• Author / Uploaded
• Yana

Citation preview

British Journal of Aesthetics, Vol. 43, No. 2, April 2003

THE INTERPRETATION OF MUSIC IN PERFORMANCE Paul Thom Musical performance, as an interpretive activity, has to be understood as relative to the material that is being interpreted. This material may or may not have the determinacy, fixity, and definitiveness of a work. Performative interpretation cannot be identified simply with what performers add to the material being performed. However, if interpretation is the assigning of significance, then in applying certain (theatrical, rhetorical, and biological) significance-endowing metaphors to integrated elements of a musical performance we commit ourselves to thinking of that performance as interpretive.

SOME philosophers think that among the forms that interpretation can take, there is one that occurs in and through the performance of works or other material, where such performative interpretations are enacted, not stated in speech. Other philosophers disagree. They think that, although it is true that in English and other languages expressions like ‘musical interpretation’ apply to what performers do, nevertheless, what is done by performers is not interpretation, in a sense of ‘interpretation’ that also applies in other recognized instances of interpretation— namely, those that are declared in assertoric sentences rather than displayed in performances. On the positive side of this dispute, Peter Kivy writes that, while a declarative interpretation says how something goes, an interpretation through a musical performance shows how something goes, and in both cases we are informed how something goes.1 By formulating things in this way, Kivy hopes to persuade us that performative and declarative interpretations, while specifically different, are generically identical. On the negative side, George Dickie writes: ‘Of course, we speak of a musician’s particular way of playing a piece as an interpretation, but this is something entirely different and not a declaration of meaning.’2 In saying that a so-called performative interpretation is ‘entirely different’ from a declaration of meaning, Dickie is rejecting precisely the thesis that Kivy espouses, 1

Peter Kivy, Authenticities: Philosophical Reflections on Musical Performance (Ithaca, NY: Cornell U.P., 1995), p. 137.

2

George Dickie, ‘Definition of “Art”’, in David E. Cooper (ed.), A Companion to Aesthetics (Oxford: Blackwell, 1992), p. 112.

© British Society of Aesthetics 2003

126

PAUL THOM

127

namely that there is a generic identity underlying specific differences among types of interpretation. For him, genuine interpretations are, one and all, declarations of meaning. The view that interpretation, strictly speaking, is a declarative activity, is also supported by Robert Stecker, who states: There is one question that can be given a quick answer and that concerns the way in which ‘interpretation’ is used in the following pages. It is confined to critical interpretation of artworks, leaving to one side interpretations that are given in performances of artworks. It is the name I give to any kind of assigning of meaning or significance to artworks.3

Both sides of the argument seem to me to be inadequately argued. Kivy’s suggestion about what it is that declarative and performative interpretations have in common, namely that they both inform us how something goes, is not sufficient to establish that both are genuinely forms of interpretation. It is not the fact that performative, or declarative, interpretations inform us how something goes, that makes them interpretations. This would be so only if it could be assumed that the conveying of information about how something goes, is of itself the giving of an interpretation. The trouble is that interpretation seems to come in af ter we know how something goes, and when we want to be told what something means. Belshazzar knows how the writing on the wall goes, in the sense that he can see the shapes; he calls the interpreter Daniel in to tell him what the writing means. It is possible to memorize a piece of music—and in that sense to know how it goes—without having formed any interpretation of it. On the other hand, Dickie appears to be simply assuming that anything other than a declaration of meaning can be dismissed as not genuinely interpretation—as ‘entirely different’. Rather than offering us any way of understanding performative interpretation, he is simply refusing to discuss it. In calling it ‘something entirely different’, he invites the supposition that he thinks it unworthy of consideration. Stecker’s discussion is more cautious than this, in adopting the stance merely of stipulating a meaning ‘for the following pages’, and in clarifying merely ‘the name I give’. But, despite this caution, Stecker’s comments are remarkably revealing. He confines the title ‘interpretation’ to a subset of declarative interpretations—namely, critical interpretations of artworks—but his stated ground for doing so might actually warrant a much wider usage. If it could be argued that musical performances actually ‘assign significance’ to the materials being performed, then the 3

