Musical Works, Improvisation, and the Principle of Continuity Lee B. Brown The Journal of Aesthetics and Art Criticism, Vol. 54, No. 4. (Autumn, 1996), pp. 353-369. Stable URL: http://links.jstor.org/sici?sici=0021-8529%28199623%2954%3A4%3C353%3AMWIATP%3E2.0.CO%3B2-7 The Journal of Aesthetics and Art Criticism is currently published by The American Society for Aesthetics.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/tasfa.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact
[email protected].
http://www.jstor.org Wed Aug 1 00:04:50 2007
LEE B. BROWN
Musical Works, Improvisation, and the Principle of Continuity "One surely cannot play what is not."
"Oh yes, one can play what does not yet exist."
Thomas Mann, Doctor Faustus.
This paper examines some central features of improvisational music.' I shall explain why it is impossible to find a place for such music within systems of aesthetics that treat artworks as obedient to what I shall call the principle of continuity. According to this principle, all artworks are continuants, and therefore subject to criteria of reidentification.* A paradigm example of such a system is that of Nelson Goodman.3 Even if it is obvious that improvisational music is likely to be marginalized by this kind of system, it is still worth considering exactly why. Anyway, for some philosophers, the statement may not be so o b v i o u ~Either . ~ way, the main aim of this paper is not to attack Goodman's view-although I shall express an opinion on that matter at the proper time. Rather, it is to use the features of his nicely articulated system to profile what is special about musical improvisations. I shall address improvisational performances as such and not-except incidentally-transcriptions or arrangements of them for later performance, or musical works that have an improvisational origin. Jazz improvisations are our local paradigm, so I shall often key my discussion to them. While improvisational performance will be my topic, the essay should have implications for improvised material of all kinds-dramatic pieces, "performance" art, or even improvised poetry, a form of which Vasari tells us Leonardo was master. Nowadays, it is generally recognized that Goodman did not appreciate the relevance of contextual considerations to our understanding of artistic phenomena. It seems clear that two identical texts will still not represent the same poem if one was set down first in the sixteenth century, the other in the twentieth. Analogous examples can be framed for musical works and painting^.^ Among these considerations are
those I shall term modal. "Modal" here has nothing to do with modal logic, but with artistic processes, as contrasted with products. It pertains to the manner in which these processes take place, e.g., in accordance with a score, spontaneously, etc. The relevance of such considerations to our topic seems natural, given the distinctive manner in which improvisational music is generated. Philosophers have suggested, quite independently of any concern with improvisation, that Goodman's categorization of the "standard" arts needs to be amplified in order to take account of modal factors.6 However, we shall see that even if we were to augment Goodman's system along these lines, it will still be hard to find a place for improvisational art in it. This essay will be developed in the following stages:
(1) After a characterization of improvisatory music and (2) a review of Goodman's division of the arts into the categories of the allographic and autographic, (3) I shall then describe an obstacle, reflecting the transitory nature of acoustic phenomena, about the placement of improvisational music within the category of the autographic. Taking some pains to understand the exact nature of this problem, (4) I shall then consider why improvisatory music will not fit in Goodman's alternative category. (5) After summarizing the argument about the placement of improvisational music in Goodman's system and (6) amplifying the contrast between improvisational performance and work-performance,
The Journal of Aesthetics and Art Criticism 54:4Fall 1996
The Journal of Aesthetics and Art Criticism
(7) I shall discuss some special features of improvisational activity. It will be important to bear in mind that for the purposes of this discussion I shall assume that recording technology does not exist-or at least I shall try to do so.' Finally, a point about terminology: since the term "work," as well as the related term "artwork," has connotations I sometimes wish to avoid, I shall now and then use the slightly more neutral expression "aesthetic object" when the occasion suggests it.
I begin with the central idea that an improviser makes substantive decisions about what music to play while playing it.8 An improviser (a) presents music intended to be worth hearing, by (b) creating it a s she plays. Given conditions (a) and (b), risk is latent in improvisational performance. Still, it is a regulative ideal that an improviser will attempt to satisfy the normative aim embedded in (a) while satisfying (b).9 While I shall generally speak of the improviser as someone acting alone, a consideration of collective improvisation underscores the character of improvisation by its resemblance to the flow of unscripted conversation. The difference is that the solitary improviser must often answer questions she poses to herself, so to say, where her answers typically lead to new questions. Now, improvisers do not create ex nihilo-a fact that might be trotted out in a mistaken effort to debunk the improvisational enterprise.10 Improvisers of the classical music of Iran must first learn some 300 musical elements that make up the repertory of what is called the rad$l Clearly, whatever the Iranian improviser invents while playing, he does not invent these materials themselves. Similarly, jazz improvisers do not take up instruments and improvise out of nothing, but take for granted a repertory of musical forms, e.g., the standard twelve-bar I-IV-I-V-I blues chord progression distinctive of the blues, or the chord progressions of preexistent tunes. Beginning conservatively, players master a repertoire of musical phrases, out of which they gradually learn to construct music of their own. On-the-spot musical invention, however daring, invariably leaves something intact. In short, I want my conceptualization to be a modest one.
While, for the sake of economy, I shall often speak of improvisers as creating music spontaneously, the hyperbole is not intended to rule it out that an improviser's decisions while playing are often guided by a larger conception of where he or she is going. Nor should we require that every authentic performance meet stringent standards in regard to musical novelty, or that it stake out utterly new stylistic territory. However, one can resist the cynical conclusion that "improvisers" do not ever really improvise. Even if all improvisatory music were made up entirely of formulaic musical atoms-a generous admission-we could still describe genuine improvisers as players who make on-the-spot decisions about how to sequence them. Of course, improvisers have aims beyond the modest ones we have described. With many genres of improvised music, there is a tacit assumption that players will try, if they can, to surprise us. This secondary regulative ideal governs classical Iranian music, for example, where it is those "parts of the radif that lend themselves to far-flung improvisation [that] are valued; those that have predictability ... are lower. ... The exceptional and unexpected is valued."12 In the same spirit, every jazz improviser acknowledges an obligation to break out of the hand-me-down repertoire of materials he has internalized. All players have heard the kind of advice Charlie Mingus gave beginners, namely to reach "beyond their formulas."13 Of course, what counts as a significant transmutation of musical materials varies widely as we move from one jazz subcategory to another. The rhapsodic kind of piano music Keith Jarrett plays would not reasonably be judged by the same criteria we would apply to mainstream "stride" playing by Teddy Wilson. Judgments about the originality of Thelonius Monk dismantling a Broadway showtune could not apply directly to the niceties of a Ralph Sutton performance of a Joplin rag. Opinions about John Lewis's partially improvised jazz fugues require yet a further shift of criteria. Although this secondary aim raises the ante on the risk that is inherent in improvisation, it would be too extreme to disqualify a performance as improvisational because it does not realize them.
Brown Musical Works, Improvisation, and the Pn;nciple of Continuity
Goodman treats his central categories of the arts as exhaustive. He states, concerning the classification of music, that it is "nonautographic, or allographic."I4 In other words, the class complement of autographic art is identical with the class of allographic art. What I have called the principle of continuity underlies Goodman's assumption that his two categories jointly exhaust the sphere of art. This claim deserves explanation. Our systematization of art would be simpler than it is if all art came packaged in the form of relatively enduring things. In that case, the concept of the same artwork would be exemplified only by works that have physical continuity. All art would then be autographic. A "work of art is autographic," Goodman says, "if and only if the distinction between original and forgery" is significant.15 The central identity-question about such art is whether an enduring entity could be traced back to the moment when it came "from the artist's hand."I6 It takes only a slight adaptation of the idea to fit multiply-instanced autographic works, e.g., etchings. They are physical descendants, so to speak, of a common physical ancestor. Where physical continuity breaks down, Goodman falls back upon the next best thing for explaining continuity. This concept we might call logical or structural continuity. It gives us the category of allographic art. In spite of the transitory character of performances of a work, their authenticity can be measured by compliance with the strictly notational characters of a score.17 As a result, a work manifested by one performance can be manifested by another in spite of their physical discontinuity. The complementary application of the two categories to art in general would insure that it exhibits one kind of continuity or the other. (And what other sort of continuity could one imagine?) So, it is understandable why Goodman treats his central categories as exhaustive. It is a tidy system, but where should we place musical improvisation in it?
