I did not say that these circumstances were imagined in detail; I said merely that we sensed their impact as a spiritual influence.) Secondly, and probably with more vividness, the word evoked the image of an emotional state—a state which is the natural and appropriate product of circumstances such as I named.

These two awarenesses were closely related—so closely that we cannot tell which was first—that of the circumstance or that of its emotional correlative. Indeed, the word tribulation may connote either experience from which feeling is generated, or feeling from which a generating circumstance is to be inferred. In either case, however, a relation of circumstance to feeling was established, and without this we should hardly say that an idea existed. This fact, that a probable cause may be inferred from a known feeling response, just as truly as a probable feeling response may be inferred from a known cause, will prove important for our future purpose.

Let me repeat also that neither your image of the experience of tribulation nor your feeling response to that image can be the same as mine. A thousand threads of association will lead to stores of kindred experience that exist in your mind and in yours alone. I doubt, indeed, that the general sense of tribulation which was evoked by the music was as various, in our several minds, as the images of that same state which were evoked by the pondered word. But in both responses these two awarenesses—that of external circumstance and that of an attendant feeling state—were present, in some magnitude, in your mind as in mine; and to the extent of their similarity, we entertained the same idea.

Indeed, from the inevitable presence of these two components of consciousness, I think we may deduce, for the purpose of our further discussion, a usable definition of idea. Let us say that an idea is at once a mental image and a valuation of experience. The image may be an immediate percept—that is, of encounter, at this moment, with actual conditions of experience—an image of reality. Or it may be evoked by various means of suggestion or portrayal—by such vehicles as drawing or painting, or by abstract symbols, such as words—what we might call a virtual image. The dimension of image, however, is not a forecast of the dimension of valuation. Even faint images, exciting what I may call tender spots in the areas of association, may call up valuations of high intensity, out of all proportion to the dimension of the image of circumstance. Also, the whole complex of image and valuation may comprise an amazing variety of detail. But both image and valuation, I believe, in some dimension, will be found in any idea.

I need hardly say that the valuation of the image is that which gives significance to the idea. And in any significant idea, we shall find that the factor of valuation—which is the measurement of significance in the imaged experience for the percipient self—is highly tinged with emotion. Even in such a problem as we are now considering, where a great number of careful intellectual discriminations must be made, and where the final decision will be one of truth or falsity, that decision—which is the valuation of the experiences we have pondered—must excite our love for that which is true (or right) or our hatred for that which is false (or wrong). If it does neither, we ask, “So what?” But even indifference, uninteresting as it is, is an emotional attitude. In the practical business of every-day living emotion is not only a sign of significance in experience, but is the sign upon which we chiefly depend to measure that significance. We work as well as play, we strive and placate and pursue and shun—not merely for immediate but also for ultimate satisfaction, not for the mere joy of momentary achievement or the relief of escape but to gain the sense that the self has been nourished.

In terms of feeling, then, rather than in terms of the more laborious intellectual discriminations out of which the feeling summary is derived, we measure the significance of rea1 or imagined experience. In terms of their feeling value, also, we store up memories of past experience. And as evidence that this swift feeling-judgment has been depended on for ages, we possess in addition considerable inheritances of instinct or intuition which may contribute to our valuation of new experiences as they arise. What we call our convictions—however cleverly we may “rationalize” them, as the young people say nowadays—are comprehensive emotional summaries of real or imagined experience. And because our convictions, or our lack of them, are visible to our friends, they accept this evidence of emotion, or of disposition to emotion, far more confidently than any brilliance of intellect, as the true index of our character.

For the purposes of art, this fact is of the highest importance. It is said, sometimes in praise, sometimes in blame, that music is the language of the emotions. But in reality, all art is a language of the emotions—a language which speaks in terms of those final emotional summaries which I have just described and which consequently loses all but a tithe of its appeal to the world at large when it fails to speak in terms of feeling. I am quite aware that feeling, amounting even to conviction, may arise from false images of fact and thus yield a false valuation of experience. Such feeling we call sentimentality. But to recognize the danger of exciting false feeling-judgments, and to attempt to substitute what is supposed to be a more intellectual attitude toward experience, may only substitute one emotion—in the last analysis, a species of fear—for another. And to assume a coldly intellectual attitude toward art—to demand of it, along with its sensuous appeal, intellectual satisfaction only—seems to me a similar error; a similar attempt to deny the normality of the emotional valuation of experience. The product of such a denial is self-incarceration in an ivory tower.

Bach possessed, and constantly exercised, a musical intellect of the highest order. But he had no love for ivory towers. He chose for his works, as for this cantata, themes (and in this case words also) which bore reference to general human experience, knowable in terms of feeling. The core of this cantata is the theme of Trübsal; both words and music bear reference to that experience. In these two utterances, the text and the music, we have thus an exemplification of two methods of expression. The method of the text—although that is perhaps hardly an art-work—is nevertheless that of the so-called “representative” arts. This method portrays or describes or otherwise recognizably represents conditions or facts of experience in that aspect which will arouse in the mind of the observer an emotional realization—a valuation—of the import of that experience. Literalness of representation may or may not be required. The merest hint of the simplest fact, in a suggestive context, may evoke in us a valuation wholly incommensurate with the dimension of the act. For a poetic example, think for a moment of Dante’s story of Paolo and Francesca—the two lovers who read together the story of Lancelot. Presently, as they read, something overwhelmed them; but the most vivid sign of that event, in Dante’s version of it, is that the book was forgotten: “In its leaves, that day, we read no more.”

A fact of experience is here portrayed in such a way that we who read will ourselves complete the given image of experience by interpreting—or as I have called it, valuing—the act. Dante knew we could complete it—he was compelling us to complete it—out of that store of experience which he knew we possessed. The essential fact in the method of representative art is that from a portrayal of fact an inference of feeling is drawn—that an image of experience has been valued—that an idea has been communicated.

Music, to be expressive, must also intelligibly communicate idea. But it is at once evident that this representation of external experience cannot be the method of musical expression. Music is almost always ridiculous when it attempts the literal portrayal of external fact. What, then, is its method? We may learn something of it by examining more minutely Bach’s passacaglia theme, since that music, like the text, did achieve the expression of the idea of tribulation.

It did not portray external conditions of bereavement or injustice or frustration or any other fact that would ordinarily induce the mental state of tribulation. It portrayed that mental state itself—or at any rate the feeling-component of that mental state. Consequently, the idea evoked by the music is vastly different from the idea evoked by the word for that state. For the word, as we normally use it, implies conditions of experience and connotes, rather than portrays, feeling. The music, on the other hand, portrays feeling and connotes conditions of circumstance appropriate to that feeling. If it were not so, we should not have been able to identify this theme as related to tribulation. We did so identify it. An image of the experience we call tribulation was inferred from what I think we may call a portrayal of feeling.

If this is true, then the method of musical expression is also a method of representation. To many of us that statement may well appear alarming and perhaps subversive. Music is generally supposed to differ from all the other arts by virtue of its complete disassociation from external experience and by many it is held to be thus exalted above the other arts. Some, following the lead of Hanslick in his little book Vom musikalisch-Schönen and of Edmund Gurney in his big and important book The Power of Sound, have argued that music really expresses nothing save the musical idea itself. Others, unwilling to abandon the generally recognized relation to feeling, have tried to retain the proud isolation of the art and still uphold its power of expression by describing music as a “presentative” art, as against the others which are “representative” or “re-presentative.” This theorem requires a moment’s attention.

