The past three issues of the Journal of the American Musicological Society have carried installments of a spirited dialogue about the need for Criticism in History.2 This session of The College Music Society meeting is to consider the case for History in Criticism. I would suppose that the two questions are the same.3

The title of this paper would suggest a manual of instruction—Vegetable Gardening in an Urban Environment—or of description—The Polish Peasant in America. In that I read the title with a question-mark I am bound to deny both implications, and what they suggest further about the author of such a manual, that he is a specialist. The subject is the concern of every member of this profession, for each of us acts on decisions about it, whether or not those decisions have been brought to consciousness. But to address it directly is to open a Pandora's box, and I can hope to do no more than consider some of the many divisive problems on both sides of the question that fly out as we do so.



I should like to frame the question first in the context of another art, by reading from two appreciations of Michelangelo's Pieta.

The first:

To this work I think no sculptor, however distinguished an artist, could add a single grace, or improve it by whatever pains he might take, whether in elegance and delicacy, or force, nor could any surpass the art which Michelangelo has here exhibited.

(Now the author expands on that word, Art.)

Among other fine things may be enumerated—to say nothing of the admirable draperies—that the body of the dead Christ exhibits the very perfection of research in every muscle, vein, and nerve, nor could any corpse more completely resemble the dead than does this. The limbs are affixed to the trunk in a manner that is truly perfect; the veins and pulses, moreover, are indicated with such exactitude that one cannot but marvel how the hand of the artist should in a short time have produced such a work, or how a stone which just before was without form or shape, should all at once display such perfection as Nature can but rarely produce in the flesh.

The author is Georgio Vasari, apprentice and biographer of Michelangelo.1 He expresses an attitude that we may take to be Michelangelo's own in placing the highest value on the recreation of the intelligible world through the mastery of anatomy and perspective.

The second passage is from a study by the great German Art Historian Henrich Wölfflin, published in 1891:2

I shall begin with the formal considerations. The first impression is always a sense of astonishment over the luxurious display of folds in the garments: the folds of the robe sinking between the knees, a piling-up of cloth over each knee, resulting in a kind of framing recess on the left knee, the same motive repeated over the left shoulder. . . . Between the fingers of the limply hanging arm the artist has placed a single fold from the Mother's robe. There is something touching in that, as though a flicker of life were still playing about the corpse.

The passages reflect a profound contrast in the standards from which these evaluations arise, one of them belonging to the artist's time, the other to our own.

The observations of both writers are accurate. With the exception of Wölfflin's last sentence they may be verified empirically. What is more, all the data gathered from both observations must have been available to each writer. Vasari, working in the Master's studio and having himself achieved some distinction as an artist, surely knew something about the planning of the formal arrangements. Wölfflin surely recognized the accuracy of the reproduction from life. The point is that each took for granted what was for the other most valuable.

We can readily understand why this should have been the case. What Vasari hailed as Michelangelo's great achievement represents a final triumph of the eye and the hand after long striving. Once achieved it could be repeated with relative ease, and it became a routine part of the artist's craft and the viewer's experience. In Art and Illusion Ernst Gombrich wrote "The history of art may be described as the forging of master keys for opening the mysterious locks of our senses to which only nature herself originally held the key. Like the burglar who tries to break a safe, the artist has no direct access to the inner mechanism. He can only feel his way with sensitive fingers, probing and adjusting his hook or wire when something gives way. Of course, once the door springs open, once the key is shaped, it is easy to repeat the performance. The next person needs no special insight—no more, that is, than is needed to copy his predecessor's master key."3 What is true for the creative skill of the artist applies with equal force to the eye, or ear, of the viewer or listener.4

Wölffiin's—and our own—emphasis on form and its bearing on expression proceeds from quite another "discovery." This had already been enunciated by John Ruskin in 1857: "Affection and discord, fretfulness and quietness, feebleness and firmness, luxury and purity, pride and modesty, and all other such habits, and every conceivable modification and mingling of them, may be illustrated, with mathematical exactness, by conditions of line and colour."5

We may generalize from these passages that the analysis of works of art rests upon a selection from all the true things that may be said about them. And the evaluation of an analysis devolves upon the awareness and evaluation of the grounds—the standards—on which the selection is made. The question of the "correctness," or the "validity" of the analysis is, compared to that, trivial. In turn, we understand those standards best when we see them in the context of the history of aesthetic judgement. We ask about the meaning of the work. But we ask, too, about the meaning of the standards on which it may be judged. And each question illuminates the other.