In spite of Goodman's classification of music in general, there is aprima facie reason for locating improvisations in the sphere of autographic art. The directness of our experience of improvised
355
music seems to parallel that of our experience of painting. When we gaze upon The Tempest of Giorgione, we are gazing upon the very object that Giorgione has made. Similarly, when we listen to Keith Jarrett improvise, we are hearing the very sounds he has created. However, we immediately run into the difficulty that music, considered as an acoustic phenomenon, is ephemeral. As a result, something curious happens when we try to pose Goodman's well-known forgery problem about improvisational performance.18 The issue here is about what Jerrold Levinson has described as "referential forgery," that is, the forgery of specific artworks, as contrasted with "inventive forgeries," i.e., forgeries of works of a given kind. ' 9 Goodman argues that no clever forgery would serve in place of an originaL20 The problem is that when we try to pose Goodman's forgery-question about a pair of improvisations, we fail, because a pair of improvisations could confuse a listener, in the way ordinary forgeries can confuse, only if the listener could bear in mind the original with which the copy is liable to be confused. This excludes superficially similar kinds of error. Ceremonial recreations of improvisations are common, of course, and unobjectionable-although such copies obviously do not play the improvisational role of the original. An uninformed person listening to Coleman Hawkins ceremonially recreate his Body and Soul solo might believe he had been hearing Hawkins improvise the music on the spot. However, this mistake has nothing to do with confusing the performance with an earlier original. The problem is that if we try to revise the story so as to introduce the right kind of confusion, the situation provides the evidence that would expose the deception before it gets off the ground. Any clue that would suggest the possibility of forgery, e.g., the recognition of the similarity itself, would give the game away. Imagine a faker, Corman Hackins, playing notes very close to those of Hawkins's performance some fifty years earlier. A listener would know that if fakery were involved, the original would have to be a performance that took place prior to the one the listener is hearing. While a painting can survive the ravages of time and confuse a viewer by turning up side-by-side with a clever forgery, it is otherwise with an improvisation. Of course, this problem would not prevent Hackins from
The Journal of Aesthetics and Art Criticism
plagiarizing Hawkins's performance. Plagiarism, however, is of little theoretical interest to Goodman. Unlike the forgeable-as he sees it-there is not even an apparent correlation between the plagiarizable and the autographic. Supposing we have identified a problem about fitting improvisations into the category of the autographic, we have not yet shown it to be a problem that attaches to improvisations qua improvisations. The same problem could be raised about a broad sphere of aesthetic events that are not improvised. Consider preconceived but nonnotated dance performance~,for instance, or fireworks displays. Even assuming a practice of attribution to identifiable pyrotechnicians, forgery projects for fireworks could not get off the ground. Now, a conspicuous difference between the two kinds of cases is the way they are generated-a modal difference, in short.21 One guesses that, because of the way improvisational music is generated, projects of forgery would run into difficulty in that sphere even if acoustic phenomena were not ephemeral. However, we will not get at the suspected difference unless we think about the matter in the right way. Consider a hypothetical subclass of paintings that are improvised.22 Like improvisations in an obvious respect, these works nevertheless involve relatively enduring material objects. Now the way such works are made is surely relevant to certain questions of authenticity we might raise about them. Imagine a project, by Jones, of making a copy of Smith's piece, Blue Flush, and presenting it as Smith's. Unlike improvisations, no problem of ephemerality would interfere with Jones's plan. Applying his reasoning about forgeries of ordinary paintings, Goodman would undoubtedly say that Jones's copy would not serve in place of Smith's original. However, whatever we think of Goodman's reasoning, modal considerations probably provide another reason for the negative reply as well. For suppose we hypothetically isolate and set aside the actual and possible perceptual differences between the two by pretending that there are none. It is arguable that Jones's copy would still not serve in place of Smith's because of the difference between the ways the two graphics were made.23 Some philosophers would urge that modal considerations probably affect informed responses to ordinary works of autographic art as well. Whatever other reasons
might be given why a forged or plagiarized painting will not serve in place of the originalsome would say-the way it is made is also rele~ant.~~ However, as described, there seems to be no obstacle to Jones's project of forging Blue Flash. So far, then, I have not turned up any special problem about classifying improvisational music as autographic, beyond the problem posed by its ephemerality-a problem it shares with fireworks displays, as I noted. If the way improvisations are made provides a more specific reason why they are not autographic, the modal factor must figure in those cases in some special way. I observed earlier that improvisational and autographic art both feature a kind of directness. However, there is a difference in this respect between the two. I shall term the kind of directness that typifies improvised music presence. Compare the following: (a) A condition of one's informed response to The Tempest is that one takes its lines and colors as being applied x-ly as one watches. (b) A condition of one's informed response to an improvisation is that one takes its sounds as being created spontaneously as one listens. (c) A condition of one's informed response to Blue Flash is that one takes its lines and colors as being applied spontaneously as one watches. Never mind how to fill the place-holder in (a). (It is not easy to give a general recipe for how to do so, except negatively, with a locution such as "nonforgedly." A version of the Croce-Collingwood theory might lobby for "spontaneously," but few will be convinced.) The point is that whatever modal conditions are relevant to our understanding of The Tempest, they obviously do not dictate that we are supposed to watch the painting being made. In spite of their relevance to an informed response to the work, they are not constitutive of the work. However, (b) seems exactly right. With improvisations, we are supposed to understand the music we hear as being created before our very ears. So, the problem I first identified about the transitory character of acoustic phenomena was an artificially abstracted aspect of a more fundamental feature of improvisation. The feature of the music that I have calledpresence suggests
Brown Musical Works, Improvisation, and the Principle of Continuity that it is over processes that an autographic principle of continuity would have to range, if we are to apply it at all. Improvisations are not excluded from the sphere of the autographic simply because their effects are ephemeral results of processes. They are excluded because improvisations are transient processes. Indeed, they are actions. (It is because he sees this that I regard Philip Alperson's recognition of the centrality of activity in improvisational music to be the special virtue of his essay on this The way the acoustic material is generated in these cases is an essential component of the genuine article. This does not entail the absurdity that we are expected to respond to improvisational processes in abstraction from the sounds generated. An improvisation consists neither of disembodied sounds nor of an activity abstracted from the sounds. It consists of the whole activity of creating a sound-sequence in the course of playing it. y of Now what about (c)? While the ~ ~ a acopy Blue Flash is made may well be relevant to certain critical questions we might pose about it, (c) is no more plausible than (a). However, if we alter the terms of the thought experiment and reconceive Blue Flash as a type of improvisational performance, analogous to musical imimprovigraphic, as one might provisations-an term it-then (c) would hold for it. The analogy between Blue Flash and an improvisational performance of music would be much closer. On its original description, though, both Blue Flash and ordinary works of autographic art have a feature we might call temporal indcference. Any modal condition relevant to an informed response to an example of either kind of work could be fulfilled at any temporal distance from the response itself. I would be in a position to respond no less appropriately to Blue Flash if I engaged it a hundred years after its creation than if I apprehended it hot off the easel. It might be replied that fireworks displays, by contrast with Blue Flash, are not temporally indifferent anymore than musical improvisations are. However, their indifference has nothing to do with the processes by which they are generated, but with the fact that fireworks probably are nothing more than the ephemeral products of those processes. We might ask why, given the similarity between autographic art and improvisatory music, we could not simply loosen up Goodman's con-
357
cept of the autographic and place improvisations there anyway. In the following section I shall be profiling the peculiarly singular character of improvisational performance. But is this not a feature of paintings as well? The difficulty with this attempt at rapprochement is that paintings are singly instanced simply because of the physical process by which they are made. On Goodman's account, there is nothing inherent about autographic art that requires works of that kind to be either singly or multiply instanced, as he himself suggests.*6 By varying those processes, as with etchings, the category of the autographic can embrace multiply instanced works. But improvisations do not just happer1 to be singularities. If I am right, they could not be otherwise. Furthermore, even if we were to classify improvisational music as autographic, improvisations would still be anomalous members of that group, given the presumed linkage between the autographic and the relatively endurable. In a symposium on the topic of work-identity, Goodman appears to weaken this linkage slightly by observing that the concept of autographic art would have application even in a world in which no "ill intent" to practice forgery could be found.27 The question, though, is whether the concept would have application in a world in which it would be impossible to succeed in such a project. Could one fall back on the position that although Hackins can forge Hawkins's performance, such projects never stand any chance of success? Maybe-but maybe not. A forgery is something that purports to be an original. Given the obstacles Hackins would be up against in trying create deception, can he even meaningfully pu1.por.t his performance to be an original? If forgery does not threaten improvisational practice, what kind of inauthenticity does? Plagiarism, of course. I restrict myself to two comments about the matter. (i) While a plagiarized "improvisation" really is not the genuine article at all, one could hardly say anything analogous about either a plagiarized or a forged painting except with the help of a theory about the matter.28 (ii) Plagiarism in an improvisational context is a distinctively vexatious matter, because it betrays the goal that is so obviously foregrounded in this kind of art. The problem is compounded by the fact that such fakery is nevertheless endemic to improvisational practice.