By the method of presentation, a composer’s emotion, somehow embodied in tone, is supposed to be re-created, intact, in us who listen—apparently by our mere exposure to the musical stimulus. Just how the composer embodies the whole substance of an emotion in the substance of tone is not clear, nor have I found any account of this theorem that recognizes the need for more than musical thought in identifying the emotions supposedly thus “presented” with the rest of our conscious existence. Daniel Gregory Mason, a spokesman for the theory of presentation, avers that music deals not with things but with “disembodied feelings, passional essences.” This is attractive-sounding nonsense, but it is nonsense all the same. What is a disembodied feeling? How are we to grasp it—how identify it as a human fact? Mr. Mason’s account of the operation is not reassuring. Music, he says, “abolishes thought, to set up in its stead a novel activity that is felt as immediately, inexplicably grateful.” Sentimental maidens who accept music as the food of love seem, by comparison, to be blindingly rational creatures. They at least associate music, however indiscriminately, with a reality of experience; but the contemplation of a disembodied feeling is indeed the abolition of thought. I doubt that rational beings will find that condition of consciousness especially grateful.

By comparison with this, the position of the purists, who find that music expresses nothing save the idea of music itself, is wholly rational. Obviously, I differ with them, but I have not time in which to set forth what appears to me to be their point of view. If, however, their opinion were generally held—if the world in general saw no relation between music and the most interesting things in life, which are man’s emotional valuations of experience—I doubt that the musical public would be numberable in millions.

I contend, then, that music in so far as it is an expressive art is a representative art. It is virtually incapable of representing things; but it does represent emotions.

This supposition may at first sight seem as incredible as the notion that music can present emotion. The turmoil of an excited mind seems, as a mental fact, incapable of definition or of any representation. But our minds are themselves somehow embodied in the twelve pounds of solid matter and the six pails of water of which we are made, and the behavior of that part of the body which is not the mind is so correlated with the turmoils of the mind itself that from that bodily behavior—and in the last analysis from no other source—we are able to know the turmoil itself. Emotion is recognizably displayed in bodily acts. Is it possible that music can represent the behavior of the emotionally disturbed body accurately enough to evoke an idea of the emotion as it exists in the mind? Unless this is possible, my whole theorem is indefensible.

Psychology has differentiated and defined and measured various features of emotion, but it has never disputed the existence of two elemental factors of emotion that were familiar to humanity—and to artists—long before the science of psychology was invented. The first of these is a perceptible condition of nervous tension—or of its opposite, nervous calm—which is an invariable characteristic of a state of feeling. Although this condition is not externally manifest as a nervous state, it is often clearly perceptible in consciousness, and a considerable variety is distinguishable in these states—distinguishable in terms of inner tension. The second, more manifest because it is externally perceptible, is a product or at any rate a close correlative of the nervous tension. This is the apparently inevitable innervation of certain muscles which we call the motor outlet of the emotion.

I believe that these two elemental factors of all emotional states can be recognizably portrayed by music. In fact, I believe I can show that in this portrayal we have two—and the only two—elemental factors of musical expression. If this is true, it may constitute the foundation of a workable theorem of musical expression.

Such representation may well seem both too remote and too fragmentary to evoke what, if expression is actually to occur, we can recognize as idea. And expression, in any rational use of that term, is the intelligible communication of idea. But these manifestations of nervous tension and motor impulse are perhaps more highly characteristic than we at first suppose, so that a really vivid representation of them may be sufficiently suggestive for our purpose. Let us inquire, therefore, into the implied meanings, as we ordinarily observe them, of these two facts.

We must remember that we are dealing with emotional states—those that we have described as the valuations of experience—and not with mere instantaneous reflexes such as occur when I jerk my hand away from contact with the hot stove. That jerk is not the product or the characteristic of an idea. It is only a reflex. But as I realize that my hand has been burned, an idea does arise—one that is in some part a product of the continuing pain, but is in larger part what I might call an overall sense of injury to myself: a feeling in which resentment, both at fate and at my own stupid carelessness is mingled with disappointment over the fact that I shall now be unable to use my hand for something that I had hopefully planned to do. These nervous stresses are as much a part of my whole consciousness of the accident as is the sight of the blister on my hand (which also adds to my distress); and they will have, either individually or collectively, their motor outlets. I am conscious of them, not as individual strains of excitement, but in their concrescence as a feeling-sign of my present state. With some effort, indeed, I can differentiate this sense of tension from the whole of my conscious awareness. These tensions, at any rate, are very different from those which would have accrued if I had barely escaped burning my hand instead of actually burning it.

If you will now compare with this simple instance of feeling and idea the much more complex awareness which is symbolized by the word “tribulation,” I think you will agree that there is only a difference in degree, not one in kind. The contributory experiences are far more numerous, but in the last analysis they are injurious contacts between my self (which is more than my body) and the world (which is more capable of doing me injury than is a hot stove). Yet, just as we can imagine the bodily injury of burning and its feeling-consequences, we can imagine the spiritual injury of tribulation.

We doubtless cannot clearly describe the nervous conditions themselves which are characteristic of the state of tribulation. Yet we do know these conditions. They were in some measure imagined by you when I asked you to ponder the meaning of that word. Indeed, an image of your mental attitude toward those conditions formed a considerable part of your whole awareness of the meaning of that word. That image was thus a part of the definition of that word as you understood it. We noted also that the word did not so much define that meaning of Bach’s passacaglia theme as that theme itself defined the word. And that was because no small part of the meaning of the word, when that word comes alive in our minds as the equivalent of a living state, is an awareness of feeling.

We do know these nervous conditions, however ill we may be able to analyze or describe them. They are, indeed, the immediate substance of all our emotions—signs through which external experience and its significance is translated into consciousness. They are so variable—so delicately adjusted to the character of each experience as we undergo it—that it is not strange that we have but few and feeble words by which to symbolize them. But, I repeat, we do know them, and know them with a certainty such that if we found them adequately represented we should at once recognize the likeness.

We do not, however, usually observe these states as nervous conditions, isolated from the rest of our consciousness. Arising in connection with, and as a product of, external experience, they are so palpably related to that experience that, instead of trying to name it by some word directly indicative of feeling, we identify or define the feeling in terms of its experiential cause.[1] The verb “to love,” for example, does imply a general mental attitude and a corresponding nervous condition; but that word has little vividness of suggestion until an object for our affection is presented. We do not just love. We must love something or somebody; and the nature of our affection is largely determined by its objects. We love men and churches and bridge and oysters—but not with the same affection; and if you will dispose your affection on each of these disparate objects in turn, you will see that in each case a considerably new complex of nervous tensions arises—a new set of overtones for the original fundamental of affection. Almost all our names for emotion are similarly vague, when used as symbols for the states they attempt to signify. Yet, once associated with their appropriate external cause, the tensions vaguely implied by these words come clear in the mind as vivid images of feeling. We do know these nervous conditions.

Sometimes also conditions of feeling arise for which no assignable cause seems to exist. These conditions we call moods; and some of them are definite enough so that we give them names, such as pensiveness, boredom, and buoyancy. Here, more precisely than with the word love, the symbol suggests the overall nervous condition—has to suggest it, indeed, because there is no fact of experience present that can explain or account for it. It is true that such words are somewhat vague, but the fact that they exist at all is proof that the nervous conditions they symbolize are characteristic and sometimes vitality important facts of consciousness.

