Stapled booklet, no cover, glossy paper. Transcription of lecture given at Black Mountain College September 4, 1944.
BLACK MOUNTAIN COLLEGE Volume 3 Number 2 December 1944
THE COMPOSER AND THE INTERPRETER by ERNST KRENEK
Lecture given at Black Mountain College September 4, 1944
The very fact that an institute like the present one, with its imposing array of experts in the various branches of music, is devoting its work to the interpretation of music, proves that an awareness exists in regard to problems coming into being when the performer sets out to make audible the music handed over to him by the composer.
In order to understand the nature of those problems, it will be profitable to turn for a few moments toward history. The further back we go in time, the less we observe of the problems besetting nowadays the art of interpretation, until we reach a point at which no problem whatever exists, because there is no difference between composition and interpretation. In the prehistoric days of music, so we have to assume, the art of composing, that is inventing organized musical sound, was one with the art of interpreting this invention, that is emitting the sound by vocal or instrumental means. In modern terms we call this procedure “improvisation”. For the understanding of our present problems it is necessary to keep in mind that for a very long time this procedure has been, consciously or unconsciously, the ideal towards which all musical activities and disciplines were converging. Music was the organized sound which was heard at the moment of the performance. I expect at this point some astonishment at my putting this idea in the past tense- music “was”. Most people still think that music is what you hear, meaning that all the other activities necessary to produce the performance, including the act of writing down what is to be played, are of a preparatory nature. This is undoubtedly true, for even those musicians who are able to hear a piece of music to some extent inwardly when they look at the score, simply imagine an actual hearing. However, the cause of our problems is the fact that the preparatory act of writing down the music in the course of history has taken on a constantly increasing significance of its own. In short, music has become literature; that is, the written score has meaning, aesthetic value and the full dignity of an art object, no matter whether, or how often, it is made audible, and it retains all those qualities even when it is only sitting on the book shelves of the library.
I think it will be considered in keeping with the democratic spirit of this college if I take the liberty to disagree with one of the points made by Mr. Charlot in his recent lecture on Abstract Art. His discourse seemed to me to imply that music was less fitted to associate itself with timeless matters than painting, because music was taking place in the dimension of time. As one of my students has put it quire philosophically, there seems to have occurred a confusion between the existence an the experience of music. It is true that music when performed is experienced in time. But so is painting, for nobody looks at a picture for centuries, but only for a limited span of time, usually less than it would take him to listen to a sonata or symphony. And just as the painting is still three even when nobody looks at it for any length of time, so is music still existent even if it is not heard. This condition certainly has never been felt by composers to imply the necessity for occupying themselves with ephemeral matters only; on the contrary, they always have thought that their art was particularly adapted to convey timeless thought.
While in olden times the music sheet was merely a mnemonic aid for the musician who combine in himself the quality of the inventor and the interpreter of the time, things became different as soon as a division of these qualities occurred, a specialization of functions, because some individuals appeared more fitted for writing music and others more capable of performing it. Still, as long as music was solidly and clearly tied up with circumscribed ecclesiastical and social functions, the composer, no matter how original, personal or even bold in technical innovations he may have been, was still supplying material for a musical manifestation that was expected to follow an established stylistic pattern and virtually could have been taken for an enormously elaborate improvisation on the part of the singers. Actually even counterpoint, according to prevailing views the most rigidly organized form of music, has been improvised by mediaeval singers, a game known as Supra Librum Cantare, when the participants in the exercise were given a cantus firmus, and each of them added a new voice to it, according to the rules which he knew.
However, the process of dissociation, separating the functions of composer and performer, went on and gained special momentum after 1600, when the center of gravity in music shifted from the set ritual of ecclesiastical services to the free and boundless field of secular thought. From then on the composers increasingly wanted to make their writings expressive of their increasingly personalized ideas as individuals. The details of this evolution belong in the field of general history of music, and a discussion of them would lead much too far in this context. For the rest, these things are more or less well known.