The Journal of Aesthetics and Art Criticism At least, this fact is conspicuous in the case of jazz. Given the demand to stand up and deliver whether ready to do so or not, soloists will understandably fall back on stretches of imitative material. Correlatively, jazz people keep their ears open to detect such lapses in each other's playing-even if otherwise harsh judgments about them will, under the circumstances, often be softened. One wonders if Goodman did not (partly) see the problem delineated in this section. He states that the "ephemeral character of performances ... favors an art's" being "allographic. ..."29 Let us see how that alternative fares.
Let us now try to apply Goodman's claim that all music is allographic to improvised music. The problem we run into immediately is that the theory of the allographic is-for our purposes here-the theory of musical works. But musical works are multiply instanced, while improvisations are not. While this statement is undoubtedly correct, simply saying so is not enlightening. Let us press the point with the help of a counterconjecture: Perhaps multiple instantiation is a feature of the improvisational world, but not a very common one-or one we just do not take much notice of. Suppose that an improvisation by Corman Hackins ( H , ) just happens to be perceptually indistinguishable from the famous "Body and Soul" solo of Coleman Hawkins (H,). Unlike the pair described in the previous section, this pair not only parallel each other perceptually, but they are equally spontaneous. I shall call such a pair a perfect pair.30 Now why could H, and H, not be regarded as compliants of some single set of notational characters? Goodman's official view is more tolerant of such a proposal than one might suspect-indeed, probably too tolerant, as we shall see. By his terms, an acoustic clone of an authentic performance of a given work, whether intentionally copied from the former or not, could count as a genuine instance of that work. Consider, for example, a pianist who plays music by copying another performance-playing it by ear, in other words. As long as the copy complies with a work-score, it would qualify as a genuine instance of that work. Indeed, the acoustic output of a performance would qualify
as a genuine instance of a musical work, however it originated. Consider an improviser who coincidentally plays notes compliant with the score of a Bach two-part invention. Perhaps the problem is that no common score exists with which H I and H, could be compliant. But does this matter? Even if an improvisation is not actually notated, it may be notatable. Transcriptions of jazz solos are routinely published for pedagogical purpose^.^ Now a host of interesting technical issues arise about the applicability of Goodman's theory to this class of cases. First, we have to reckon with the fact that his theory of notationality is an idealization of a European musical tradition. No other culture, he says, "has developed any comparably effective musical notation over the c e n t u r i e ~ . "Surely, ~~ however, the applicability of the methodology to the wider range of musical practices found around the world is problematic. Many of these practices, it should be noted, make themselves felt in profound ways in American jazz. Both tonally and rhythmically, jazz is a sort of vectorial resultant of European and African practice, each of which, considered in itself, is a stranger to the other. For instance, jazz contains a huge class of notes that are, according to European practice, systematically skewed from a tonal point of view-the socalled blue notes. Another example is the class of "swinging" eighth notes that are represented inaccurately in transcriptions of them-as, e.g., dotted eighth notes plus sixteenths. We would have to consider how such features are to be represented by a Goodmanian notation. One possibility is that such features might be represented by what Goodman calls supplementary instruction^.^^ Such instructions include nonnotational marks, e.g., diminuendo "hair pins," words written over the bar-line, traditions handed down verbally, and performance agreements made on the spot. So, perhaps we could include the tacit agreements that enable jazz players to interpret the inevitable misrepresentations of jazz music that lurk in European transcriptions of it as belonging to these supplemental instructions. For example, trained players will know how to interpret the standard scored misrepresentations of "swinging eighths" and make appropriate adjustments in order to get the required effects. The problem is that the compliance of a per-
Brown Musical Works, Improvisation, and the Principle of Continuity formance with supplemental instructions does not, for Goodman, play any role in identifying it as an instance of a musical work. Goodman claims that such instructions account for the expressive features of performances of standard musical works. However, these features are, for identificational purposes, n o n f ~ n c t i o n a lEven .~~ if this were a plausible approach to many of the salient features of work-performance, would we want to advance an analogous position about the wide range of "eccentric" sound-events that pervade a typical jazz improvisation? It would be absurd, surely, to treat them all as elaborately nuanced compliants of the ordinary European note-heads that make up the imaginary score that supposedly defines the hypothetical work in question. Nor do we want to adopt Theodor Adorno's outlandish view that such soundevents must be regarded as failures to comply with given European note-heads-as errors, in sh0rt.~5The problem, in sum, is that features of a jazz improvisation that are hard to dismiss as playing identifying roles fall into limbo on Goodman's account. This is understandable, given that some of them, e.g., blue notes, exist only in the cracks between the keys of a tempered keyboard. While these are interesting theoretical questions, I do not want to become further absorbed by them. For two related reasons, the real problems lie elsewhere. First, all readers of Goodman eventually find themselves wondering if his idealization provides conditions that are either necessary or sufficient for the identification of instances of ordinary musical works. Even in this sphere, Goodman is forced to admit that, for example, tempo specifications, being nonnotational, "cannot be accounted integral parts of the defining score. ..."36 (Again, they belong to the supplemental instructions.) Second, we need to see why improvisational music is peculiarly resistant to the Goodmanian analysis, for one's intuition is that work-performances and improvisations really are different in kind. For the sake of argument, then, I shall table the issue about the problems Goodman faces applying his idealized picture of notationality to music-whether of an improvised sort or not. I shall assume one of two possibilities: either Goodman's methodology for identifying pieces of music-including that of improvisational origin-is workable; or some other method of identifying a common
359
structure or work-type in its various tokens is feasible.3' Flint Shier has suggested that Goodman's theory about compliance with a notational scheme is just his suggested method of insuring the identity of a common structure or work-type in its various tokens.38 In favor of something like Goodman's own approach is the fact that standard musical practice does have it that being true to a work customarily involves being true to its score. I want to explain why, however this matter be resolved, there would still be a gulf between improvisational music and work-performance. However, Goodman's account of work-performance stands in need of another kind of emendation before it can be usefully compared with improvisational music. Although Goodman's theory of notationality is based on European notational practice, his appeal to it is curiously disconnected from the role scores actually play in that tradition-that is, from their relevance to what I have called modal conditions. Without augmentation by reference to such conditions, Goodman's theory faces potential counterexamples even in the sphere for which it was devised. For instance, on his view, should a burst of radio static happen to comply with the score of Beethoven's opus 11 1, it ought to count as an authentic performance of that work. So, in order to continue with our thoughtexperiment, we must augment its application to ordinary work-performances. On Goodman's behalf, Jerrold Levinson defines a concept of the autographic-autographic, art, we might term it-as "art such that the question of the authenticity of works of that type is not determined at all by compliance with the characters of a n0tation."~9He further defines a concept of the allographic-allographic, artas "art such that the identification of instances is partly determined by notational c ~ m p l i a n c e . " ~ ~ On the augmented version of Goodman's view, genuine instances of a given work must not only token a given type, but stand in "intentional and/or causal relatedness" to the composer's "dated act of indication."41 (Scores, obviously, are the typical repository of such "indications.") A bonus is that these redefinitions stand a better chance of being coextensive with an alternative pair of definitions that Goodman also employs: autographic, art is "art such that even the most exact duplicate of such a work cannot, under any conditions, 'count as genuine,' since it would not