Robert Stecker, Artworks: Definition, Meaning, Value (University Park, PA: Pennsylvania State U.P., 1997), p. 113. Levinson sometimes writes as if he is a supporter of this view, for example when he uses the expression ‘performative interpretation’ to mean a set of answers to the question ‘How should various prescriptions of rhythm, tempo, dynamics, and so on be precisely realized within their permissible ranges?’ But at other times he takes a performative interpretation to be ‘a considered way of playing a piece of music’—something that he says is not propositional. See Jerrold Levinson, ‘Performative vs. Critical Interpretation in Music’, in Michael Krausz (ed.), The Interpretation of Music: Philosophical Essays (Oxford: Clarendon Press, 1993), pp. 33–60 at pp. 35, 36, 38.

128

THE INTERPRETATION OF MUSIC IN PERFORMANCE

ground on which Stecker restricts interpretations to the declarative variety would turn out to warrant applying the term to musical performances. I will return to this thought. The idea that musical performance is an interpretive activity is one that can be explored from a number of vantage-points. Since interpretation is standardly thought of as somehow going beyond that which is interpreted, I shall begin by asking what it is that gets interpreted in musical performance. Then I will consider one simple answer to the question of what the extra elements might be in a musical performance, beyond that which is interpreted, that make it an interpretation. Next, I will examine some of the metaphors that are commonly applied to musical performances. Finally, I shall offer answers to questions of whether there is such a thing as performative interpretation, and what are the conditions of its possibility, relative to a certain concept of interpretation. I. WORKS AND OTHER MATERIAL

Gunther Schuller once wrote: Despite the limitations of musical notation, a score by Beethoven or Schoenberg is a definitive document, a blueprint from which various slightly differing interpretations can be derived. A jazz recording of an improvised performance on the other hand is a one-time thing, in many instances the only available and therefore ‘definitive’ version of something that was never meant to be definitive.4

That which is performed by musicians may be of various types. It may be more or less determinate.5 It may be more or less fixed. And it may be more or less definitive. Determinacy concerns the content of that which is performed, fixity concerns its vehicle, and definitiveness the type of authority it carries. At one extreme, there is what is commonly called a work for performance,6 where what is performed is thought of as being absolutely definitive, highly fixed, and fairly determinate. Performances of works can be definitive, but that does not in any way erode the work’s definitiveness. Similarly, performances of the work may vary over time, but the work remains fixed the same forever, or perhaps until there is no longer any practical way of recovering or accessing its set of 4

Gunther Schuller, Early Jazz: Its Roots and Musical Development (New York: Oxford U.P., 1968), p. x.

5

I think of a work for musical performance as a sequence of specifications for action. Such a sequence is indeterminate to the extent that there are actions—actions relevant to the project of executing that sequence of specifications—that are neither prescribed nor proscribed by the specifications in question. See Paul Thom, For An Audience: A Philosophy of the Performing Arts (Philadelphia, PA: Temple U.P., 1993), pp. 32ff.

6

Some musical works are not works for performance, for example some electronic music; and some musical works mix elements that are for performance with other elements that are pre-recorded, or electronically generated, or naturally occurring.

PAUL THOM

129

specifications. Finally, musicians might perform a work in different ways, but much of what they perform has already been determined by the work. This is not to deny that the performance is more determinate still than even the most highly determinate work. It is to deny that the increased determinacy of music in performance can make the work performed any more determinate. At the other extreme, that which is performed may not be highly determinate, may not be fixed, and may not be thought of as definitive. Such is the case with traditional folk-songs passed on in an oral tradition. The specifications for singing the song are not precisely specifiable. They are not fixed by the existence of an authoritative score. The music is not thought of as definitive, but as open to indefinitely many versions. There is a range of intermediate cases. Our three variables—determinacy, fixity, and definitiveness—are found in texts as well as in music, where for example God’s message to the Israelites contrasts with a teenager’s text-messages to a friend: while both are quite determinate, the former has great fixity and authority, while the latter has very little. In music, there may be different versions of the one work—the commercial release versus the director’s cut, the first versus the second folio, the Prague versus the Vienna version. Here, determinacy and authority are not in question, but fixity is. Then, there are pieces, like many of John Cage’s, which have a low degree of determinacy but are fixed in definitive scores. Jazz is different again. Gunther Schuller once proposed the following definition of a jazz singer: . . . a singer who is musically creative with the capacity to forge an original style (not merely mannerisms); one who can and does (but doesn’t always have to) improvise; and one who can reshape the material into something personal and individual.7