The essential consequence of this process is that now the performer is given a material that is not any longer supposed to sound like his own invention at the moment of the performance, but has meaning and dignity in itself before it becomes audible. The problems involved arise from the fact that the performer, according to his very nature, which I don’t have to analyze psychologically in this circle, still strives at making the material sound as if he were creating in the act of performing it, and I wish to go on record here to the effect that he is absolutely right in this aspiration of his.
The composer, even if he acknowledges and respects this state of affairs- which many composers are not ready to do- feels the need of making his own intentions as clear as possible, be it only for practical reasons. In the first place, he can hardly rely on established stylistic conventions any longer, since such conventions gradually disintegrated during the nineteenth century when originality and uniqueness became more and more desirable as principal attributes of the work of art. Furthermore, music is now as a rule being written not for the purpose of one particular performance, but at least theoretically for an unlimited number of performances removed from the control of the composer in both time and space.
Let us see what means are at the disposal of the composer for transmitting his ideas to the interpreter of his works. Obviously these means are almost exclusively graphic signs. The oldest and most elementary of those signs are known as notation, designating originally pitch level and later also time value of the notes to be played or sung. Looking at our present system of notation, it seems so obvious that we are almost at a loss to understand why it took many centuries of long and tortuous detours to evolve that system. Of course, these detours were by no means due to the ignorance or incapability of earlier musicians, but they were very logical in the light of the purposes that notation had to serve in various periods, according to the character of the music, and that is what makes the history of notation a fascinating field of study, by no means limited to purely philological research.
The first thing for which we are nowadays used to look if we want to find out how to perform a piece of music is the tempo, the indication of the speed. Historically, however, this question did not arise until very late. It is easy to assume that mediaeval ecclesiastical music was meant to be sung in one certain base speed that everybody knew. Yet it would be wrong to believe that no variations of that speed were used. Willi Apel, in his book on the history of notation, has brilliantly and in my opinion convincingly pointed out that the so-called proportional system, which was dispensed with by so many earlier musicologists as one of the craziest bubbles produced by scholastic brains, actually served the purpose of indicating variations of speed. According to the tenets of the period these variations were expressed in terms of proportions referring to the all-pervading basic speed. The artistic and intellectual boldness of some of the mediaevalists went far enough to make them use such fantastic directions as, for instance, that in one voice of the contrapuntal fabric after a certain point 13 units should cover the same time span as did 9 units before that point. Taking into consideration that the other voices would still proceed in the old speed, rhythmic combinations would result of a subtlety and complication that makes the dreaded rhythmic complexity of modern music appear fairly simple. It indicates an entirely different and by no means inferior way for musical thinking.
The modern equivalent of the proportional system is the metronome, by which the varying individual speeds of the music are expressed in relation to the abstract unit of one minute. In the eyes of the composer this should be a great improvement over the vague descriptive indications like Allegro, Andante, and the like. For reasons to be discussed later, the improvement brought on by the metronome is not so unequivocal as one may think.