The Journal of Aesthetics and Art Criticism would qualify as two instances of that work. be the one that came from 'the artist's hand.'"42 (And perhaps nothing should rule it out either.) Correspondingly, allographic, art would be "art Any analogous diagnosis of a pair of soundin which both one sound-event and its clone events emanating from a pair of improvisations could, under certain conditions, serve in place of would be incoherent. each 0ther."~3Most important, Goodman's categories, with their revised definitions, now ex(iii) An improvisational action is an aesthetic haust the artistic field. Or so Levinson c l a i m ~ . ~ singularity. ~ If HI and H, really are improvisaI shall evaluate Levinson's larger claim later. tional in character, then each harbors its own generative act. Essential to H, is its being this We now have enough of the pieces to make a comparison between improvisational music and spontaneous action; essential to H, is its being that one. H I and H, each possess a kind of aesGoodmanian allographic art. thetic indexicalitx so to say. While H I and H, do (i) The most obvious effect of Levinson's token many common action-types, e.g., playing emendations on our comparison is that the two middle C, there is simply no type-improvising types have different intentions. A piece of improvised music is not generated the way that the this-of which they could be tokens. augmented theory tells us work-performances All in all, then, the thought-experiment is disare generated. True, H I and H, share a general tortive. It invites us to take a point of view of the aim, namely, to improvise some music or other. matter that is framed, so to say, sub specie aeHowever, two performances of a common work ternitatas. From that point of view, HI and H, W are bound together by the aim not merely of seem to be composed of a set of acoustic events playing music-or even musical works-in which are, furthermore, determinate in characgeneral. They are bound by a specific common ter. From a perspective on H, and H, as the proaim, namely, to token the antecedently identicesses they really are-a point of view in media fied type W. It is no accident that scores for H, res, so to say-there is no determinate pair to or H, would be merely hypothetical ones-or, if match up in the way we have supposed they are actual ones, capable of being written down only matched up. Each member of the pair hangs in after the fact. The hypothetical unity of H I and the balance as long as the performance lasts. H2 is just an artifact of the thought-experiment. When each becomes determinate-and if, by a (ii) One might be tempted to take the foregofluke, they were to do so in qualitatively indising point of view on H I and H, anyway, if we tinguishable ways-a moment essential to it has begin by thinking of the pair as composed of passed. With improvisational music, the perforsheer acoustic events. However, our discussion mance process gnaws into the very essence of of presence in the previous section suggested the aesthetic object, so to say. that any criterion of reidentification that might The foregoing considerations all fit the charapply to improvisation would have to range over acter of an informed interest in improvised actions, not over acoustic events. On Levinson's music. They all reflect the fact that the point of amended version of Goodman, no less than on such performance is to create music in the the original account, criteria of authenticity range course of playing it. The connection between over "sequence[s] of s0unds."~5Obviously, the the music's telos and (i) is clear. The connection typical way of satisfying the "intentional and/or with (ii) and (iii) will be developed in section causal" conditions is by means of a standard seven where I shall further address the compotype of intentional action, e.g., playing the nent of activity in musical improvisation and piano by hand. However, there is nothing in our reception of it, in contrast with our reception Levinson's revision that requires that such a perof work-performance activity. Already, though, formance play an intermediary role here. In the it should be clear that someone who approached days of player-pianos, some piano rolls were cona piece of improvisational music as if it were structed without the intermediary agency of any merely a string of sounds-or even such a string performer at all. Holes were simply punched in as exemplifying a musical structure or patthe paper corresponding with the details of the tern-would not be in a position to respond to it score of that musical work. Nothing in Levinin an informed way. son's account clearly rules it out that a pair of sound-sequences produced by such a piano roll
Brown Musical Works, Improvisation, and the Principle of Continuity
In sum, improvisations are not subject to either kind of Goodmanian reidentifiabilitv. The comparison of them with autographic art shows an improvisation to be a certain kind of ephemeral process, namely, an action. The comparison of them with allographic art amplifies our description of them by profiling their singularity. They lack either autographic or allographic continuity. So, it seems that we cannot follow Levinson in his solidarity with Goodman's opinion thatgiven amendments of the sort he suggestedthe allographiclautographic duality exhausts the artistic field. The larger moral for any aesthetic system that treats art as coming packaged in reidentifiable works is that there simply is no entity with which an improvisation can be reidentified-nor any entity with which an improvisation might be mistakenly reidentified. The argument does not imply that we have no means of picking improvisations out, of course. Scholars of recorded jazz do it quite naturally, by referring to, e.g., Miles's "Funny Valentine" solo played on June 12, 1959. If I am right, identifications of improvisations begin and end with such characterizations. It is not easy to decide what Goodman himself would think-or what he should thinkabout the relationship of improvisational art to his system. In The Languages of Art he neither granted nor denied that the sphere of art spills out beyond the territory mapped by his categories. Indeed, artistic phenomena to which Goodman does not intend his study to applyassuming there are such-were made conspicuous in LA by their absence. In a work written later with Catherine Elgin, the authors briefly contrast musical variation with improvisation. The latter, they say, is a "spontaneous invention having nothing to do with variati0n."~6 Once mentioned, the topic is set aside, and the discussion proceeds within the purview of the central allographic/autographic dichotomy. In Of Mind and Other Matters, Goodman takes note of pieces by John Cage, scores for which do not come close to satisfying strict notational criteria. While he acknowledges them, as a matter of "clarification" or "stress,"47 he does not seem to regard the cases as serious counterexamples to contend with, and the theoretical waters close over them. Of course, we do not need to exam-
36 1
ine avant-garde art to find problematic cases. Consider the world's vast output of "ethnic" and "popular" music, for instance. There seem to be two directions Goodman's view might take. Isolated later comments by him suggest that his disjunction is exhaustive only for the category of artworks.48 (This may be Levinson's view of the matter as we1L49) But are the Cage examples not works? If not, what are works? Just the things to which the dichotomy applies? Perhaps works are cases of art that are subject to reidentification. (Indeed, on at least one occasion, Goodman seems to put it this way.50) If this is his view, it would have been helpful to provide the reader of LA with some useful caveats about the matter rather than simply describing his subject as the "arts."5l Anyway, the move only slightly widens the apparent circle. On the other hand, Goodman sometimes seems to regard his system as applying to all art without exception. In a passage of speculative history, he proposes that originally all the arts might have been autographic, but that notation was devised "to transcend the limitations" of time, i.e., the ephemeral and "the individual."S2 This is striking, because these "limitations" are, of course, typical features of improvisational music. (Perhaps Goodman would regard such music as a vestige of an undeveloped stage in art's history.) However, unless we set aside the assumption that the autographic concerns the relatively endurable-and hence the forgeable-we would have the same kind of trouble applying the label to ancient ephemeral art as we have had applying it to improvisatory music of our own time.