Our verbal vocabulary for all states of feeling is feeble. The states themselves, however—whether “transitive,” like love, or “intransitive,” like boredom—are important, always to our own selves, and sometimes to the selves of others. An adequate vocabulary through which important facts of experience can be expressed and discussed is of incalculable value for every man. The nervous states which are aroused in us by experience are the signs of significance in experience itself. Thus if we can find in music an adequate vocabulary—an effective means of communication—regarding these states, not merely as feelings but as indices of significance in experience, it seems to me we shall have added incalculably to the effective vocabulary of experience. For it is through such a vocabulary—through the knowable communication (which, literally, is a “sharing together”) of that which is in other minds than our own—that the spiritual self is nourished.

The remaining element of emotional experience (considering emotion as a psychological process) is the inevitable motor outlet of accumulated nervous tension. This outlet is more readily observable than the inner tension itself, since it often appears as the visible motion, or as a characteristic of the motion, of various members of the body. These motor outlets, it is true, are seldom the most direct, or as we might say, natural, products of the originating tensions. They are very often redirected or inhibited motor acts. Indeed, it is probable that we do not feel them unless they are inhibited. But in many cases these inhibitions are purposive, arising from conscious effort toward control or concealment; and since we all find it desirable to adapt our behavior to that of our fellows, the result of all this careful redirection is a kind of socialized behavior. It is not, however, on that account less interpretable as an index of feeling. I do not strike the pretty lady whose sharp heel has ground into my toe on the crowded bus. Instead, I smile sweetly—or sourly—making a grimace which, unless she is very stupid, she will recognize as an effective denial of my lying assurance that it didn’t hurt.

This will suffice, I think, to indicate the vast possibility of motor behavior as an index of feeling—of feeling which, as a valuation of experience, is a vital portion of idea. I need not dwell upon the extent to which we interpret the ideas of our fellows in terms of their conscious or unconscious motor reactions. We watch for these signs and depend on them, in almost any encounter, to supplement our understanding of what is being communicated to us in words. Consider your reliance, for the interpretation of the whole state of mind which accompanies an expressed thought, on posture, gait, facial expression, and that most subtle motor release which is manifest in the inflections of the speaking voice, and you will be sufficiently aware of my present purpose.

It is also worth noting, however, that just as we have inexplicable nervous conditions which we call moods, we have also motor impulses for the execution of which we have no muscular equipment. In spite of this lack, some few of these impulses are so vivid that, like some of our moods, they have been symbolized in words. Consider the etymology of the word “elated” which means “borne up”—off the earth, and kept there in defiance of gravitation, by some mental image of spiritual excitement. Ecstasy similarly is “being outside oneself”; depression is “being weighed down”; transport is “being carried over” into a region we should never be able to reach in the body. I can find but few words that imply such inoperable muscular impulses as these; but these few will show that such impulses exist, and they are probably more numerous and more important for the whole fact of consciousness than we realize. But taking them along with the vastly greater body of recognizable motor impulse, we find again a fact of consciousness for which an adequate vocabulary is in the highest degree desirable.

These two facts of nervous tension and motor release are essential factors in the whole phenomenon of emotion. No matter what the stimulus, they appear; and they are therefore elemental constituents of emotional states. We are seeking, in the substance of music, for an effective vocabulary for these states—for that which in some way can represent or portray them. It will be natural, then, to search first for those features of the musical substance which show direct relation or some indubitable similarity to those elements of emotion. If we can find them, and if, in the ordinary reception musical utterances, their presence and their correlation with the elements of psychological experience is readily recognizable, then we may hope to have discovered the elements of musical expression.

What these elements really are; how far they may need to be complemented by other factors; and how, as a cooperative body of elementary and secondary suggestion, the whole vehicle of musical expression functions, is a problem at least as complex as that of the identification of the elements of expression. But if we can with some certainty identify these elements, the direction of our study will have been indicated. Our next section will take us far along this road.

Elemental and Secondary Factors in the Process of Musical Expression

Our search for the bases of a tangible process of musical communication appears to have yielded, thus far, the following results:

Having agreed that music can relate, not to the facts of external experience but to the mental response aroused in us by our confrontation with those facts, to the idea formed in our minds, we have defined idea as at once a mental image and a valuation of confronted experience. That valuation is an estimate of the significance of the experience for the self that encounters it. It is both an intellectual and an emotional valuation, for while the facts are recognized as facts, there is also a deep concern for their effect on the well-being of the perceiving self. That concern is largely emotional, but it is also very complex. It is compounded not merely out of regard for the immediately confronted facts but also out of our memory of innumerable former occasions of experience, similar in some degree to this one, and out of our certainty that similar occasions will be encountered in the future.

This feeling, whether anxious or hopeful, is not apprehended merely as feeling. We find it an appropriate response to the situation that confronts us. If, then, this emotion can be portrayed—whether in music or through any other medium—a dependable inference may be drawn as to the nature of the experience to which the feeling is appropriate.

We have now to see (1) whether such musical representation is possible and (2) whether such an inference, from such a representation, can reliably be drawn; for without this inference we shall have an apparent valuation—a valuation of the unknown, which is undesirable, if not actually unthinkable.

We found that common sense, corroborated by psychology, distinguishes two elemental factors of emotional states—a characteristic nervous tension and a correlated (and thus equally characteristic) motor impulse which is the outlet of the tension. We have seen that these nervous and motor manifestations are of immense and subtle variety and are closely related in consciousness to our awareness of the external circumstance which is their cause—so closely that they constitute a part of our idea of the experience, and contribute incalculably to what we call the “definition” of the name by which we call the experience. We are now to explore the substance of music to see whether it possesses properties which enable it to portray these elemental facts of motion and tension. If we find them, it will be natural to suppose that we have identified the elements of musical expression.

But we must not suppose that if our search is successful, it will at once reveal the whole detail of the process of expression. Elemental facts are by no means always obvious. They are often so completely hidden by what is on the surface that we do not even suspect their presence. This is true of our present problem. There are a large number of expressive devices, used by every interpreter and familiar to us all, which I have as yet not even mentioned. I am as keenly aware as you of the value of these devices; but we shall find that, like the parallel devices used for inflection and emphasis in speech, they are dependent—and therefore secondary—factors in the whole process of expression. They are a manner related to a matter; and the matter to be uttered must determine the manner of its utterance.

The indubitable designation of the emotional facts of nervous tension and motor impulse as elemental determines precisely the direction of our immediate inquiry. We are to see whether there is that in the substance of music which can recognizably portray or effectively represent these two characteristics of emotional behavior.

No musician needs to be told that the substance of music is in its very nature an extraordinary complex of fluctuant tone-stresses. Indeed, the correlation between these and the stresses of feeling is probably obvious to everybody. But more is needed for actual expression than a possible correlation. The tone-stresses must be shown to be capable of that precise suggestiveness which will portray recognizably that emotional tension which is a part of our valuation of an experience. To show so precise an adaptability, we must explore the various features of tonal arrangement and relation which convey the sense of stress—or of its opposite, the sense of calm, which may be equally important for expression.

Three types of tone-stress are universally recognized—indisputable features of the musical substance and familiar to everybody; but their differentiation is often unobserved. I will therefore take time for a brief summary of the values useful for our purpose inherent in each of the three types.

The first of these—the simplest, the most obvious, and the least precise for expressive suggestion—is that of relative height and depth of pitch. I need hardly say that high tones normally suggest greater stress than low tones, or that progression from low to high generally implies increasing tension, and progression from high to low, relaxation. The maintenance of a single level of pitch has also its natural implication of unchanging tension. Other factors of the musical substance may alter or even negate these primary implications—for example, a rising line which diminishes in intensity or fades into obscurity. But such progressions are exceptions, and you will hardly find in all musical literature a tense climax which is achieved by descending progression.