Another objective of graphic indications approached by the composers still later and with increasing solicitude is dynamics. Not only does the notation in itself not suffice any longer to suggest to the performer what should be loud and what soft, but also the meaning of the music increasingly depends on a very accurate distribution of dynamic shadings. This purpose is achieved by a great number of symbols known as expression marks, which in the romantic period were more and more supplemented by descriptive adjectives, adverbs and whole sentences aiming at conveying extramusical associations of a poetic nature which the composer wanted to evoke in the performer and eventually in the listener. This procedure implies a fair amount of confidence in regard to a generally shared pattern of emotional reactions and to the automatism of psychological processes. If the composer indicates that a certain phrase is to be played “with determination” and another “with tenderness”, no technical directions as to the manner of playing are involved. But he hopes that the player, if he generates in himself the attitude of determination or tenderness, will hit upon the technical means necessary for the desired effect. I shall speak later of still another type of graphic signs that the composers use in order to clarify their intentions. This refers to phrasing and articulation. If this is as far as the composer may go in establishing a text as clear as possible, let us now see how far the interpreter may go in making this text appear his own creation, or nearly so, at the moment of the performance. It will be easy to dispense with the extreme cases in which the performer does not shrink from arranging or paraphrasing the given text by putting in features which would allow him to indulge in mannerisms in which he knows himself to be particularly successful. They may reach the naïve habit of dwelling unduly on high notes on the part of certain singers to actually writing down special versions, concert arrangements and paraphrases of the given material such as have been left by Franz Liszt and other virtuosos. The more elaborate such arrangements become, the more they take on the character of compositions in their own right and become in turn objects of interpretation. I don’t wish to go into the problem as to whether such manipulations should be condemned under all circumstances, or whether they may be justified in some cases. It is well known that in the 19th century in general few inhibitions were felt in this respect, and nowadays we shudder at the thought that, for instance, Bach scores were amplified with clarinets and trombones. However, we don’t feel any such misgivings playing the Well-Tempered Clavichord on the modern piano, in spite of the tremendous difference between that instrument and Bach’s harpsichord.
While one may readily admit that the listeners to the romantic arrangements of Bach’s music received a picture far remote from that which the original score conveys, we must also admit that Bach’s music has successfully survived any kind of manhandling inflicted upon it, including the swing arrangements which caused such a fuss a few years ago, and if for nineteenth century audiences the only possibility of hearing Bach was to receive him through the medium of the teste of the period, it was perhaps better than nothing at all. It seems plausible enough that each period perceives the art of ages past in the light of its own aesthetic tenets, and the honest efforts made in our own time to get as closely as possible at the originals may involve as great a number of errors as the innocent enthusiasts of the romantic school committed, who, as far as they were honest, were just as convinced as we are that they did their best to serve the true intentions of the great masters.
We may at this point break up the discussion of these affairs, for they transcend the field of what we wish to consider here under the heading of interpretation: that is, the actual rendition of music previously written. In other words we shall confine ourselves to analyzing the problems of the interpreter who wants to carry out faithfully the instructions given to him by the composer through the system of the graphic symbols discussed before. As a reaction against the era of the virtuoso who used to manipulate the compositions coming under this hands according to his fancies, we have no established the idea of work-fidelity; that is, we consciously try to carry out the wishes of the composer. That seems to be simple enough a proposition, since apparently we have just to obey the graphic signs employed by the composer, and at this point I wish to narrow down still further the scope of this discussion by focusing on those cases in which the composer to the best of his knowledge has put down all the indications that seem to be necessary in order to insure the proper rendition of his work. In general this will apply to contemporary music, as most composers today are aware of the problems which we are discussing here, so that we ultimately will examine the relationship of the living composer and his contemporary performer.
It will appear that the advice just to obey the directions put down in the score does not solve all problems. A comparison which dramatic literature will elucidate the point. Take the following lines from Hamlet’s monologue:
“To die: to sleep
No more; and by a sleep to say we end
The heart-ache and the thousand natural shocks
That flesh is heir to, ‘tis a consummation
Devoutly to be wish’d. To die, to sleep;
To sleep; perchance to dream: ay, there’s the rub…”
Obviously there are a hundred ways of speaking those lines, as far as pitch level, distribution of accents, speed, number and length of breaks are concerned. The repetition of “To die, to sleep,” ad the immediate repetition of the second “to sleep” clearly require some special treatment in regard to expressive nuances, and doubtless many different possibilities could be chosen and defended. If we would demand that of all these possibilities the one and only should be chosen which Shakespeare had in mind, we would be quite at a loss to decide which one it was. Shakespeare does not give the slightest stage directions as to the mood succession of the monologue. Our only guiding line is the text itself, the succession of thoughts expressed in the words and the context of the dramatic action, from which we have to draw our conclusions as to the state of mind of Hamlet at this particular point. We may try to get some light through examining why he says “To die, to sleep” instead of putting the stronger alternative later. When he repeats “to sleep” for the third time, is the increment of emphasis due to the growing attractiveness of the idea? Or may, on the contrary, some doubt have crept in as to whether sleep would be so desirable at all, so that a slightly interrogative inflection would seem in order? “To die, to sleep; to sleep- to dream perchance? To be sure, Shakespeare did not put any question mark there, since that punctuation sign might be much too strong for the subtle nuance suggested by the following “Ay, there’s the rub…” by which the idea of sleep is finally discarded. Which one of these alternatives is completely true to Shakespeare? I don’t think that any amount of research will give a definite answer.