Comparing improvisations with work-performances, rather than with works themselves, seems awkward; but it seems just as awkward to make the comparison the other way. Perhaps improvisation is composing in the course of performing-as has been suggested.53 However, this will not do either. Of course an improvisation is a process of creation; and it does eventually generate a determinate result-which could be transformed into a full-fledged work. However, while musical works are subject to reidentification, this is not the case with improvisa-
The Journal of Aesthetics and Art Criticism tions qua improvisations. Could an improvisation be treated as a work-type that just happens to have a single performan~e-token?5~Not likely. We can imagine opus 11 1 having been played only once. However, it is inherent in the concept of it as a work that it could be instanced any number of times. But it would not have made sense to ask Charlie Parker to improvise a given solo a second time. It has to be granted, though, that considerations relevant to the sphere of works and their performance and those pertinent to the territory of improvisatory music do blur into each other-or seem to. The matter can be considered from two directions. (i) Improvisations are often performed in the course of playing modest works. The jazz field, for instance, contains a broad spectrum of musical material, e.g., jazz "standards," along with specialized jazz compositions, and I can only sketch one or two points about it. Many cases are full-fledged musical works containing improvisational components, similar to both modern "serious" works and older ones, e.g., concertos, as played when their cadenzas were not yet fully written out. However, much mainstream jazz consists of improvisations on X-where "X" denotes a familiar tune.55 They seem to be work-performances and improvisations at the same time. However, saying so is not like saying that someone at the same time both improvises and plays opus 111. First, the sense in which improvisers perform works is much looser than the sense in which Serkin plays opus 11 1. George Russell's deconstruction of the pop tune "You Are My Sunshine" is only one obvious example.56 A listener who said of one of these performances, "That's just not 'You Are my Sunshine,"' because it did not conform to a scored representation of it, would misunderstand the point of the performance. Second, for many jazz ears, what identifies a "work" is its pattern of harmonic motion. For example, the strains of the A-A-B-A tune "Koko" that became the basis for a series of Charlie Parker improvisations do not have the same notes as the strains of the tune "Cherokee." So, from one point of view, a player who has performed one of these tunes would not have played the other. However, since the two havenot by accident-basically similar patterns of harmonic motion, we might say that a perfor-
mance of one is a performance of the other. The same point goes for the blues. There is a sense in which every twelve-bar blues performance in Bflat is an instance of a common work, and a sense in which it is not. The problem is not just that we lack any theory about how to recover the improvisations themselves-the problem discussed in section fourbut that we do not even have any very strict antecedent guidelines for individuating the items that are the bases of the improvisations. Certainly, jazz players pay little attention to the simplistic representations of them in published sheet music. Unlike the situation with fullfledged musical works, there is no rigorous practice that theory ought to reflect. (Certainly, it would make no sense to detail ways in which any such item succeeds or fails to conform to strict Goodmanian criteria of allographic authenticity.) Given the built-in lack of precision about what counts as playing the work either in the improvisational or the nonimprovisational context, it seems harmless to say, with this kind of jazz at any rate, that an improvisation on such a work does two things at once. The criteria for its being a work-performance simpliciter are too loose to get in the way. Jazz practice lives with this lack of precision. Indeed, it thrives on it. Jazz players continually reshape jazz's fluid musical materials for new musical purposes. (ii) Informed criticism often touts not only interpretive flexibility but even unpredictability in full-fledged work-performance. Wilhelm Fiirtwangler, unlike more rigid conductors such as Arturo Toscanini, was said to let his performances "breathe." Or consider a recent press report that Luciano Pavarotti was challenged by the BBC for having lip-synched a televised performance. The New York Times complained that Pavarotti's fakery makes a mockery of the spontaneity of the real thing, of the "thrill of the moment," the "gamble on the unknown."57 So, is spontaneity not relevant to the sphere of workperformance as well? One problem is that not all informed parties will agree with this picture of the thrills of live work-performances. Glenn Gould, for example, had such serious doubts about the possibility of serving works well in live performances that he retired from the concert stage. About the vaunted value of risk in performance, he acerbically replied that the "purpose of art is not the release of ... adrena-
Brown Musical Works, Improvisation, and the Principle of Continuity line, but ... rather the gradual ... construction of a state of wonder and serenity."58 But here we need to distinguish kinds of cases. First, the Times author is probably fretting about the fact that a s taped, we are deprived of the delicious possibility that Pavarotti might fall off his vocal high-wire-make a terrible mistake, in short. Gould was probably worrying about something more subtle. He describes, for instance, how one of his live performances could be deformed by impulsively plunging in at too fast a pace to suit the subsequent phrasing he thought the work required.59 And here we must make a second distinction: not all eccentric performances are idiosyncratically inflected and phrased on the spot-extemporaneously, in other words. Let us think about those that are. Given a focus on the manner in which a work is performed, as contrasted with an interest in the performed work per se, might our earlier argument not be turned against us? Consider an example of a perfect pair again, but this time composed of two performances of opus 1 11 that equally qualify as performances of that work and are coincidentally nuanced the same way and with the same degree of spontaneity. Following the same reasoning as I used earlier, we might conclude that neither performance could serve for the other.60 In that case, the difference between a work-performance and an improvisation seems diminished. About this thought-experiment, one might say, first, that identifying6l work-performances in this way would be a very specialized project, as compared with identifying them as, e.g., Cortot's regularly idiosyncratic way of playing the "Butterfly Etude." Either way, though, the effect would hardly be to break down the distinction between work-performances and improvisational performances. A performance by Cortot could not meaningfully be called "idiosyncratic" if construed simply as a bit of piano playing. It is idiosyncratic as a way of playing a sharply definable work, e.g., the "Butterfly Etude." So, the specificity that we won on behalf of such performances is compromised by the fact that a description of one of them would still be predicated upon its ordinary identification a s a performance of a given work. This fits the fact that when such personalized performances go so far as to intentionally depart from
363
scores, the gesture begs for an explanation. This could hardly be said of a jazz improvisation. We must avoid a possible misunderstanding about the difference between the two kinds of cases. News about a high degree of acoustic similarity between two performances of a given work would hardly come as a big surprise to their performers. By contrast, should Hawkins, for example, make such a discovery about the perfect pair described in section four, he would probably be puzzled. (Indeed, he might wonder if there was not something fishy about the situation.62) However, this contrast can be misleading. The difference between the two types is not simply a function of the degree of sheer acoustic difference between a pair of improvisationswhere each is based on the same tune, as in jazz, for instance-and the degree of acoustic difference between a pair of full-fledged performances of a musical work. Even if pairs of ordinary improvisational performances of the sort just described did as a rule diverge more from each other acoustically than pairs of performances of a given work do, we would certainly not assign the pairs to different kinds on that basis alone. Rather, acoustic differences are profiled in different ways in the two types of cases because the two are antecedently assigned to distinct categories. I might reasonably believe that, by contrast with a pair of up-coming improvisations, both an afternoon and an evening performance of opus 11 1 will generate much the same music. But that expectation would be a partial function of the fact that I antecedently categorize the pair as performances of a given work. Regarding them in that light, I might be inclined to overlook the multitude of acoustic differences that might differentiate the two. Admittedly, those differences would loom large should I consider the pair in altered terms-in light of their liveliness, or their idiosyncrasies, a s work-performances. However, this fact does not erode the boundary between the two types of music. Consider the different ways-and responses to those ways-in which my expectations can be unfulfilled in the two cases. Listening to a stretch of a work-performance, I might take a modulation as a cue for the recapitulation. Instead, suppose, I am surprised to hear some entirely new material. In some respects, my registry of the unexpected in this case is not utterly different from my confronta-