The second tone-relation suggestive of stress depends for recognition on the prior acceptance of a tonic, or at any rate of a tone-center, in relation to which other tones stand in various degrees of stability or instability. In our musical system (with which alone this discussion is concerned) the tonic is the most stable note—the gravitational center of the whole tonal mass. Our almost invariable dependence on harmony in music-making has reinforced this sense of stability in the tonic, making it more positive than was the final in the old modes. By virtue of the shift from modality to tonality, with consequent emphasis on the sense of major or minor key, we have also come to feel that the third and fifth of the tonic chord, whether that chord is major or minor, partake of the stability of the tonic. This is so far true that in what may be called “standard” melody—the three notes of the tonic chord are distinguishable from the others as “rest” tones, while all the other notes of the scale, diatonic or chromatic, appear as “active” tones. Analysis of standard melody will show that, in principle, active notes tend to proceed to the nearest rest tone. They have, of course, great liberty of progression, so that melody from an active tone may not only execute considerable gyrations before it proceeds to its destined rest point, but may also evade that point altogether—with all the advantage of novelty and logic combined. Because of this high variability, it is impossible to ascribe precise values of activity, or even of rest, to any one of the notes of the scale. The seventh note, the “leading tone,” is as such the most active note of the scale. But this same note, heard as the fifth of the mediant chord, loses much of its tension. Similarly, the tonic, dissonantly harmonized, loses much of its value of rest. But it is obvious that without this high variability, music would be unequal to the suggestion of many subtleties of stress, a great number of which suggestions must be available if the fact of expression is to be attained.

The third type of stress suggestion is of course that of harmonic concord and discord. These, palpably, suggest rest and activity. Like rest and activity in the notes of the scale, and to a far greater extent, these qualities are fluctuant and variable, and description would be merely tedious. It is worth noting, however, that active notes, in melody, may be harmonized by either concord or discord, with great variety of suggestion; that rest tones may be similarly subtilized; that the rhythmic position of tonal stresses may be adjusted to almost any purpose of intensity or its opposite; and that in consequence these three primary values of tone-stress, working in combination, present a pretty adequate equivalent of the immense variety of nervous stresses which are a part of our amazing equipment of reaction to our daily contacts with the world.

Appropriately to his expressive purpose, also, the competent composer, in a given musical movement, will provide a general norm of harmonic intensity—the “environment,” so to speak, of his more specific thematic utterances. This environment may contribute greatly to the vividness of the thought expressed, or may of course be so badly handled as to interfere with our apprehension of that thought. A high norm is not, in itself, an infallible index of high intensity of expression, and if such a norm is established at the beginning, it is obvious that still higher excitements must be sought at the expense of more and more unusual harmonic tensions. And whether the end result is an actual gain in expressiveness is a question that can be finally answered when the real idea of the composition is valued, not in terms of the ingenuity of the language but in relation to other ideas—to other images and valuations of experience.[2]

The various types and combinations of tone-stress may of course be exhibited—and interestingly exhibited—without reference to any external experience. So rich a resource for pattern-making cannot but appeal to the born patternmaker—which every artist is. But that is not our question. Without destroying the musical interest of the fabric, can they also be so disposed—have they been so disposed—that at the same time they pointedly suggest states of feeling? I believe they can. And their power is further enhanced by another resource which, so far, I have all but ignored. This is the fact of motion—in itself possibly representative of activity or calm—which may subtly qualify the purely tonal suggestions of stress. We are as keenly aware of the motion and the rhythmic propulsion of music as of the pitch, activity, or dissonance of the tones; and the whole quality of a given stress-moment may be profoundly affected by the rhythmic energy which propels it or the lack of energy which impedes it. Surely there is enough here so that the skillful and imaginative composer may use these resources for expression—for intelligible reference to, or portrayal of, those facts of nervous tension which are elemental to emotion itself.

The possibility that we have here an actual element of musical expression seems to me evident. Our recognition of the fact of motion as a component of what is largely perceived as tone-stress will somewhat blur the identity, as an element, of the musical fact we are observing. But it seems to me that for the purpose of verbal discussion we may call this element by the name we have already often used—that of “tone-stress.”

We found that the fact of motor release was another elemental fact in the psychology of feeling. The correlative of this should prove to be another element of musical expression.

It may well appear to you that this element has long since been identified. You will point at once to the familiar fact of rhythm as the fact we are seeking. No one, at any rate, will deny to music the power of exhibiting, and so possibly of portraying, rhythm. For since rhythm, whether in the creeping of the worm or on so vast a scale as the precession of the equinoxes, is invariably characteristic of motion, that which portrays rhythm can be understood as a portrayal of motion.

But again, recognizable portrayal of characteristic motion, such as is required for the purpose of intelligible expression, is not attained by a mere general exhibition of motility. Music, like a human body, is capable of both, and a discrimination is necessary.

The conventional sense of the word rhythm, which designates an element of musical structure, barely comprises the fact of motility, and certainly lacks the implications, needful for expressive suggestion, which I believe to be present in the musical substance as we all normally apprehend it. The usual process of musical analysis defines or exhibits rhythm in terms of marked instants of accent and nonaccent—essentially as a series of detached thumps of varying weight—which mark the instants and the relative energies of propulsion in the musical mass. Doubtless, musical analysis implies more than this; but the word “rhythm,” in this context, still fails to connote all that fact of motion which the word in its original etymological sense conveyed.[3]

The fact of motion in music, even to the ordinary observer, appears as far more than the succession of thumps—the mere instants of propulsion—which the marks I have mentioned indicate. For just as the body of the dancer as visible between the accents that define its propulsion, so the “body” of the music rises and falls and gyrates and floats; and it is this motion of the whole body, human or musical, which is interesting. What happens between these instants of propulsion, indeed, is usually more indicative of the real character of the motor-effort than is the pattern given by the thumps. Any given rhythmic pattern—dactylic, iambic, or whatever—may be exhibited in music which moves with energy suavity, timidity, grace, hesitance, or a hundred other characteristics; but I think you will agree that these characteristics, far more than the mere basic pattern (which is no more than a poetic foot in comparison to the poetic thought) are the really compelling features of the music.

Structural analysis can usually classify rhythmic patterns in terms of the poetic feet from which at least the principal meters of music are derived; but musical motion, even when exhibited an a single poetic meter, is so diverse that no given pattern of rhythm may be said to retain any intrinsic feeling character. Even such a general classification as that which we found possible in tone-stresses is thus impossible. But that does not in the least minimize music’s power of motor suggestion. It is unfortunately true that the qualities of motion I spoke of (energy, suavity, and so on) are not as positively indicated in musical notation and design as are the facts of tone-stress (dissonance and tonal activity) which we have already discussed. The interpreter, therefore, comes into our picture, and his presence must ultimately be recognized; but for the moment we shall try to do without him—to show, indeed, what he, as executant, often fails to perceive.

Musical motion is of all possible degrees of speed—lagging far behind or far outrunning our normal two-legged gait. But motion at almost any speed may appear as impeded or impelled, laborious or effortless, or anything between these extremes. The tensions of the musical substance, as well as the nature of the melodic line (legato, staccato, syncopated, or whatever) will greatly influence the character of the motion. The substance of harmony, likewise, not merely by virtue of its consonance or dissonance but by its obviously heavy or tenuous mass, may affect both the superficial motion of the melody and what I may call the apparent momentum of the whole musical substance. These things are exhibited in the rough in our present system of notation; but even when the actual substance of music (in the given notes) and its rate of motion (in the composer’s designation of tempo) are preestablished, a keen interpretative insight is needed if the intended character of the musical notation is to be appropriately exhibited.