Returning to music, we face somewhat similar problems. Although a musical score is infinitely more exact than literary text in regard to the elementary facts of pitch level, rhythm, dynamics and the like, there is still a wide margin left for interpretation.
To speak for a few moments of one of the most elementary factors only, the tempo, we will have to face the fact that even the most accurate manner of indicating it, that is by metronome figures, is not quite unequivocal. In the first place, the ideas of the composer on a certain tempo of his own composition are not at all times absolutely the same. This does not imply that he is muddle-headed or that he is unable to make up his own mind. It only illustrates the elementary fact that a work of art, in spite of its inexorable logic, or rather beyond that logic, has various and varying aspects. I would even go so far as to maintain that the number of possibilities in which a work of art may be interpreted convincingly is an indication of its greatness. Only small and insignificant things have only one aspect, allowing only a single interpretation. Secondly, using the metronome in order to specify the tempo of a composition just written requires an amount of self-denial and insensitivity against noise that only a few hardy individuals can muster. Everybody who has played with, or rather against, a metronome will know that after two or three measures he either gets off the beat, or he has the strange feeling as if he were playing under an anesthetic, not being quite aware any longer of the music he is trying to perform. Obviously based on this experience was a very wise remark that Beethoven attached to one of his own metronome markings to the effect that it ought to be obeyed strictly only for the opening measure of the piece. I am indebted for this information to my friend, Nikolai Graudan. Thus, if a composer writes a metronome marking like, for example, “quarter notes equal 96”, it means probably that 78 would be far too slow, and 120 much too fast, but according to circumstances 88 might be just as admissible as 102. From my own experience I know that in several cases when interpreters suggested to me a tempo somewhat different from the one I had indicated, I was quite satisfied with their version. It is well known that Wagner after some disappointing experience with interpreters who followed faithfully his metronome markings decided to dispense with that kind of indication altogether. Similar things could be said about the relativity of dynamic directions and so forth. All this points to the fact that even careful indications on the part of the composer leave a substantial margin to the interpreter.
Technically speaking, the filling out of that margin is largely assigned to processes which I like to sum up under the heading of articulation. In the set of definitions which I contributed to the Dictionary of the Arts to be published by the Philosophical Library I offered the following about “articulation”: “devices of composition aiming at clarity of design; purposeful arrangement of musical elements according to the functions which they have to fulfill in the context, especially with a view to properly balancing their metrical weights; phrase construction.” I believe that this definition applies to the work of the composer as well as to that of the performer, and according to my experience the main problem of interpreting contemporary music lies with articulation. Of course, necessity of articulation is by no means limited to modern music, and the excellence of a performance of old music depends to a great extent on clear and intelligent articulation. However, while classical music by and large remains intelligible even if this point is badly neglected, modern music becomes perfect nonsense if not properly articulated. The reason for this is that the basic facts of the tonal idiom, the standard relationship of chords, are familiar enough to make the elementary structure of such a piece perceivable even if the deeper sense remains hidden. In new music written in an unfamiliar idiom the material itself, the strange chords, unusual progressions, the dissonant intervals and so forth, do not evoke any associations with previously known musical features, so that making any impression at all on the listener depends entirely on the eloquence of the rendition, which is more than mere accuracy.