The Journal of Aesthetics and Art Criticism tion with the unexpected in the sphere of jazz improvisation, as any eccentric modulation in such a performance ill~strates.6~ However, the differences between the two categories in this respect are still overriding. At leisure, I can go back again-and again-and examine the unexpected twists and turns in a musical work. Soon, of course, I will learn to expect them. Further, the fulfillment of these expectations, on the whole, contributes positively, not negatively, to aesthetic response. In a fictional commentaryon Beethoven's opus 11 1, in fact-one of Thomas Mann's characters refers to this "here it comes again" feature of listening, as we might describe it.64 With improvised music, things are different. True, some improvisations may seem predictable just because they unfold with such striking musical "rightness." That they do so may seem uncanny-indeed, surprising-surprisingfor an improvisation, that is. By contrast, it would be strange to describe a piece of music as sounding surprisingly "right" for a piece of composed music. Furthermore, consider the difference between responses to performances of the two kinds that are predictable because they soundfamiliar. Such a feeling about an improvisation will naturally cast a shadow over it, since it will suggest the possibility that the performance might be more borrowed than an authentic improvisation ought to be. True, a sense of familiarity with a musical work can also be a tip-off that it might be plagiarized, or perhaps composed in the style of another work. But in the latter case, an alternative explanation is always available that is not available in the former case. A work can sound familiar because I really have heard it before, by means of earlier performances of it. By contrast, if an improvisation is genuine, I could not have heard it before. There may have been a time, even in Western music, when the distinction between the two kinds of performance would have made little sense. Theodor Adorno tried to identify underlying social reasons why, in the eighteenth century, musical practice became increasingly bound by the ever stricter authority of scores, with correspondingly less and less space for improvisation. He also noted that, alongside modern work-based practice, relics of a freer, more improvisational tradition continue to play a subsidiary role in later work-performance prac-
tice.65 Nowadays, if we are looking for echoes of a time before the distinction hardened as much as it has, we might well consider jazz.66 However, history did make its difference. That the dichotomy is not just an artifact of exaggerated zeal in border patrolling we can see by the way we describe the mixed cases-namely, as mixed. However, such cases also illustrate the fact that goals central to the one kind of performance can play relevant if subsidiary roles in the other. VII
The difference between the two categories of performance can be further profiled by reflecting on the way we regard performance activity in cases of each type. Activity may turn up in a work-performance as an aesthetic quality of the work itself-as exemplified by passages of music possessing a "running" character, for instance, or as a musical image of musical activity, e.g., the sound-picture of piano playing in SaintSaens's Carnival of the Animals and Strauss's Burlesque. (Such examples are not unknown in improvisational music, e.g., the vocal sounds that could be heard in Eric Dolphy's sax playing.) However, with some music-if not with all-we are expected to take an interest in the actual activity involved in the performance. One's recognition of the brilliance of a performance of Mephisto Waltz is partly a function of one's sense of the difficulties posed by the piece.67 However, as already suggested in sections three and four, informed responses to improvisational music involve a distinctive kind of registry of performance activity. With improvisational performance, one's concern is about how a player's on-the-spot decisions and actions create the very music unfolding as one listens. This focus is evidenced by the interest we take in collective improvisation in jazz, where players often try to unhorse fellow players by throwing them off balance.68 However, such cases are only striking instances of a general feature of improvisational performance. In typical jazz improvisations, players can be heard probing and testing possibilities latent in the music they are making.69 (The jazz pianist and composer George Russell refers to this self-monitoring feature of a typical improvisation as its intuitive quality.70) Correlatively,
Brown Musical Works, Improvisation, and the Principle of Continuity we take a special kind of interest in this activity-in how a performer is faring, so to say. If things are going well, I wonder if the player can sustain the level. If he seems to be getting into trouble, I worry about how he will address the problem. When he pulls the fat out of the fire, I applaud-as when Louis Armstrong rushes too quickly, if thrillingly, into the first notes of the introduction in his famous Okeh recording of West End Blues. Of special interest in that performance is, first, the spontaneity of Louis's gambit, and, second, the way he copes with his self-created problem with an on-the-spot restructuring of the meter.71 To feel the relevance of this self-monitoring feature of improvisational activity, compare a hypothetical species of improvisational performance in which players improvise-by pressing buttons on a machine, say-but in which they cannot take note of the accumulating musical effects until after the entire improvisational process has been completed. What would be missing in such an unmonitored improvisation, by contrast with an ordinary one, would be the feedback loop by which improvisers check what they are playing as they play it. The relevance of this self-monitoring is obvious when we consider that improvisers normally work on a background of music played by other people, e.g., a baroque thorough-bass, a jazz rhythm section, a band "riffing" behind the soloist, or other players improvising simultaneously. Of course, stretches of unmonitored music can be laid down in ordinary improvisations before becoming the object of this self-regard. But an improviser told she could not check what she plays of her performances until after she has finished playing would soon feel the difference between this kind of improvising and the standard way of doing it-even if she only plays alone. The cybernetic feature that this hypothetical contrast profiles in ordinary improvisational performance is undoubtedly an element in our concept of improvisational music as music created as it is played. As noted earlier, some improvisers manage to create well-organized musical material that sounds relatively free of the searching qualities I have described. (Bach's improvisations were said to come off as smoothly as if composed.72) In such cases, the self-monitoring feature is relatively inconspicuous. This fact does not under-
365
mine the over-all point, but illustrates it. It would be odd, surely, to describe a Chopin Ballade as sounding composed; it is composed. Smooth as some genuine improvisations are, knowing listeners will not regard it in the same way as they would regard a work-performance. As Nat Hentoff said about Paul Desmond, we could hear in his playing a mind moving with "eerie swiftness" as he spun out his remarkably consistent l i n e ~ . ~Hentoff's 3 point is that we attend to the way Desmond keeps his music on track even though he is creating it as he plays. It should be obvious, finally, that given an understanding of the kind of performance involved, we will not evaluate acoustic strings generated by improvisations and performances of works in all the same terms. Heard in a workperformance, the kind of notes played in the initial sequence of the opening notes of the Armstrong solo (and the metric adjustment that follows) might strike a listener as a gaffe it would be best to overlook or forget-as Gould's concerns about such matters reminds us. Heard as the improvisation it is, the performance has strikingly positive features, as we saw. VIII. EPILOGUE