This case suggests a complex of preliminary conclusions and premises that I should like to review before I proceed any further.

The notion of an "independent," "objective," or "contextual" analysis, based only upon observation of the work itself, and framed in categories and concepts suggested only by the work, is an illusion. The work of art has no existence apart from any interpretation of it.

The works we value most are those that continue to interest us through shifting interpretations. We cannot simply make a categorical decision for one kind of interpretation because it carries the authority of the artist's standards, or for another because it has the greater meaning—for us—of our own standards.

We are interested in the continuity of the work through a multitude of interpretations. This is a question of Criticism, and of History, and of the History of Criticism. In that these subjects constantly illuminate one another, no one of them can be pursued fully in the absence of the others. They constitute, then, a single broad subject.



I should like now to approach the question of "Music Analysis in an Historical Context" from the standpoint of the purpose for which the analysis is undertaken, for each purpose will suggest its own criteria of value. I shall suggest four broad possibilities that do not necessarily exclude one another, but that differ in important ways as points of departure.

1. The establishing of the "hard and irreducible facts" about the musical work, the laying bare of its events.

With this as a goal, the analysis of music may be said to be a search for consistency of practice. It would seek to distinguish the significant from the fortuitous, design from accident, or, as it is often put, from "coincidence."

This is a safe formulation. It may be translated into a formulation that is, nowadays, far less safe: The analysis of music is the attempt to reveal the intentions of the composer. If we were to put it that way we should stand accused by some of having committed the "Intentional Fallacy,"6 for, they would argue, the intention of the artist is in principle not available, nor is it relevant, as a standard for judging his work.

But the two formulations are in essential respects equivalent, for in both the emphasis is upon consistency and design. The difference reduces to the question whether, having observed consistencies in the work, we care to attribute them to a human agent; or whether design in a work is taken to be the result of purposive behavior on the part of a human agent. This is probably what positivists call a "pseudo-question," like the question whether the rubber-band, having been stretched, may be said to have recalled its original form, just because it has been observed to return to that form.

To telescope the argument: let us read, for intention, significant design, circumventing the unanswerable question of the agent. Then historical evidence is relevant to the determination of criteria of significance. If we can recognize, on whatever evidence, what the consistencies are, or are likely to be (what the composer is likely to have intended), then we shall have one kind of standard for the analysis of the individual work of music. In these circumstances the "intentional fallacy" charge loses much of its substance.

I should like to insert as one example of the possibilities the insights which Allen Forte has derived from the sketches for Beethoven's opus 109, in his monograph The Compositional Matrix.7 This is a promise of what lies ahead once the vast fund of Beethoven sketches is subjected to analysis. But Forte himself denies that even the sketches provide adequate background. He writes, "Any technical study of Beethoven's works must recognize that his compositional technique cannot be understood apart from certain concepts of musical structure which reached a definitive stage of development about a century before the composing of opus 109. Many of these concepts are expressed within the practice of thorough bass. . . . Unless certain of the more basic concepts implicit in thorough bass are grasped it is almost impossible to cope with the complex structure of composed tonal music at any other than the surface level."8

In this study of a single work, the author appeals, not only to the principles of order that are manifested in the work itself, but also to evidence on two further levels that would direct us to a more sophisticated understanding of the functioning of those principles.

2. The interpretation of the musical work. With this as a goal, analysis is a search for the values and schemata that condition the apprehension of works of music, for values enter into the apprehension of facts just as facts enter into the determination of values. The interpretations of the Pieta demonstrate well enough the extent to which the reporting of objective facts so called—in this case the physical characteristics of the sculpture—is dependent upon the values of the reporter. We find meaning, expression, and purpose in works of music through an interpretive transformation of facts.