I am using intentionally the term “eloquence”, because it touches at the sphere of rhetorics. The means by which music becomes articulate are closely related to the linguistic devices that impart to speech the sense-making logical, and expressive shadings. The first requirement here is to discriminate between elements of first importance and such of subordinate significance. Naturally in any good composition every single note is necessary and therefore important, but in any kind of context, be it a combination of simultaneously sounding tones or voices, or a succession of phrases and sections, certain things will have to stand out against certain other things, or else the effect will be muddy and dull. While in homophonic music it is usually easy to discover the leading melodic line, it is not so in a complex fabric of several such lines where the emphasis may shift quickly and frequently from one to the other. The composer may try to indicate the differences in significance by various graphic means, but it will be the performer’s tasks to grasp intelligently the proper shadings of emphasis.
As far as the structure as a whole is concerned, it is just as imperative that the performer be aware of the structural function of the different sections, for it is here that the composer can guide him only in a general manner. Everyone knows that a thematic statement has to be played differently from a transition passage, but the problem is how to tell a statement from a transition. In my opinion the secret of a perfect performance consists in making all the details so lively and variegated that the interest of the listener is attracted by what happens in each moment, and at the same time keeping the continuity of the broader span in evidence. Both tasks are the business of articulation. The performer should act as guide that takes the listener through a new landscape, inviting him to stop and look around at certain points, enjoying the nearer surroundings, then again to keep going steadily until a new point of special interest is reached, preparing him for the strenuous exertions of a climax, making it clear to him that an area of relaxation will follow, or that the end of the journey is near. All this is written down by the composer, to be sure, but most of these things cannot be expressed through special descriptive symbols; they are embodied in the music itself and must be brought to life by someone who understands the musical process.
I said in the beginning that I thought it was the legitimate desire of the performer to make the music appear as if it were created by him in the act of performing it. This he will achieve only if he re-enacts mentally the work of the composer. Composition consists largely in treating, according to certain principles, material furnished by the creative imagination, that is, It is an action going on the spheres of both freedom and discipline. These spheres are not neatly separated, that is to say: the invention of the material is not entirely free from assumptions, it always is based on some pre-established selection of basic elements, and the treatment to which the material is subjected according to certain principles is not at all devoid of the element of imagination, for in the first place the principles are the expression of a more inclusive artistic vision, and their application allows for a great number of equally good and aesthetically satisfactory variants. In order to strike the balance between vague and vain virtuoso business and dry and deadly pedantry, the performer ought to put himself in the place of the composer, trying at any given point to reconcile the impulses of his imagination with the principle that he has set up for himself, that is, to obey the prescriptions of the composer as well as he can.
Many composers live under the impression that interpreters are their natural enemies, almost as bad as publishers and critics, and they think that the only remedy of the evil would be if the interpreters would stick more faithfully to the text, and nothing else. In his book Poetique Musicale, which is a collection of his Harvard lectures, Igor Stravinsky takes great pains to make this point, admonishing his interpreters to forget all about their own individuality and to obey blindly the directions of the composer. In the case of the highly mechanized type of music that Stravinsky writes and advocates such an attitude on the part of the interpreter may indeed be most conductive to producing an adequate picture of the art object. But in mind other cases the composer insisting upon faithfulness to the best would seem to be carrying coals to Newcastle, for contemporary interpreters are anyhow more than intent upon playing what’s written on the music sheet, and nothing else. Paradoxically as it seems, I venture to say that if there is any trouble with interpretation, it is mainly due to the fact that interpreters are trying too hard to suppress their own imagination in rendering music. Of course, giving rein to one’s imagination is not identical with not paying any attention to the composer’s indications. But you remember that I decided to talk on those interpreters who are above the level of neglectful ignorance.