A phenomenology of the experience of improvised music would profile what I called in section three presence. The sense that a unique, unscripted, event is taking place as I listen gives an improvisatory performance a sense of moment. I have to be there at the right time to hear a specific improvisation; yet, I cannot plan to hear that one. Being there at the right time, I have a special sense of that music's birth, as I listen.74 A feeling of indeterminacy is built into the situation and cannot be dissolved by an alteration of my epistemic relationship to the music. The excitement the experience engenders is enhanced as secondary regulative ideals make themselves felt-as when players take risks that are conspicuous.75 Is presence just a value that improvised music has? Not merely, since it is a feature both of imaginative and uninspired performances. It is a reflection of the way informed listeners address the music-a way that guides our judgments about the comparative merits of given improvisations. However, it is undeniable that presence is a kind of over-arching feature of the music
The Journal of Aesthetics and Art Criticism that can be savored, even in mediocre performances. There is nothing incoherent here. One can savor the cinematic characteristics of movies in general without judging any one of them to be a superb film among films. I said at the beginning that I would do my best to assume in the argument of this paper that recording technology does not exist. I failed, of course, since, in order to avail myself of a common currency of examples I had unavoidable recourse to recorded ones. This spots a problemnot a trivial one, alas.76 In a world without recording technology, we might never raise Goodmanian identity-questions about improvisations. But of course, we do live in such a world. The problem is that, as recorded, the music may have an entirely different phenomenology from that of the living thing. Indeed, it may have a different ontology. For we should consider the likelihood that recording technology has made possible a type of musical art hitherto unknown, and deserving of its own ontological analysis. One might call these new entities-completely unknown until the invention of recording technology-works ofphonogr ~ p h yThere . ~ ~ is every likelihood that these artworks do obey the principle of continuity. After all, what is the function of recording technology but to transform a living process into a soundprint, multiply it, and scatter it everywhere?78 (Works of phonography would seem to be a species of multiply-instanced artworks with interesting similarities to et~hings.~9) Among unconvinced readers of this essay would surely be numbered those whose first thoughts of improvised music come from the world of recorded sound. Their feelings about the matter are understandable. The situation is ironic. Our understanding and appreciation of the living music seem inextricably entangled with our understanding and appreciation of the canned version. Consider how much of our understanding of improvised jazz-indeed, jazz's evolving understanding of itself-is based upon its recorded history. Yet, once recorded, one of the most important groups of improvisations we know may become fodder for a species of artwork that does obey the principle of continuity. Once embedded in the grooves or bytes of recording media, improvised music is in danger of becoming seriously alienated from itself. It takes only one punch of
the "repeat" button to bring home the fact that the effect of recording on what I have called "presence" is corrosive. However, these matters are a topic for another essay.80 LEE B. BROWN
Department of Philosophy The Ohio State University University Hall 350 230 N. Oval Mall Columbus, Ohio 43210-1365
INTERNET:
[email protected] 1. I owe thanks to Stephen Davies, Emily Foster, David Goldblatt, Ted Gracyk, Roosevelt Porter, Diana Raffman, and to JAAC's outside readers for their helpful comments during the writing of this paper. 2. The theoretical policy to which I am objecting can also be formulated in terms of a pair of cooperating principles: a. All art comes packaged in artworks. b. All artworks obey the principle of continuity. If I am right, at least one of these principles has to give way. My reference to a principle of continuity is diagnostic. Nelson Goodman nowhere uses the phrase. However, it does seem to catch the spirit of a central organizing thought in his aesthetic system. 3. Goodman's system has often been challenged in its details, for reasons that have nothing to do with my present issue. I am interested in it only vis a vis improvisational music. Further, I am interested in it as perhaps the most worked out instance of a type of theory that it shares with other instances, e.g., Richard Wollheim's working hypothesis in Art and its Objects (Cambridge: Cambridge University Press, 1980). (Henceforth "AO.") In Wollheim's treatment, the territory of art is divided into two kinds of entities,physical objects and types. 4. Philip Alperson, for one, in his penetrating article "On Musical Improvisation," The Journal of Aesthetics and Art Criticism 43 (1984): 18-19. (Henceforth "OMI.") Alperson writes about the central categories of Goodman's system as if they might be congenial to improvisational music. His objection to the applicability of Goodman's system only gos: so far as to say that the two "stages" involved in Goodman's view of allographic art are blurred in musical improvisation. 5. The point is made in various writings of Arthur Danto and Richard Wollheim, for instance. See the latter's Essay I1 in AO. 6. See, for instance, Jerrold Levinson's "Autographic and Allographic Art Revisited," in Music, Art, and Metaphysics: Essays in Philosophical Aesthetics (Cornell University Press, 1990). (Henceforth "MAM.") 7. I shall have something to say about that matter at the end of the paper. 8. This is in line with standard usage, as in a recent textbook by King Palmer, The Piano (London: NTC Publishing Group, 1975), p. 109: "Improvisation is music which is created as it is performed, without previous preparation or detailed notation."
Brown Musical Works, Improvisation, and the Principle of Continuity 9. On the usefulness of the concept of regulative concepts in understanding the European musical tradition, see Lydia Goehr, The Imaginary Museum of Musical Works: An Essay in the Philosophy ofMusic (Oxford: Clarendon Press, 1992). (Henceforth "IM.") For Goehr, the concept of a musical work is a regulative idea, instanced both paradigmatically and derivatively. Although this essay does not show it, I am sympathetic with Goehr's general approach. I must defer to another occasion a consideration of specific points on which I might, in the end, have to disagree. 10. As in the final chapter of the otherwise insightful work by Winthrop Sargeant, Jazz, Hot, and Hybrid, 3rd ed. (New York: Da Capo Press, 1975). 1 I. See Bruno Nettl, The Radifof Persian Music: Studies of Structure and Cultural Context, rev. ed. (Champaign, Ill.: Elephant and Cat, 1992). 12. Bruno Nettl, "'Musical Thinking' and 'Thinking about Music,"' The Journal of Aesthetics and Art Criticism 52 (1994): 142. 13. David H. Rosenthal, Hard Bop: Jazz and Black Music (New York: Oxford University Press, 1992), p. 137. 14. Nelson Goodman, The Languages of Art: An Approach to a Theory of Symbols, 2nd ed. (Indianapolis: Hackett, 1976), p. 113. (Henceforth, "LA.") 15. Ibid., p. 113. 16. Ibid., p. 116. 17. See Goodman, L A , chap. 111 (Section 4), chap. IV, and chap. V (Section 2), in which the theories of notation, compliance, and their application to a theory of work-defining scores are explained. It should be noted that identity questions concerning the literary arts, the other main subclass of allographic art, will not concern us here. Identity questions concerning such works have to do with determinations about the identity of characters in a notation rather than the identity of compliants with notational characters. 18. In the context of the argument to follow, Goodman's remark ( L A , p. 1 18) that work-performances could be forged should not be misunderstood. Although he speaks of "forgery" here, he does not mean that such fakery would make performances autographic. It is only a category of artworks that could be called either autographic or allographic, and it would be a misunderstanding of the theory to regard a category of works as allographic, but authentic instances of them as autographic. The p. 118 remark only draws attention to a case of fakery that we might confuse with the kind that is of real theoretical interest. Someone could falsely represent an authentic work-performance as, e.g., a premiere. But such a performance would still be a genuine instance of the work of which it is a performance ifcompliant with the notational features of its score. 19. Levinson, MAM. p. 103. 20. See Goodman. L A , chap. 111 (Section 2). Goodman argues, controversially, that the potential perceptual difference between the original and the copy is grounds for an actual difference in one's perceptions of the two, and that this in turn is grounds for granting an actual aesthetic difference. Clearly, Goodman expects the moral of his discussion of referential forgery to rub off on inventive forgeries, as illustrated by his discussion of the famous van Meegerens forgeries of paintings in the style of Vermeer. 21. There may be no particular modal story to tell about the fireworks. In that case, the difference is that with improvised music, there is such a story.