If facts have to do with knowledge, values have to do with preferences. One deals with "what is," the other with "what ought to be." The determination of preferences provides another kind of standard for the analysis—on this level we might say "criticism"—of music. If the standard which conditions our apprehension of the work differs from that which conditioned its creation, we can hardly afford to ignore either one for the other. For each can be recognized and can take on meaning only in the context of the other.

3. The explanation of the musical work. With this as the goal, musical analysis is a search for the causes of works of music.

We seek a balance of understanding between the work considered in itself and the work considered as the resultant of a multitude of forces outside itself.

A truly discerning analysis, in the sense of the first of my suggested categories, is a manner of explanation, for it reveals the significant events of the work and the manner in which they are related to one another. Events and relationships may, then, be regarded as causes for the whole work which is the outcome. In this sense "explanation" may be read as "understanding."

But perhaps to the polarity of the significant and the fortuitous there ought to be added a third category, the normative. This would refer to those aspects of a style that may be found in every work belonging to it—hence that cannot be considered fortuitous—and that belong among the données of that style—hence that would be significant in the characterization of the style, but not so much so in the characterization of each piece. If we recognize such a distinction in principle, it can be given substance for any single work only by reference to other works of similar provenance, and to non-musical documents that give evidence about what was normative.

Of course we do recognize such a distinction, but thereby we give ourselves the license to avoid the responsibility of understanding works in themselves and to explain them instead through the identification of conventions and antecedents. This has sometimes had two very serious and unfortunate effects: first, the acceptance of superficial resemblances as evidence for genetic relationships; second, the abstraction of conventions from the works themselves, and their re-constitution in the mystical notion of a "style" with a life-cycle that follows its own immutable laws—from zygote to tadpole to frog. It has sometimes meant, not explaining, but wholesale explaining away. This is surely one of the most sensitive questions bearing on the relation between Criticism and History.

The misuse of the concept of the normative is responsible for the hostility that the study of conventions—indeed that the so-called "historical approach" itself—has engendered in many corners. Still, if our aim is the recognition of what is significant in individual works, it is difficult to imagine how we can do without that concept. To clarify my meaning I should like to cite one illustrative example. In an essay on Dufay's tonal system9 I called attention to one interpretation of melodic organization in the Italian monodic Lauda and the early two-part madrigals. Here is that interpretation: "The orientation of each section of the melody toward a goal-tone and the ordering of those goal-tones in terms of one another—above all the dominance of final and upper fifth—give stable polar points toward which we can orient ourselves in the sound-space. Thus there is realized in the early two-part madrigals and the monodic Lauda an ordering of the sound-space that must be recognized as the basic form of tonality."10 Now the features of Italian melody which are here singled out—about the description there can be no disagreement—are in general characteristic of European melody after the end of the first millennium. What is more, they are directly called for in expositions of the modal system from that time. In short they are normative in every sense. With the experience of the music of several intervening centuries behind us we may feel that we are witnessing in those repertoires the birth of Tonality. But that would ignore the evidence of the theory and practice of their own time, which shows them to be fully developed specimens of a related, but quite different, system. It is one thing to recognize the caterpillar as that creature which will one day be transformed to become a butterfly, but it is surely quite another to mistake him, as he is, for a butterfly.

There is yet one more aspect to the problem of the illumination of works of music by historical information about norms. It is raised by music that, in some sense, incorporates common material: Christian chant, composed in many cases on established models or from a common stock of melodic formulas; liturgical and secular music of the Middle Ages and Renaissance employing an array of cantus firmus and parody techniques; music of the 18th and 19th centuries, where there may be a sharing of materials and techniques from the most specific to the most general. The study of one work in any of these categories informs the study of the other works of its kind.