I have doubtless mentioned no aspect of musical motion which you have not long since perceived, nor any which is not a recognized feature of music considered merely as music. But again I pose the question: can this suggestion of motion be understood—has it been understood—by composer and listener alike, as the correlative and even the representative, not merely of such physical facts as the rending of the veil in the St. John Passion, but of the subtler motor acts which reveal our mental states? If so, I shall be justified in regarding this portrayal of motion, which may appear in a variety as endless as that of actual motor impulse within ourselves, as a second element of musical expression.

Since not only the continuity of the represented motion but also the characteristic propulsive or impeding stresses perceptible in it contribute to the whole motor fact we are considering, the word “rhythm,” in its conventional, technical sense, seems to me unsuitable as a name for this element. I have therefore borrowed for it a phrase from Edmund Gurney, and call it “ideal motion.”

These two facts of tone-stress and ideal motion in music are possible correlatives of the two psychological facts of nerve-stress and motor impulse Because the psychological facts are elemental for emotional states, I believe their musical correlatives to be elemental for the musical expression of those states. Because the elemental facts in the psychological response to the stimulus are two in number, I believe that there are two, and only two, elements of musical expression. They seem to me to offer—without detriment to the interest of the musical substance as such—dependable reference to the world of general experience; and I think I have sufficiently shown that expression—the intelligible communication of idea—can hardly occur without reference to the world of general experience.

I should have liked to go on from this point to a discussion of the manner in which these elements function—to describe their actual working, as I see it, in the process of expression. But I cannot yet do that. I am building up a hypothesis, and it is not yet complete. The elements, as I have called them, are indeed the fundamentals of that hypothesis; but they function conjointly with other contributory factors which remain to be described.

And there is another reason for delay. Even the elements, in the perspective in which we have seen them, have an unwonted look. We have seen them—for clearness of identification—as no more than the correlatives of certain details of emotional experience. But they are also—and first of all—musical phenomena. They appear, not in isolation (as for identification I have described them) but as if in solution in the fluid musical substance. Although I may have, as I may say, precipitated these elements, they do not function for expression in this state They function only when in solution.

I contend that the substance of pure music can with no detriment but rather with high addition to its musical interest, contain in solution the active principles of expression. It may also contain these principles in an inactive—that is, an unsuggestive—state. And it is also possible, even though the active principles be present and active, to ignore their presence and see the music merely as pure. Bach’s passacaglia theme is as well constructed as music can be. I believe you also found it expressive; but to look at it as a sample of pure music will perhaps keep you from thinking that I have no interest in this aspect of the art, and it will also offer an opportunity for something like a judgment of the contribution which expressiveness may make to pure music, or of the derogation from its pure value, if you find that that occurs.

What does this theme yield, when observed as pure music? You can see a coherent overall design, based on the motive announced in bars 1 and 2—the rising line of a triad, followed by a sudden descent. In bars 3, 4, and 5 this pattern is pursued in a kind of diminution or foreshortening—the note completing the drop being also the first note of the next rising triad. Variety in this continuation is gained by the use of two diminished triads instead of the minor triad of the opening; these two are followed by a major triad; and the quasi-cadential value in bar 2 reappears only in bar 6, where the pattern of bar 2 also appears in a foreshortened state.

The middle part of the theme, which may be seen as the B-section of an A-B-A form, ensues. It is made of the diminution of bar 2, just mentioned, ingeniously combined with the swerving figure B-flat, F-sharp, G; the motion is now in continuous eighth notes; and the characteristic fact of the sudden drop is exhibited at every strong beat. But the notes which complete the drop form a definite pattern of ascent (B-flat, C, and D) to the E-flat, upon which a pause is made. The E-flat, however, is an unstable note (the submediant) so that continuation is clearly implied. The final A-section is entered upon by a kind of inversion of the swerving figure of the middle part; there is another rising triad (uncertain in that it is again diminished) and another sudden drop to the lowest of the many low notes thus emphasized. The cadence is striking. It departs from the motive patterns so far persistently used into a straightforward figure, rhythmically—and by implication, harmonically—decisive. Although no harmony is actually heard, the triad arpeggios and the rising line in the middle section delineate unmistakable harmonic progressions, and these, almost at every long drop, are given a high degree of unexpectedness.

Certainly, with no thought of expression, this theme presents a compelling musical idea. Nor is this pervading musical thought irreconcilable with the definition of idea to which I have bound myself. There is an image and a valuation of experience. The experience, it is true, is not portrayed. It is presented. But it does yield an image—an image of form. And our valuation of that image is partly in terms of our satisfaction with any patently perfect organization and partly in terms of the sensuous interest of the tones themselves, in their contrasts of register, their unexpected behavior, and other aspects.

Since we have already associated this theme with a nonmusical idea, we may find it difficult to view it in a nonexpressive aspect. But I have tried to be just in my appraisal of its purely musical value. If you prefer to accept only the purely musical idea, I shall have no ground upon which to complain; but I doubt that that will be your decision. However, this excursion has led us astray from the straight line of our inquiry, and we must find our way back. You will not have forgotten that in addition to the elemental factors of expression, of which we should now have a pretty definite idea, are an undetermined number of accessory or contributory factors. To these and their function we must now return.

We shall find most of them akin to the endless varieties of inflection and intonation which contribute to the meaningful utterance of verbal ideas. Their contribution, whether in music or in speech, is so great that we often quite forget the fact that they are essentially contributory to the conveyance of thought but do not themselves contain the thought. However eloquent may be a verbal inflection, it is so because it reveals the true significance of the word to which it is applied; for it is evident that if this same inflection were applied to a word of different meaning, or to an almost meaningless word like a preposition, its effect would be ludicrous instead of expressive.

The same may be said of musical inflections. I hardly know how to name all these secondary or accessory resources. Of them all, I believe appropriate dynamic shading to be the most important; but you may equally be right if you feel that first place should go to precision of rhythm, and to that kind of rhythmic shading which, though all but imperceptible, contributes incalculably to the vitality of the motion aspect of music. Ritardando and accelerando are obviously in this category. When this kind of shading occurs within a phrase, we call it tempo rubato. When it becomes too conspicuous, we call for bicarbonate of soda. In the wake of all these, which must be an interpreter’s incessant care, come such resources as tone color, contrast of register (as when a violinist plays high on the G-string), lightness or weight in attack or release (really, but not always apparently, a dynamic device), and many others which I need not catalog. Of all these, only the shadings of dynamics or tempo require any discussion.

Everybody knows—even those performers who, when they play, seem often to forget it—that minute control of dynamics is of the highest importance for musical utterance. Nor is a mere indiscriminate placing and shading of dynamic stress sufficient. Inappropriateness is almost as annoying as dynamic indifference. But for a judgment of appropriateness we must answer the question “appropriate to what?” and the answer must be “to the sense of the phrase.” That phrase, as we have seen, may by different interpreters be regarded either as purely musical or as also expressive, and the character of the phrase may be considerably altered according to this opinion. In general, however, the syntax of a musical phrase is a structural fact which remains the same in any enlightened view; and it may be worthwhile to describe what seems to me an illuminating analogy with verbal phrases.