367
22. Some Oriental works, in which a drawing is completed in a single spontaneous brush stroke, come close to fitting the description, as do some de Kooning paintingsat least on a certain view of them. 23. Such considerations, it might be noted, would also help explain why plagiarisms of Blue Flash-including selfplagiarisms-would not serve in place of originals either. 24. One might feel that modal conditions are relevant to questions of identity about all autographic art-ordinary paintings, for example, as well as Blue Flash. I must defer this larger issue here. However this plays out, I would argue that such conditions still impinge on improvised music in a distinctive way. Jerrold Levinson states that if confronted with a forgery of Giorgione's The Tempest, we would "acutely" feel the "absence of physical connection to Giorgione ..." (see MAM, p. 106). With a slight modification, Levinson's nice description would seem to apply to plagiarisms of autographic art as well. Suppose we were to learn that the "original" The Tempest, although painted by Giorgione, was a deceptive copy of a work by Lorenzo Lotto. Then, although the direct connection with the Giorgione would remain, we would feel a kind of "absence" in such a case anyway-namely, an absence of the right kind of connection. 25. See "OMI," p. 24, n. 21, where Alperson stresses the nature of improvisations as actit'ities. 26. Goodman, L A , pp. 114-1 15. 27. Goodman, "Comments on Wollheim's Paper," Ratio 20 (1978): 49. This piece is reprinted, in substance, in Of Mind and Other Matters (Harvard University Press, 1984) p. 140. (Henceforth, "MOM.") 28. For instance, that an inauthentic "painting" is not really art, and hence not really a painting either. Croce and Collingwood come to mind again. 29. Goodman, MOM, p. 140. 30. Of course, considered as shaping our conception of his oeuvre, Hackins's solo might not play the same role as Hawkins's solo would play in shaping our conception of his career. So let's agree to bracket and set aside this consideration. 3 1. Nothing in Goodman's analysis of musical works, it should be noted, entitles us to rule some scores out of court on grounds of the nature of their origins. 32. Goodman, L A , p. 179. One wonders what the criterion of "effectiveness" is here-beyond its usefulness for Goodman\ theory. 33. Goodman, L A , pp. 237-238. 34. Ibid., pp. 187-188. 35. Theodor Adorno, "On Popular Music," Studies in Philosophy and Social Science 9 (1941): 26. 36. Goodman, L A , p. 185. 37. I leave it open here how structure would be measured on this alternative. The matter is complex. Any piece of music-whether composed or improvised-possesses a multiplicity of structures, typically organized in hierarchies. See Kendall Walton, "The Presentation and Portrayal of Sound Patterns," in Human Agency: Language, Duty, and Value: Philosophical Essays in Honor of J. 0 . Urmson, eds. Jonathan Dancy, J. M. E. Moravcsik, and C. C. W. Taylor (Stanford University Press, 1988). Here, we are conjuring with structure in a specialized sense-a single structure that can be regarded as the real definition, so to say, of a musical work. 38. See Flint Schier, Deeper into Pictures: An Essay on
The Journal of Aesthetics and Art Criticism Pictorial Representation (Cambridge: Cambridge University Press, 1986), p. 29. 39. Levinson, MAM, p. 101. 40. Ibid., p. 101. Levinson's exposition addresses a number of concerns not directly relevant to the present discussion, including the application of Goodman's concepts to the literary arts. My summary and adaptation is a simplification of Levinson's many-faceted discussion. Without Levinson's restriction, this was the concept that was put in place in section two of this essay. 41. MAM, p. 98. Levinson works out his intentionalistic view in the essay cited and in other essays in MAM in terms of "compliance with a structure-as-indicated-by-X-at-t." While sharpening up Levinson's definition takes some doing, it could probably be construed so as to accommodate the case of the performer who plays the piece by ear, while excluding objectionable cases, e.g., the astral static. 42. It was this concept that we employed in section two of this paper. 43. Without Levinson's redefinitions of the one pair of distinctions, he argues, it would not be coextensive with the other pair. 44. Levinson, MAM, pp. 101-102. 45. Ibid., p. 97. 46. See Goodman and Elgin, Reconceptions in Philosophy and other Arts and Sciences (Indianapolis: Hackett, 1988), pp. 66, 73. 47. Goodman, MOM, pp. 139-140. 48. Ibid., p. 139. 49. Levinson, MAM, pp. 101-102, where he says of his revised version of Goodman's dichotomy that it is exhaustive for the "paradigm" arts. 50. Goodman, MOM, p. 139. 51. Goodman, LA, pp. 113, 264, to pick two references at random. 52. Ibid.. p. 121. 53. The view expounded by Alperson in "OMI." This idea may be the explanation for Alperson's congenial attitude toward Goodman's autographic/allographic duality. If an improvisation really is a species of composing, then what is composed must be a musical work obedient to the principle of continuity. 54. If I understand it, a suggestion affirmed by Alperson, "OMI," p. 26. 55. The concept can be amplified, of course. Such performances typically involve several improvisational solos. 56. As recorded at a live performance of "George Russell at Beethoven Hall," BASF recording MC ZS 125. 57. Bernard Holland, "Pavarotti Lip-Synchs to the Sound of Controversy," The New York Times, October 27, 1992, sec. B, pp. 1 , 4 . 58. Glenn Gould, The Glenn Gould Reader, ed. Tim Page (New York: Knopf, 1984). p. 246. 59. In an NPR broadcast of a talk by Gould, probably originally recorded by CBC. 60. Of course, a mannered work-performance by the likes of Vladimir de Pachman would not have played the same aesthetic role in his oeuvre as a perceptually identical one by Artur Schnabel would play in his. As with our perfect pair, I shall then discount contextual considerations of this sort. 61. In keeping with my thought-experiment, we would only be identifying them in the sense of giving a temporally indexed definite description of them, e.g., "Cortot's version
of 'Butterfly Etude' on March 3, 1932-not, anymore than with an improvisation, identifying them in the sense of providing a basis for their reidentification. 62. A listener might wonder if both performances might not have a common ancestor-that each was unwittingly plagiarizing a third performance. Hawkins might wonder if Hackins was not plagiarizing him. 63. Of course, jazz improvisations will almost certainly never seem unexpected because of the fact that they violate, or even test, the kind of large-scale formal principles of European composition. 64. Thomas Mann, Doctor Faustus, trans. H. T. LowePorter (New York: A. A. Knopf, 1948), pp. 51-57. 65. Theodor W. Adorno, Gesammelte Schrifren, 20 vols., ed. Rolf Tiedemann (FrankfurtIMain: Suhrkamp, 197086), vol. 18, pp. 753-754, for example. in an essay on the "social situation" of music. See Goehr, IM, for her important recent discussion of these matters. 66. Given his notorious contempt for that music, Adorno would not agree, of course. See my "Adorno on Popular Culture: The Case of Jazz Music," The Journal of Aesthetic Education 26 (1992). 67. A type of point nicely illustrated by Kendall Walton in his "Categories of Art," as reprinted in Aesthetics: A Critical Anthology, 2nd ed., eds. George Dickie, Richard Sclafani, and Ronald Roblin (New York: St. Martin's Press, 1989). 68. As noted by Alperson, "OMI," p. 28, n. 21. 69. Hence, the nice title given to a guitar solo recorded by Eddie Lang-"Feeling My Way." As Francis Sparshott puts it, a jazz improviser will often be "trying to do two things at once, changing his mind about where he is going, starting more hares than he can chase at once, picking up where he thought he had left off but resuming what was not quite there in the first place, discovering and pursuing tendencies in what he has done that would have taken a rather different form if he had thought of them at the time." See his The Theory of the Arts (Princeton University Press, 1982). p. 255. Cited by Alperson in "OMI," pp. 23-24. 70. Cited by Max Harrison, A Jazz Retrospect (New York: Crescendo Publishing, 1976), pp. 15-16. 71. The restructuring, as Gunther Schuller shows in a patient analysis, stands in a deliciously "right" relationship with the original meter. See his Early Jazz: Its Roots and Musical Development (New York: Oxford University Press, 19681, pp. 116-1 18. 72. Goehr, IM, p. 189. Goehr is citing the testimony of Lorenz Mitzler, as recorded by Hans David and Arthur Mendel, The Bach Reader: A Life of Johann Sebastian Bach in Letters and Documents (New York: W. W. Norton, 1966). 73. Cited by Ted Gioia, The Imperfect Art: Reflections on Jazz and Modern Culture (New York: Oxford University Press, 1988), p. 88. 74. In terms of his own, Jean Paul Sartre conveys this aspect of jazz in his vivid essay on the topic. See his "Jazz in America," Frontiers of Jazz, ed. Ralph de Toledano (London: Jazz Book Club, 1966). 75. The phenomenology is rather complex. We find ourselves slipping back and forth between our hopes for the ultimate quality of the music and our fascination with the activity by which it is generated-even when those actions appear to threaten the quality of the resulting music. Curiously, the strain contributes to the music's interest rather than detracting from it.
Brown Musical Works, Improvisation, and the Principle of Continuity 76. The implications of Edison's machine for musical aesthetics have scarcely begun to be digested. See Evan Eisenberg, The Recording Angel: The Esperience of Music fiom Aristotle to Zappa (New York: Penguin, 1988) for a refreshing discussion that will quickly disabuse anyone who does not think that sound recording has introduced whole new aesthetic dimensions into music and the philosophy thereof. The term "phonography" appears to be Eisenberg's invention. 77. Examples of such a category are discussed by Theodore Gracyk in his Rhythm and Noise: An Aesthetics of Rock (Duke University Press, 1995), chap. 1. Gracyk's description of this sphere was intended as an account of the ontology of rock music. (His view, to oversimplify, is that the artwork, in typical cases, is to be identified with the recording. The music of the Beatles' Sgt. Pepper's Lonely Hearts
369
Club Band album, for example, is not really performable, although it is reproducible. I see such works as examples of a much larger category of art, including, e.g., musique concrete, computer-generated music, etc. The delicate question is whether all musical material-including the improvised kind-must be converted into elements of works of phonography. 78. An instance of a general condition profiled in the writings of Walter Benjamin. 79. Whether allographic criteria would impinge upon decisions about the authenticity of instances as well is a nice question, though. 80. Such a paper, "His Master's Voice," by this author, exists in draft form.