In the first case—Christian chant—the composite character of melodies is of their essence; they cannot be regarded as though they had been through-composed, for the order of events in those melodies is to be recognized by reference to the strategic position of standard opening, extending, and closing formulas far more than from the persuasiveness of any inner logic or syntax.

In the last case, recurrent ideas are frequently those which lend a measure of referential meaning to the works in which they occur: the horn call at the beginning of Beethoven's Lebewohl sonata, the sarabande rhythm at the beginning of his Egmont overture, to cite two simple and obvious examples. Of course those clichés can and must be explicated in terms of the structure of the piece, but we lose a part of the meaning of the work if we ignore the reference.

Other works share modes of expression that are not nearly so concrete in their meaning, but that point nonetheless to vocabularies to which we can gain access only through the study of the several works which share them. One splendid example is the celebrated affinity, in manner and affect, between Mozart's G minor Quintet (Köchel 516) and his G minor Symphony (Köchel 550). In considering them together we can only gain in our sensitivity to their composer's expressive means.

While analysis deals, and must deal, with individual works of art in their wholeness and uniqueness, still there may be a sense in which each work is but a moment, frozen, in the creative life of the artist, and in that sense each work belongs to a larger whole. I do not see that it is necessary to regard these two points of view as mutually contradictory. Referring to his own practice, Picasso has remarked "Paintings are but research and experiment. I never do a painting as a work of art. All of them are researches. I search constantly and there is a logical sequence in all this research. That is why I number them and date them."11

"Words to the public, of little use to the scholar," we might remark. Yet perhaps we behave on some such belief, all but the most pertinaciously new-critical among us.

4. The coordination of the musical work with the world of which it is a product. With this as a goal, musical analysis is an investigation into music's functions and environmental relations.

This statement of aims appears quite naturally to belong to the special discipline of ethnomusicology. But I should like to quote, and support, the conclusion of Frank Harrison in the volume on Musicology of the Princeton Studies, "Humanistic Scholarship in America."12 Having made a strong appeal for the study of the institutions, the circumstances, and the beliefs that have nurtured and conditioned music, he wrote: "Looked at in this way, it is the function of all musicology to be in fact ethnomusicology."13

If we could believe that it were possible always to achieve the aims of musical analysis in the three senses to which I have so far alluded, then Harrison's suggestion would merely serve the filling out of the picture. But I strongly suspect that his conviction is born of long experience in the study of Christian liturgical music, and of the certain knowledge that analysis, even in the sense of laying bare the facts, is hardly possible without reference to the rituals which music served. Here, then, is at least one area in which the study of music's environmental relations is a crucial step in its analysis. But this must be the case wherever music has, by common consent, served some ritual or broadly defined expressive purpose.

Analysis in this sense would seek to comprehend 18th century tonality not only as the remarkable force for the creation of coordinated and directed musical structure that it is, but also as an instrument of the Doctrine of Affects. It would seek to comprehend the rhythm of the Ars Antiqua, not alone for its achievement in imparting both vertical and horizontal unity to the polyphonic texture, but also in the light of the force with which the notion of "Perfection" held the composers and theoreticians of that time. What, in the first instance, amounts to an aesthetic, in the second a metaphysical position, having called forth musical practices for their realization, and must be regarded as clues for the understanding of those practices.

Now I must return to my first category, the analysis of music on the level of facts, for there is a point of view that I have so far left entirely out of consideration: It is that the goal of musical analysis is the conditioning of the listener, for perceptive hearing. With this goal, musical analysis is the search for principles that will "codify previous hearing and extend and enrich the listener's perceptive powers by making listening more efficient and meaningful, by explaining the formerly inexplicable." (The quotation is from Milton Babbitt's review of Felix Salzer's book "Structural Hearing.")14

This point of departure differs radically from all the others that I have suggested so far, for it recognizes as its object and as a unique mode of experience the response to the work of art. While I come to it last here it is the first, and usually the exclusive consideration when the analysis is pursued, so to speak, for its own sake and not in the context of an historical investigation. But are historical considerations irrelevant in such a case? Is it, on the other hand, an approach that is hors de combat where the historian is concerned? To phrase the question in the better exercised battle terms, is there an antagonism between a "musical" and an "historical" point of view?