A story was passed around about the great German playwright, George Kaiser, who when he was pestered by a certain stage director with requests to permit him to change one line of Kaiser’s play and then another line and still another line finally lost his patience and wired back: “Change anything you want except the title of the play and name of the author,” for these were the only two things necessary in order to collect royalties for the copyrighted item. I am not suggesting that a composer as a rule should take so cynical an attitude towards what happens to his work, but my own experience has shown me that composers frequently are much more tolerant in regard to interpretation than the interpreters are themselves. Composers who think that there is only one single way of interpreting their music could perhaps derive some hope from new developments in the province of electrical instruments. In my book “Music Here and Now” I outlined an interesting situation in which the acts of composing and interpreting might again coincide in a manner entirely unforeseen. Even in earlier times composers, such as Mozart for instance, wrote pieces for mechanical instruments, but in these cases the composition was written in the usual way on paper, and afterwards transferred onto the mechanical device by engineers. Only in our time a few composers like Hindemith have tried to cut their own rolls for a pianola. Now theoretically it should not be impossible to cut a phonograph record without the music’s ever having been played beforehand. However, it seems more than unlikely that a proper mechanical process could ever be devised for producing the necessary grooves accurately. It seems much less unlikely that a contrivance could be invented which would enable a person to design a sound curve onto the light-sensitive film of a sound track. How is could be done, I don’t know. It would certainly require infinite study to manage properly all the finesses of such a complete curve in order to get in the full range of tone colors and dynamic shadings that we want to apply in our music. Obviously we would eventually be able to obtain unheard-of tone colors and sound combinations such as could never be arrived at on any of our present instruments. Certainly one could not draw such curves with a clumsy pencil, but perhaps the physicists would present us with a maneuverable light beam that could leave the desirable traces on the film. The point that seems interesting in the context of this discussion is that the composer would obtain an authentic interpretation by eliminating the interpreter altogether. His act of writing down the music by means of the magical light beam would not produce a set of graphic signs to be read and interpreted by someone else; its result would be the sound-producing device itself. The performance would not contain a single element that was not put there by the composer.
That sounds very interesting and attractive for purists, but the essential question is whether the composer is really interested in such a type of unique and unchangeably set performance, or whether he should be so interested. It Is true that the composer is, and ought to be, convinced of the uniqueness of his message, and he naturally is most anxious to get it across to the listeners in undistorted and unadulterated fashion. It is also natural that he might be overanxious in this respect, seeing that he is practically helpless as soon as he has handed over his music to the interpreter. During the act of the performance in which his message undergoes the supreme test of facing the reaction of those for whom it was destined, the composer is a passive onlooker. However, he should have confidence enough in human nature so as to enjoy rather than to fear the medium of personal life through which his message is filtered before it reaches whom it concerns. The personality of the interpreter is not necessarily a stumbling block on which the work of art goes to pieces, although in only too many unfortunate cases it may be one; in the good cases which ideally should be the rule, that personality vouchsafes an increment of vitality that is not only desirable, but truly necessary in order to put the message across. The interpreter is the first with whom the composition has registered; he, as its first recipient, has had an experience akin to that of the listener, only makes times more intense. This enables him to communicate the message to the audience. If composers frequently despair of the possibility of such desirable conditions, it is due to the fact that interpreters of contemporary music more often than not are incapable of fulfilling their assignments. I don’t wish to duplicate the efforts of my friends and colleagues who have worked the better part of this summer here to clarify these problems and to suggest remedies for the shortcomings of the art of interpretation. The composer may contribute to the improvement of the situation if he takes the trouble of teaching the interpreters, making them acquainted with the nature and meaning of the creative processes and showing them how a composition originates and of what it is made. I am glad that Black Mountain College has offered me the opportunity of discussing some problems of compositions with the young performing artists assembled here, and I hope it has been to their benefit as well as to that of the art that is so close to our hearts.