In this vein, he writes about Billie Holiday’s ‘ability to reshape (re-compose) a given song to make it wholly her own’: She did this on two levels—almost always simultaneously: on the larger structural level by freely reinventing both the melody and its rhythms, on the smaller level by embellishing these with her own wholly original vocal adornments.8

I think it is better not to try to stretch the concept of a work to cover all these diverse possibilities. Instead, I think it is preferable to follow Schuller in speaking of ‘material for performance’ rather than a work for performance. The term ‘material for performance’ can be taken as the more general term, with works for performance being a special case of such material. An analogous situation arises in connection with the playing or reproduction, as opposed to the performing, of 7

Gunther Schuller, The Swing Era: The Development of Jazz 1930–1945 (New York: Oxford U.P., 1989), p.528.

8

Ibid., p. 532.

130

THE INTERPRETATION OF MUSIC IN PERFORMANCE

musical materials. Theodore Gracyk has shown some of the ways in which preexisting material—for example pre-made backing tracks—may fail to have the status of a work.9 Such material—if it gets used at all—may undergo a succession of changes in the multi-layered process that leads to the creation of a recording, which is then a work. So, even though the scores of musical works are in varying degrees indeterminate, their existence as scores ensures their relative fixity, and the fact that they embody works ensures their definitiveness. By contrast, the material for certain types10 of jazz performance—the tune or the song—is not definitive in this sense, but can be varied from one performance to another in the hands of creative performers. Of course, the recording of such a performance is fixed, as are all recordings—even recordings of fixed works—because recordings themselves have the fixity, even if they do not always possess the definitiveness, of works. Recordings, however, are not works for performance, but (we may say) works for playing.11 So, it is indeed true that some jazz recordings constitute ‘the only available and therefore “definitive” version of something that was never meant to be definitive’; however, that of which they are a definitive version, and which itself was never meant to be definitive, is not the recording but the material (or the performance). II. THE RESIDUE THEORY

Beyond the work or the folk tune, or in general the material, the performance includes various elements; and it is among these, if anywhere, that we will find the performative interpretation. We need to ask two questions here: which aspects of the performance form part of the performative interpretation, and what is it about those elements that make them interpretation? The former is the question of extension, the latter the question of intension. Let us start with a simple suggestion. Could it be that the interpretive element in a musical performance is a kind of residue, namely everything about the performance that is not contained in the material being performed? This idea seems to be accepted by Stephen Davies when he writes: . . . faithfulness to the work is consistent with significant differences between performances, and these differences will be attributable to the performer’s interpretation.12 9

Theodore Gracyk, Rhythm and Noise: An Aesthetics of Rock (Durham, NC: Duke U.P., 1996), pp. 46–50.

10

Not all. Compare Schuller on Duke Ellington’s works: Gunther Schuller, Musings: The Musical Worlds of Gunther Schuller (New York: Oxford U.P., 1986), ch. 9.

11

See Thom, For An Audience, pp. 55–57.

12

Stephen Davies, ‘The Multiple Interpretability of Musical Works’, in Michael Krausz (ed.), Is There a Single Right Interpretation? (University Park: Pennsylvania State U.P., 2002), pp. 231–250 at p. 238.