Such an antagonism would rest, presumably, on differences between the terms in which the work was conceived and those in which it can be most fully comprehended. As one such case I should like to cite the analysis by James Randall of Haydn's Quartet opus 76#5, published in the Music Review for 1960. In this analysis, all the progressions of tonicizations in the quartet are shown to be derivable from that of the opening passage by the operations of transposition, retrogression, and what is called "rotation" (the tonics of the progression are arranged in a circle, and one has the option of starting the process from any point on the circle).15 Let us assume that the assertion is empirically verified. Even so, the analysis in its present terms would not have been possible, or, at least, it would have been highly improbable, before Schenker and Schoenberg. Is it necessarily irreconcilable with what would have been possible? Not, it seems to me, if we proceed with sympathy. Musical analysis is communicated in a conceptual language, but the modes of cognition of the composer at work may be quite different. The translation from one to the other is an ongoing process. The composer may hold all the aspects of his work in suspension in what may be quite abstract cognitions until he commits them to a score, where they continue to reside and await discovery. In far more colorful terms, but in a similar vein, the psychologist Rudolf Arnheim has written this: "Artists, in particular, have learned to tread cautiously when it comes to reporting the internal events that produce their works. They watch with suspicion all attempts to invade the inner workshop and to systematize its secrets. Surely, creative processes are not the only ones to rely upon impulses from outside the realm of awareness, but they are unique in that their results give the impression of being beyond and above what can be accounted for by the familiar mental mechanisms. To the artist himself, his accomplishment is often a cause of surprise and admiration, a gift from somewhere rather than the traceable outcome of his efforts. It is viewed as a privilege that might be forfeited like the golden treasures of the fairy tales, which vanish when curiosity ignores the warnings and peeps at the miracle-working spirit. The privilege and nuisance of relying on helpers who do not take orders require those abnormalities of behavior for which artists have been known: those fears of power failure, those irritations and despairs, the agonies of waiting, the manic delights of success, the elaborate rituals necessary to create propitious conditions."16 Simply stated, there are differences between the creative and the re-creative processes. There is no prior reason to expect that the thinking of the composer should resemble in form the thinking of the analyst.

If there is any question to be raised about analyses of the type that I cited just above, it is not about their anachronistic character—that is about the inaptness of their terms for the period in which the work was composed, but once again about the significance of their observations. That is the question that suggests study beyond the single work in isolation.17

On the other hand, if there is a residual—as there must be—when we measure individual works against contemporaneous standards, it begs the question to account for that residual with those popular paradoxes, the composer ahead of his time and the critic or theorist behind his.

We may ask ourselves how much we understand about the music of our own time, even when composers, in their new pedagogical roles, have given us more abundant documentation of their intentions than ever before in this history of music. Their lectures, journal articles, and program notes have left us in the main unsatisfied, if what we expect of them is guides for our hearing. More than likely that is not what they mean to give us, for they are understandably preoccupied with matters of technique; not with the thing itself, but with how it comes to be. And certainly not with how it may be comprehended! Future students of our music will have to read them with a sensitivity for their place in the total musical context of our time. In our search for guides to the music of the past we are obliged to exercise the same caution with respect to non-musical documents. We must recognize, for example, speculative theory which, as Guido of Arezzo wrote in the 11th century, "is useful to philosophers, but not to singers," or practices which fail of recognition by theoreticians who, like medieval illustrators, copy not from nature but from their predecessors.

The true historical approach to music is, as it were, yesterday's musical approach, and conclusions from today's musical approach will add to the fund of tomorrow's historical knowledge.

I have taken too many twists and turns to offer a summary. But if my arguments bear on any single conclusion, it is what everyone knows: The analysis of music, like the analysis of anything, is best conducted in the context of all the information that relates to it. Polemics about the Musical and the Historical points of view would flounder if only we could agree upon the need for a sympathetic and canny, yet irreverent approach to evidence of every cast.