In any phrase of words, you will find that the precise utterance of your meaning demands that some one word in that phrase be given predominant emphasis. While this may be an emphasis of sudden softness, the most frequent demand will be for emphasis of loudness, with subordinate points of stress placed as the sense demands. The position of verbal emphasis is clearly determined by the meaning of the word. Notes have no individual meaning such as words have, and their sense is created by their relation, in intelligible sequence and combination, to other notes. Yet a phrase of notes—as is indicated by our unquestioning use, in music, of this same word “phrase”—has a similar contour to that of a phrase of words, and for effective utterance it must display similar points of predominant and subordinate stress. The exact dimension of a phrase (which may have two notes or an indefinitely greater number) is sometimes a matter of legitimate dispute. The carelessness with which the slur was used as a mark of punctuation by almost all composers until at least the second half of the 19th century adds to the difficulty. Yet, I believe that if the notation is obscure, musical common sense will generally provide at least a tolerable solution of the problem. But to determine the stress-point is not enough. The quantity of stress is as important as its position; its dimension in relation to other notes is by no means a matter of rule; and the relation of this to the surrounding phrases, and at last to the whole design or sense of the movement, poses incessant problems which only keen intuition—and of course perfect technique—can solve.

The problems of rhythmic shading are so similar that I hardly need to dwell on them. It is probably true that dynamic shadings are more closely related to the element of tone-stress and that rhythmic shadings relate more frequently to the element of ideal motion; but this is an almost useless generalization. To extend the catalog of these values would result in the utter boredom of us all, who are familiar with niceties of utterance which, described at length, would appear but as two grains of wheat hid in two bushels of chaff. It is time for synthesis, not for more analysis.

The apparatus of expression, as I have described it, seems to consist of two elemental factors, tone-stress and ideal motion, whose function is the portrayal of conditions of nervous tension and motor impulse within ourselves. That apparatus also includes various contributory or accessory factors, usable for the heightening of the portrayals given by the elements. Through this apparatus, if its parts are made to work harmoniously to a single end, I believe it possible to convey intimations as to actual states of mind—intimations which, rightly apprehended, may be enlarged into images and valuations of experience. That enlargement is not provided for by the music, which gives only the intimations. The recipient mind must accomplish the enlargement for itself, and it will do it in its own way and according to its powers of inference. Images of experience, immediate or represented, must be erected into idea through this process of inference. Images of the feeling characteristics induced by experience may similarly be erected into ideas by the process of inference The only difference in principle is that the process of inference, with music, works in the opposite direction to that which occurs when, as is usual, we are confronted by an actuality or an image of experience.

Neither, I believe, is this reversal of the direction a serious impediment to exactness. Words themselves are far from conveying with absolute precision the thought of him who utters them; and when we deal with the highly suggestive words out of which poetry is made, the inferences we draw are as various as those we draw from music. Not only you and I but actors like John Gielgud and Maurice Evans, and also all the eminent Shakespearean scholars, past and present, will draw different implications of meaning from Hamlet. But beneath these differences lies a deep well of unanimous opinion which they and we share and which is the best and perhaps is the only final evidence of the significance of that tragedy. Analysts can show us details which substantiate that significance; but neither they nor we can express the whole of that deeper opinion which we all share. We are reduced to the dull statement that Hamlet is a great play.

The reason why we cannot express it is, of course, that what I have called an opinion is both less and more than that. It is less, because an opinion is a reasoned judgment, and reason is not evident in that banal assertion of greatness. And it is more, because the reasons—which we could give, if they were demanded—are swallowed up in our embracing consciousness of greatness. That consciousness is too comprehensive to be shaped into a single clear image of experience out of which most of our puny ideas are generated. But it is not the less an idea—an acknowledgment, not lightly offered, of the spiritual nourishment that our starving selves have received.

I have set before you a hypothesis which, if it really works, should add appreciably to the soundness of our judgments of musical value. Whether it will work or not remains to be seen. We shall look once more at this theme of Bach’s to which I have so persistently referred. I have too much respect for the principle of sufficient reason to suppose that in one demonstration—however successful it may prove—the validity of this hypothesis can be finally proved. But I shall hope that it will be convincing enough so that you may use it for yourselves, in your own valuations of the spiritual nourishment which music offers.

A Test of the Process of Expression

What I have presented is a hypothesis—and is no more than that—of the process of musical expression. It has been, I hope, a reasonable hypothesis; but many an unworkable theorem can be reasonably argued. This hypothesis is valueless if it will not work; and I shall try to put it to as severe a test as my opportunity will allow. At first contact, my theorem will probably have appeared to you very complex. The process, in operation, should however appear fairly simple; but you would hardly expect so significant a fact as the communication of idea to be accomplished by the mere turning of a psychological crank.

I have obligated myself to show that an idea—once more, if you will allow me to repeat my definition, the image and the valuation of an experience—can be communicated by the vehicle of music. In the first section of my paper I offered an example of such communication—one which the composer had indisputably intended as an expression of a state of mind which is indicated by the word Trübsal. We recognized, I believe, that Bach’s passacaglia theme accomplished that purpose. That theme—which, as we saw, lacks no value ordinarily attributable to pure music—may now be explored in the light of our expressive theorem. If it seems to you that I have chosen too convenient an example, and one which takes my theorem for granted, I will ask you to remember that it was originally a purely instrumental piece, and that the association with the word Trübsal was established later. Bach perceived the association as valid. I shall try to show why he was right—not because he needs justification but because his judgment, analytically justified, will also justify our search for expressive purpose in other cases where association such as exists here cannot be invoked.

In the substance of this theme we are to observe especially two supposedly elemental facts, tone-stress and ideal motion. These, our hypothesis assumes, may be perceived as portrayals of two elemental facts of emotion as we know it—nervous tension and motor impulse. Amplified by normal associative accretions, the fragmentary suggestion given by the elements may assume the dimension of an idea.

The amplification is of course essential; for nervous tension and motor impulse alone do not form the substance of idea. They are merely phenomena characteristic of a mind in which an idea is present. But they are characteristic; and they may evoke interpretative inferences. We gained the idea of that which overwhelmed Paolo and Francesca merely by amplifying, inferentially, the fact that they read no more in their book. We should likewise be able to grasp Bach’s expressive purpose, if, in what is obviously a cogent musical idea, we attend to what I believe was also designed as a portrayal of the feeling appropriate to a given type of experience.

Perhaps I should say, before we begin, that I am not supposing Bach to have conceived the problem of musical expression in the terms in which we are now considering it. We are observing, analytically, a process which the artist executed synthetically, and probably with little or no thought of the process as such. Bach, as his adaptation of the concerto theme to the cantata proves, devoted the resources of his incomparable musical imagination to the utterance of what may also be conceived as a nonmusical idea; but he no more pondered the things we shall observe in the light in which we shall observe them than Milton pondered the principles of English grammar while he was composing the highly periodic sentence with which Paradise Lost begins. Bach, like any master of language, was utilizing resources of his language which were put there long before he learned it by other users of the same medium of expression. Let me emphasize that the facts of nervous tension and motor impulse were not invented by the psychologists but were only named by them. They are universal characteristics of feeling states, and to perceive the fitness of music for their representation, it is not in the least necessary to see the substance of music in our present analytical light. I did not invent the facts of tone-stress and ideal motion. They, also, have been there from the beginning.

In our observation, let us ignore, as far as we are able, the implications of Trübsal which we found in the music. Let us look, with active sensibilities but without any definite expectation, at what, merely as portrayals of tension and motion, our hypothetical elements of expression have to say. Our first question will be what sort of tension is exhibited by these notes, in those relations of height and depth, activity and rest, or consonance and dissonance, which we cataloged previously. Our second inquiry will explore, similarly, the characteristics of motion. Then, assuming that these are possible portrayals of tension and motor impulse within ourselves, our next question will be: What feeling-state do they characterize? And the last—if we need to go so far—will be: Out of what conditions of experience would such a state arise?