PAUL THOM

131

This cannot stand as an answer to the intensional question. An understanding of performative interpretation merely as that (in the performance) which is not contained in the musical material cannot supply any reason for thinking that performative interpretation is or embodies interpretation. Nor can it explain why we value performative interpretations, so understood. In order to supply a reason for thinking that performative interpretation understood in this way falls under some general concept of interpretation, we would have to subsume the idea of that in the performance which is not contained in the material being performed under this general concept of interpretation. This is what is lacking in the current proposal. It fails to imply anything positive, or of positive value, about performative interpretation, let alone that such interpretation has anything in common with other types of interpretation. The extensional question has two sides. It is indeed true that the actions that are constitutive of a given performative interpretation are not all specified by the material being performed. That this is so becomes absurdly clear if we try to imagine the opposite. Imagine a piece of music that specifies that a particular action be done and specifies that this be done by way of interpreting the music. I do not mean that the score at a certain point includes a directive to interpret—a directive such as cadenza or ad libitum. There is a clear sense in which such a directive specifies no particular action. By contrast, what I am asking you to imagine is a score that at a certain point does specify a particular action, at the same time as directing that this action be executed by way of interpretation; and I take it that the absurdity of such a supposition is palpable. So it is true that the actions constitutive of a performative interpretation are not all mandated as such by the material that is being performed. But this is not the same as saying that the performative interpretation includes everything about the performance that is not specified by that material. The latter is a distinct assertion—one whose denial would not obviously lead to absurdity. To deny it would simply be to state that there is some feature of the performance that is neither mandated by the material itself, nor part of its performative interpretation. What might such a feature be? Stan Godlovitch’s account of musical performance is helpful at this point. Godlovitch discusses a number of ways in which a musical performance can fall short of the ideal,13 because of wrong notes, incompleteness, temporal discontinuity, changing personnel, changing audience, or interpretive inconsistency. These are all plausible candidates for features of a performance that are contained neither in the material itself nor in its performative interpretation. Even the last of these—the case where performers change their interpretation in midperformance—is arguably not an instance of performative interpretation, if by this last phrase one means a single interpretation. 13

Stan Godlovitch, Musical Performance: A Philosophical Study (London: Routledge, 1998), pp. 31–38. See also Levinson, ‘Performative vs. Critical Interpretation in Music’, p. 36.

132

THE INTERPRETATION OF MUSIC IN PERFORMANCE

A defender of the residue theory might very well respond to this by saying that it is not surprising that a performance’s imperfections do not turn out to be part of the performative interpretation, since there is a sense in which they are not really part of the performance, and that therefore the question, whether there is some feature of the performance that is neither mandated by the material itself, nor part of its performative interpretation, is still open. So let us look at another implication of the residue theory. If the content of a performative interpretation comprises everything about the performance beyond what is in the material, then it must include those aspects of the performance that constitute what we might call its rich texture. I mean here highly specific features of a performance that cannot be specified in a score, features that may be unique to an individual performer. Roland Barthes wrote about the ‘grain of the voice’ in the recordings by Charles Panzéra, contrasting this with a different vocal texture in Dietrich Fischer-Dieskau’s recordings.14 In a similar vein Schuller writes about Billie Holiday’s ‘unique olive-toned timbre’.15 Such features are of crucial importance, not only because they enable us to distinguish one vocal performance from another, but also because in many cases they are among the singularities that we prize most about particular performances. True; but are they part of the performer’s interpretation? A defender of the residue theory might argue that features like this are not part of the performative interpretation, or of the performance, because they are not chosen by the performers. Rather, they are just the way performers are, given that they have trained their various bodily attributes for the task of performance. Following this line of thought, the defender of the residue theory might say that that theory has to be understood as applying, not to absolutely all features of the performance but only to those that are chosen by the performers. I believe that this way of putting things makes a wrong assumption about the logical grammar of the verb ‘to choose’. It assumes that the verb takes a nominal object—we choose things, in this case features of our performance. But in truth, the object of the verb ‘to choose’ is not nominal but propositional—we choose that something is to be the case. This implies that there may be a plurality of objects to my choice, all of which in some way involve my singular personal features. I cannot choose whether I have those bodily features. But I might choose either to include or not to include them as features of my performance. To choose that they be part of my performance would be to choose to highlight or project them in the performance; and this can be done even in relation to personal attributes over whose possession I have no control. It should not be assumed, however, that to incorporate some of my own bodily attributes in my performance is necessarily 14

Roland Barthes, ‘The Grain of the Voice’, in Image–Music–Text, trans. Stephen Heath (London: Fontana, 1977), pp. 183–185.