1This essay was written for the annual meeting of the College Music Society, Ann Arbor, 1965.

2Joseph Kerman, "A Profile for American Musicology," XVIII (1965), 65-69; Edward E. Lowinsky, "Character and Purposes of American Musicology; A Reply to Joseph Kerman," XVIII, 222-234; communication from Joseph Kerman, XVIII, 426-427.

3In using the terms "Analysis" and "Criticism" I do not presuppose the hierarchy—with "Criticism" at the top—that is suggested by Kerman. It is admittedly difficult for musical scholars to agree on the meaning of the word "criticism," for as Kerman has observed, it is a foreign word in the language of our discipline. But the study of music, considered as art rather than as document of history (the latter is possible and legitimate, but it presupposes the former) is in any case criticism. On the other hand, to put criticism at the end of the scholarly process, as Lowinsky does in writing of "the enormous work that needs to be done before fruitful and responsible criticism can be practiced in all areas of music history" is to suggest the procedure of the young piano pupil who learns the notes first and puts in the feeling afterwards.

In any case, a belief in the separability of the observational from the interpretive and the evaluative seems to be a minimal commitment in maintaining such distinctions. The following is meant to question seriously the soundness of that position. For now, I should merely like to ask whether an analysis that does not attain the level of criticism fails in a radically different way than does, let us say, an analysis that is made from an incomplete score. For it seems inescapable that any analysis—even the choice of one system of analysis over another—will embody judgements that must be regarded as "critical."

1The quotation is taken from Vasari's Lives of the Most Eminent Painters, Sculptors, and Architects, V (London, 1892), 239.

2Die Jugendwerke des Michelangelo (Munich, 1891), p. 23. Translation mine.

32nd edition, revised (New York, 1961), pp. 359-360.

4Gombrich's thesis—that the inventions of artists are accompanied by an evolution in the reading of images—has not, I believe, had any parallel expression in music criticism. But it seems clear that responses to musical sounds have changed through the history of the species. We may consider a single example. While it is certain that during the Middle Ages the octave—considered as an interval—was heard as we hear it today, it is also clear from medieval melody that our doctrine of octave equivalence was not in effect (i.e. two members of a pitch-class could not have been heard to share the same function). Similarly, it was through Rameau's teaching that we learned to hear inverted triads functioning as versions of their root positions.

5The Elements of Drawing, printed in The Works of John Ruskin, ed. E.T. Cook and Alexander Wedderburn (London and New York, 1903-1912).

6See Monroe C. Beardsley and W.K. Wimsatt, Jr., "The Intentional Fallacy," in the latter's The Verbal Icon (Lexington: University of Kentucky Press, 1954).

7Music Teachers National Association, Baldwin, New York, 1961.

8Ibid., p. 15.

9"Tone System in the Secular Works of Dufay," JAMS, XVIII (1965), 132-169.

10Heinrich Besseler, Bourdon und Fauxbourdon (Leipzig, 1950), p. 72.

11Quoted by Rudolph Arnheim in Picasso's Guernica (Berkeley and Los Angeles, 1962), p. 12.

12Englewood Cliffs: Prentice-Hall Inc., 1963.

13Op. cit., p. 80.

14JAMS, V (1952), 260-265.

15Examples: the original progression D-b-G-e-D; its retrograde form D-e-G-b-D; the upward transposition by minor third of the retrograde, beginning with the fourth element d-F-g-vol6id597-d.

16Arnheim, op. cit., pp. 1-2.

17In this instance it might be a question about whether Randall's statement of the tonal relationships in the work in question—attractive as it is for its own elegance—is as significant as would be one in which symmetrical progression by third (retrogression and rotation follow readily from that as first principle), with consequent emphasis of the subdominant, is seen as an alternative to progressions based on the dominant relationship. Then the non-unique case of the Haydn Quartet would be illuminated by consideration of other tonal works—e.g. Beethoven's Hammerklavier Sonata, the Ninth Symphony, Brahms' Fourth Symphony.

4983 Last modified on November 14, 2018