It would be more precise to take all these types of inquiry in the order in which I have stated them; but it would also be tedious. I think we may safely observe the notes or the phrases in their order, noting what each yields of tension suggestion. I shall restrain, also, my impulse to interpret these suggestions, but I shall sometimes yield to the temptation. You, however, will feel entire liberty of disagreement and, if you like, of verbal protest. We shall cover the ground pretty slowly, and this will amount to the magnification, by many diameters, of the values we perceive. But this magnification, without which many significant values might go undetected, we may rectify by thinking the music though at its normal rate. We shall then perceive, in its proper perspective, what the magnification has revealed.

The music begins with three rest tones. Their only hint of stress lies in the fact that they form a minor triad (faintly stressful by comparison with a major); secondly, that they rise; and thirdly that they rise very slowly. (This slowness is more obviously a fact of motion; but you will remember that we could not wholly divorce the fact of motion from the fact of tonal stress.) No very distinctive quality of stress is suggested by this upward progression. Indeed, we might gain from it an impression of calm, if it were not for the faint echo given by the second of the repeated notes in each beat—a hint of some impediment in the motion which, at this slow speed, implies effort or hesitancy.

But the next four notes forcibly intensify the whole character, and belonging as they do in the same phrase as the three rest tones, cause those notes also to be seen in this new light. The laborious climb of the rest tones (from which all superficial excitement is expunged by virtue of the fact that they are rest tones) ends in a precipitate drop—a descent far longer than was the whole dragging ascent. The drop is also to an active tone, so that the thud with which the melodic line falls to the E-flat, is perceived as a fact of motion; but its unexpectedness is certainly a fact of stress. The heaviness of this fall is nothing less than alarming; and it is the more so because we shall hardly have understood the E-flat, at the instant of the fall, as possibly grouped with the g-minor triad in a single, dissonant chord. If you now think these notes as forming a major seventh-chord, you will find that the force of the fall is much weakened; but you will also see that Bach meant you to feel its full force. And you can measure the alarmingness of the long drop by imagining the high D as rising to a higher E-flat, instead of falling to the lower one.

This active note E-flat at once descends (further emphasizing the length and weight of the fall) to a rest tone, D. But there is no sense of rest in that tone. It is the lower octave of the high D from which the musical body fell so precipitously; but that fall showed the sense of rest in the high note to be precarious, and the low D, displaying no sense of rest at all, consequently returns to the E flat as if it might find there something of rest, and on this note of uncertainty the phrase, with extraordinary vividness of suggestion, is silenced. Perhaps I only imagine the tension which persists throughout this silence as being portrayed by the silence itself. But even though it is determinably the product of the pause on an active tone, the silence is eloquent. In these seven notes, as we noted before, lies the underlying structural motive of the whole theme. We noted the consequent impression of structural coherence in the music; but we may now also observe that the logic of persistence extends also to the expressive sense of the music, which never departs wholly from this motive until the end.

Now, as if it had taken some hidden path in the silence, the melody emerges again, far above, on the active note A. Uncertainty or a kind of groping obscurity seems evident in this unexpected emergence. The active A now shows no disposition to seek a neighboring rest tone. Instead, it begins anew the pattern of effortful ascent, and the fifth of the triad (now the unstable diminished chord) is again followed by a longer fall. This time, however, the drop is cushioned by alighting on a note immediately associable, in a harmonic sense, with the preceding diminished triad. Accordingly, the ascent can begin again, and it proceeds once more in the short steps of the minor third. The next drop is longer—a minor seventh; but as it falls to the dominant and rises once more on the arpeggio of that triad, we almost sense stability in the midst of all the uncertainty that has so far ruled. But this suggestion is deceptive. Another drop of a major seventh makes the B-flat (really a rest tone) active; and the strangeness of this note, refuting the hint of stability in the dominant triad, is almost appalling. You will note that from the A on to this B-flat, the trend of the whole design is downward, with unusually forceful implications.

We observed previously that the eighth notes execute the same melodic curve as that in bar 2, but foreshortened and with no ensuing pause. This curve, whose motor energy is increased by the foreshortening, seems now to instigate a new condition of activity, in continuous eighth notes, and centered on the higher octave of the B-flat. This center is considerably stabilized by the repeated leading-tone progression, F-sharp–G (which, however, hardly sounds like a true leading-tone progression). Each swerve to G is followed by a fall—a motion fact which has by now acquired a pretty distinctive significance in the theme; but the successive drops outline a progression of three rising notes (C, D, E-flat)—the only varied notes in this whole momentary pattern—and this gradual but determined rise, combined with the more kinetic eighth-note motion, gives what I think you will perceive as a more purposeful sense than has hitherto appeared.

The E-flat, the goal of all this progression, is really an active note, but it has a considerable implication of rest—probably by association with the Gs and B-flats so often reiterated, and also because of the stoppage of the kinetic motion. Tonally, however, we are lost. We do not know whether a modulation has been made or not. And this indecision is at once terribly heightened by the weary resumption of melodic movement on the legato curve, G–A-flat–F, the descent into the dark obscurity of the diminished triad, and another long pause on the tonally indeterminate note, A-flat. This is the lowest note we have heard, so far, and you will agree that everything conspires to make its lowness impressive. In view of the whole character of what has occurred, nothing could be more unexpected than the firmness of the final cadence.

I have tried to speak of these notes in terms of their actual musical tensions—values which they possess for any experienced musical ear—and with little reference to the whole expressive sense we are still to seek. For that sense does not literally reside in the tones. It must be inferred from them, and inferred similarly (if not identically) by all of us, else no fact of expression will have taken place. I doubt that you will have disagreed substantially with my findings thus far. I suspect, indeed, that you may have gone much farther than I in the direction of inference.

But we have yet more facts to consider before we release our faculty of inference—the facts of what I have called ideal motion. These may prove, in themselves as suggestive as the facts of tension; but they must also be integrated with those facts, and this integration may considerably enhance the whole value of meaning which, through critically controlled inference, we may draw from the music. We have indeed made some mention of the value of motion, since that value is sometimes indissolubly bound up with active tension; but the theme must also be studied in its motor aspect primarily.

The motion is slow, and I think you will agree that it is heavy. There is no hint of that energy or assurance with which imagined limbs would move if they were impelled by a state of exaltation. The sudden alarming drops, viewed as motor facts, have the precipitancy of actual falling. This does not mean that we have here the physical representation of an unsuccessful attempt to scale a cliff, but we can hardly follow the persistent repetition of these descents without some awareness of earnest, yet always defeated motor purpose. The pauses, whether of actual silence as in bar 2, or on long note as in bars 9 and 11, also need not be visualized as pauses for breath during physical effort; but they do contribute to and define that motor purpose.

Although this theme is wholly melodic in texture, we have seen that the implications of harmony are frequent and compelling. Because of this texture, the music acquires none of those propulsions which harmonic accompaniment usually provides for melody, and which, whether of special harmonic significance or not, are ordinarily absorbed into our awareness of the melodic character. Yet we do sense in this music a peculiarly weighty mass, such as accompaniment usually provides, and no small portion of our total impression of motion is that of the momentum of this mass. The middle portion of the theme—that where the continuous eighth-note motion is present—yields, both in its more kinetic energy and in the rise of the three notes, C, D, E-flat, an impressive manifestation of this momentum. For the three notes represent three triads, rising toward a position of apparent stability (the long E-flat). But the music topples from that position, even during the continuation of that unstable note, and falls ominously, in spite of its futile rise on the diminished triad, to the low A-flat. The departure from the established patterns of motion which occurs at the cadence is in the highest degree striking; but that, I think, we need not now interpret.