15

Schuller, The Swing Era, p. 530.

PAUL THOM

133

to incorporate those features in a performative interpretation. We have not as yet been given any justification for using the word ‘interpretation’ in relation to musical performances. In fact, a musical performance may easily fail to qualify as embodying any sort of interpretation on the ground that it lacks the kind of integrity that we demand of interpretations. It may be disfigured by quirky mannerisms that are not integrated with the musical material. The singer may have failed to make the song her own. She may (as Schuller reminds us) have failed to turn mannerisms into a style. In the light of all this we should not yet accept the proposition that, extensionally speaking, the performative interpretation of musical material comprises everything—even everything chosen— about the performance that is not specified by that material. We need to identify features of performative interpretation that link it with a more general, and positive, concept of interpretation, and that link it with the aspects of interpretation to which we attach value. What can be said about the general concept of interpretation? III. A CONCEPT OF PERFORMATIVE INTERPRETATION

Elsewhere,16 I have argued that the most general feature of interpretation to which we attach value is its capacity for making sense of that which is interpreted by representing it within a significance-system. In short, to interpret something is to make sense of it. Thus, I am in agreement with Stecker’s definition of interpretation as assigning significance. In contrast to Stecker, however, I have argued that there exists a spectrum of ways in which this can be done, ranging from adequational interpretations (where the bearer of sense pre-exists and its sense is discovered in the process of interpretation),17 through to constructive interpretations (where the making of sense is at the same time a making, or a changing, of the bearer of that sense). Constructive interpretations predominate in the performance of musical material lacking the status of a work. Adequational interpretations are most common in the performance of musical works, but music-lovers will be aware that constructive interpretations are found here too, in improvisations based on works or in deliberately transgressive performances such as many of Stokowski’s. What many music-lovers are not aware of, because they have no knowledge of the scores on which performances are based, is that many performances that present themselves as adequational are in reality more or less constructive in their approach; and we have to thank Gunther Schuller for providing detailed evidence of this in his book on the art of conducting.18 In the 16

Paul Thom, Making Sense: A Theory of Interpretation (Lanham, MD: Rowman & Littlefield, 2000), ch. 2.

17

The terminology comes from Joseph Margolis, Interpretation Radical But Not Unruly: The New Puzzle of the Arts and History (Berkeley and Los Angeles: University of California Press, 1995), pp. 21ff.

18

Gunther Schuller, The Compleat Conductor (New York: Oxford U.P., 1997).

134

THE INTERPRETATION OF MUSIC IN PERFORMANCE

case of elucidatory interpretations, it is a sine qua non of a successful interpretation that it is faithful to the musical material being performed. But that is not the end of the story. The task of making sense of the music cannot be achieved merely by reiterating what is in the score. The interpreter always goes beyond the score, even in elucidatory interpretation. Thus an injunction to play the music as it is written—if it is to be understood as demanding good performance—should be taken as stating a necessary but not a sufficient condition. If it were supposed to be a sufficient condition of good performance, then it would be a statement of a kind of musical fundamentalism. If we assume this account of interpretation, the question of whether performative interpretation is really a form of interpretation comes down to this: are there elements in what we call performative interpretation that are capable of making sense of the work or materials performed? (I should emphasize that I choose the somewhat vague term ‘making sense’ deliberately. I do not mean to imply that musical performance has meaning in the sense of standing in some relation of denotation to extra-musical entities. Musical performances make sense by virtue of certain qualities that they have, not by virtue of any relation they stand in to extra-musical entities.) We already know that if there are such elements then they will be a proper subset of those performative elements that are not in the work or materials. What we have not yet established is whether there are elements in the performance of music that link that activity with interpretation understood as the making sense of given material. Progress can be made by re-reading Roger Sessions’s lecture on the performer.19 Sessions portrays musical performance as an activity that has dual aims. On one side, the performance aims to be faithful to the material being performed; this is the aim of fidelity. On the other side, the performance aims to be informed by a quality or qualities that he identifies variously as personality (ME, pp. 78, 84), eloquence and conviction (ME, pp. 78, 85), or ‘the vitalizing energy of a genuine impulse’ (ME, pp. 84, 86). W ith reference to these two aims, Sessions summarizes two broad ways in which a performance may fail: ‘W ithout fidelity a performance is false, without conviction it is lifeless; in other words, it is hardly music’ (ME, p. 78). Looking at things another way, we may say that a successful performance must exhibit both fidelity and creativity. Clearly, this account is biased towards the case where what is performed is a musical work. But that is not what interests me at the moment. What interests me is the metaphors that Sessions uses to characterize musical creativity. These metaphors fall into three groups. First, there are terms that are drawn from the art of the actor. In this group falls the idea of personality. Second, there are terms deriving from the art of rhetoric—terms such as ‘eloquence’ and 19

Roger Sessions, The Musical Experience of Composer, Performer, Listener (Princeton, NJ: Princeton U.P., 1971), ch. 4 [hereafter ME].

PAUL THOM

135

‘conviction’. Third there are the biological ideas of vitality and energy. Terms drawn from the language of the theatre are commonplace in relation to musical performance: we speak of musicians as exhibiting ‘presence’, as performing with ‘character’. Indeed, such terminology naturally suggests itself to the extent that we think of a piece of music as analogous to a person—in respect of expressive potential and narrative identity.20 For, if we think of the music in this way we should think of the musician as one who adopts a persona and a narrative identity; and this is to think of the musician (even the instrumental musician) as a kind of actor. Equally commonplace in descriptions of musical performance is the language of rhetoric. W ithout giving it a second thought we speak of ‘pause’, ‘climax’, or ‘dialogue’ in musical performance.21 The general attributes of persuasiveness, clarity of structure, effective gesture and emotional power, along with specific rhetorical figures of patterning, gradation, and emphasis, all apply to musical performance—in some cases in historically very specific ways. The comparison between the musical performer and the orator is an old one, much used in the baroque period. Quantz, for example, opens his chapter on good execution in general in singing and playing with the observation that ‘Musical execution may be compared with the delivery of an orator.’22 He and other theorists of the period developed the rhetorical metaphor in some considerable detail. Schuller freely uses the language of rhetoric in analysing Billie Holiday’s style: . . . the reshaping (mostly the stretching) of rhythms to press the maximum of expression out of the words; the little dips, scoops and sags with which she inflects (and thereby emphasizes) particular words and pitches; the strongly felt implicit beat and swing, for all her exterior rhythmic freedom . . . ; the direct, unsentimental, all the more poignant delivery.23

Thirdly, biological ideas such as that of vitality are connected with the very idea of a ‘live’ performance. Vitality is clearly a valued quality in performances and it can take various forms, only one of which is evidenced in the following description of the piano-playing of W ilhelm Backhaus: Backhaus in his best playing displayed a wonderfully muscular animality in which a superb life force was at work.24

The liveness of a live performance is seen by Hegel as having to do with the very 20 21

22

23 24

See Stephen Davies, Musical Meaning and Expression (Ithaca, NY: Cornell U.P., 1994), pp. 367–368. For an excellent example of the latter see Levinson, ‘Performative vs. Critical Interpretation in Music’, pp. 40–41. Johann Joachim Quantz, On Playing the Flute, trans. with notes and an introduction by Edward R. Reilly (London: Faber & Faber, 1966), ch. xi. Schuller, The Swing Era, p. 530. David Dubal, Liner notes to Brahms (Grand Piano Series), Nimbus Records NI8806.