Here, I think, are the principal facts of tone-stress and ideal motion. You apprehend them, in the actual process of hearing, simultaneously, and in a hundredth of the time it has taken me to describe them. But I have described them only to make sure that they are perceived as there—inherent in the stuff of this music. What do they mean? Their musical interest is great; but is this their only interest? And was an awakening of purely musical interest the real purpose of the composer? Dare we go outside the field of purely musical perception to look for more meaning in the field of human experience? Shall we do violence to the music by such a search? Or shall we do violence to the composer if we fail to make that search?

The result of this inquiry will be an inference, drawn from the facts of motion and tension which the music exhibits. The shape of our inquiry will be the question, what would be our mental state if it involved such tensions and such motor impulses as we have just observed in the music? And along with the awareness which will be the answer to that question we may combine another—the awareness, also inferential, of the external circumstance or experience out of which such a condition of mind would normally arise. Together, these awarenesses will comprise an image and a valuation of experience. The music has represented only a few features of that valuation and has left us to complete it, and to form the image of experience, by inference; but this reversal of the usual direction in which inference is drawn need not prevent our acquiring, through the medium of music, the full stature of an idea.

Even if we had not established, at the beginning, an association with the word Trübsal, I suspect that you, if asked to tell what state of mind these notes depicted, would have found at least an analogous word to that. I, at least cannot but feel in the slow climb of the first phrase and its ensuing dreadful drop the tensions and the sudden loss of grip, not of groping hands but of a spirit that labors and is heavy laden. I do not have to seek for this impression. I am compelled to it by the music; and I am not released from that compulsion by all that follows. The middle section, with its more active motion, does not alter the basic character, but it does add a sense of kinetic purpose, without which the earlier phrases would appear as a portrayal of despair. It is in the light of this section, I think, that the cadence, which certainly departs from the tone of all that precedes, begins to come clear.

Before we interpret that, however, I should like to make one more comment. I said I saw a spirit, not a body, that was heavy laden. I could easily visualize a body thus burdened and could manufacture a series of events corresponding to the various indications of movement and effort I have noted. But if I once began to impersonate these inferences, I should end by making up a story about the person I had imagined. Various things happen, as music goes on, and it is all too easy to take a sequence of musical details as implying a parallel sequence of actual events. But there is no real reason why this parallel should be recognized. A musical idea must, indeed, unfold itself in time; but that which is said last in music is not on that account the sequel in time or causality of that which preceded, and the disposition to make music into a story is an error, not merely because music does not portray event, but because it ignores this fact in its possibly justifiable inference of event.

The cadence of our theme, which portrays a very different vein of feeling from the rest, seems to me a case in point. If we were to interpret this theme in a narrative sense, we should have to accept the cadence as the reversal of all that had gone before—as a kind of deus ex machina, such as always appeared in 18th-century opera to straighten out impossible tangles in the plot, and figuratively to dry the tears and smooth the sensibilities of the aristocratic ladies who had suffered through the harrowing production. There is no need to do that. This music is not a story.

The end belongs to the beginning. However unexpected and dramatic it appeared, our very acceptance of its musical sense argues that somehow we knew that the end belonged to the beginning. Yet that cadence is no portrayal of tribulation. Was it, possibly, the conclusion toward which all the music pointed? If there had been in Bach’s mind only the image of tribulation, would he have given throughout so unmistakable an intimation of strength? Is there not, along with the persistent slow climbing and the sudden loosening grip, an equal intimation of indomitable endurance? Was I not misled into a sentimental preoccupation with the obvious when I called the music the very stuff of tribulation?

Bach was thinking of more than that. The word Trübsal was not, for him, the key word of his text. The real sense is at the end, both in the music and in the verbal phrase; and in both the end brings the rest into focus: Wir müssen durch viel Trübsal in das Reich Gottes eingehen.

Notes

1. Poets, wisely, use the same method for the evocation of feeling. Compare, for instance, Goethe’s somewhat ineffective line, a description of feeling itself:
   “Es schwindelt mir, es brennt mein Eingeweide” with Coleridge’s image of a revolting reality:
   “And slimy things did crawl with legs
   Upon a slimy sea.”
   and compare your bodily responses.
2. This comparison, extendedly pursued, would lead to another problem—that of taste. This problem is so intimately related to that of expression that it cannot be ignored, even though it is too large for more than brief mention here.
   What is taste? It is obviously a product of experience, and takes the form of a complex habit of choice. But the word taste is also our name for one of the five senses; and this implication—of something given before our individual experience began—still remains when the field is transferred from that of more sensation to that of artistic enjoyment. A cultivated taste is generally thought of as one which discriminates finely in the field of pleasure. In that view, pleasure and its refinement would be the true object of all the long effort of cultivation. A philosophy of pleasure, which is called hedonism, is the possible product of this pursuit.
   But it seems it me that such a philosophy ignores one of the primary functions of taste. That sensory function was indeed given for discrimination, and was also, from the first, doubtless concerned with pleasure; but primarily it was given for the discrimination, not of that which is pleasant but of that which is nourishing. And this holds, whether the question is of the body or of the spirit. Appetite and taste are obviously related—again in both these fields—and I need not describe the disasters, physical or spiritual, which result from overindulgence. Age has also its influence on both appetite and taste, as have the many idiosyncracies of body and mind which are built into us and which we cannot resist. Even allergies—inexplicable bodily repudiations—have their parallel in the spiritual field; and since there can be no agreement about the things to be discriminated, for or against, the old Latin proverb (which says, not that tastes cannot, but that they ought not to be disputed) still holds.
   But if the question of nourishment is allowed to enter, and particularly that of the nourishment of the spirit, the adjective “good” begins to expand its meaning beyond that fairly large area which it covers, even at the sensory level. The spirit has an appetite for beauty; and this appetite, cultivated under favorable conditions, may rise to a point where beauty, as the satisfaction of a discriminating taste, may be called the highest Good. This, again, is hedonism—or, if you prefer a more palatable name, is eudaimonism. But it seems to me that the possibilities of nourishment, where pure beauty is proposed as the spiritual offering—are here not wholly admitted. If they are admitted, however, the word good will again expand its meaning. The antithesis between good and bad will not merely imply the distinction between the beautiful and the ugly but will begin to imply also the distinction between good and evil.
   The lover of pure beauty has an allergy for this implication. But if we explore again the implications of Bach’s theme, I, for one, cannot see how the question of moral value can be wholly excluded from the field of reference. Neither have I the least objection to its presence. The music has reference to the experience of tribulation. That experience—as St. Paul, and after him, Bach, conceived it—is unthinkable in a world where goodness is beauty and the pleasure of beauty only and evil is ugliness and the distress that ugliness brings to the spirit enamored of beauty.
   I am not contending that it is the business of art to moralize. But if experience, as we have been thinking of it, is the proper field of art, then I cannot see how the moral implications of any fact of experience can be excluded from our final valuation of the work of art.
3. On the etymology of this word, and also on its ancient meaning, see Werner Jaeger, Paideia, I, 125 f.

From The Musical Heritage of the Church, Volume VII (St. Louis, Mo.: Concordia Publishing House, 1970). Copyright Concordia Publishing House. Printed by permission. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Concordia Publishing House.