136

THE INTERPRETATION OF MUSIC IN PERFORMANCE

stuff from which the performance is constituted. He saw the performer as entering into the work of art in his full humanity,25 and thus saw the performed work as having something living woven into its very fabric. The metaphors of the actor, of the orator, and of life are all highly charged ideas. The idea of the actor gathers around itself a rich network of signifying elements, including the idea of the theatrical, of playing a character, of the comic and the tragic, and of emotions communicated. In a more extreme vein, Nietzsche says of the actor that he is ‘no longer an artist, he has become a work of art’.26 The idea of the orator has similar connotations to that of the actor. The orator is ‘moved by a divine influence’ (Plato),27 and has mastery over ‘the hearts of his listeners’ (Quantz).28 Life, according to the Greek philosophers, is something mysterious, and akin to the divine—something that is explained by the presence of a soul (Plato).29 Against this background, if the performance gives life to the musical material, then it is the music’s soul. If the metaphors of the actor, the orator, and of life itself are charged with significance, and therefore with value, it follows that by applying these metaphors to musical performances we attribute significance and value to those performances. (We attribute significance to the performances, but doing so does not commit us to saying that those performances stand in some relation of significance to an entity that we can regard as their ‘meaning’.) Given the account of interpretation that I am advocating, doesn’t it then follow that by using these metaphors we are committing ourselves to speaking of those performances as embodying performative interpretations? Not quite. In order for a musical performance to qualify as a performative interpretation, it would have to be more than a performance charged with meaning. It would have to meet two further requirements. Firstly, the meaning in question would have to be projected by the performers as part of their performance, rather than being merely in the mind of a critic, or for that matter merely in the minds of the performers.30 This is because a performative interpretation needs to be a performance, and performance involves projection. Secondly, in order for this meaning to be part of a performative interpretation of the material being performed, it would have to be integrated by the performers with other aspects of 25

G. F. W. Hegel, Aesthetics: Lectures on Fine Art, trans. T. M. Knox, 2 vols (Oxford: Clarendon Press, 1975), vol. 2, part 3, sec. 3, ch. 3 C.C. §2.

26

Friedrich Nietzsche, The Birth of Tragedy, or Hellenism and Pessimism, in The Birth of Tragedy and The Case of Wagner, trans. with commentary by Walter Kaufmann (New York: Vintage Books, 1967), p. 37. Plato, Ion 533d, trans. Percy Bysshe Shelley in vol. 7 of The Complete Works of Percy Bysshe Shelley, ed. Roger Ingpen and Walter E. Peck (New York: Gordian, 1965), 533d. Quantz, On Playing the Flute, p. 119.

27

28 29 30

Plato, Phaedo, trans. and ed. David Gallop (Oxford: Oxford U.P., 1993), 105c. To project an interpretation is rather more than to reflect what is in the performer’s mind. This latter expression is Levinson’s (‘Performative vs. Critical Interpretation in Music’, p. 38).

PAUL THOM

137

their presentation of the material. This is because a performative interpretation needs to be an interpretation, and interpretation generally aims at coherence.31 Coherence may be global or local, depending on whether we are talking about the performance of the whole piece or of a particular passage; and it can be achieved in a performance in a variety of ways, for example through unity of style, or by using means–ends relationships like those that Schuller finds in Billie Holiday’s performance, speaking of ‘the reshaping . . . of rhythms to press the maximum of expression out of the words’, and ‘the little dips, scoops and sags with which she inflects (and thereby emphasizes) particular words and pitches’.32 So the most we can say is that to apply the metaphors of the actor, the orator, and life itself—or indeed any other metaphors charged with significance—to the performance of music is potentially to categorize that performance as interpretive; but if that potential is to be actualized, then the further conditions of projection and integration have to be met. What we have arrived at is a statement of the conditions of possibility of performative interpretation. At the same time it is abundantly clear that those two conditions are actually met at least some of the time: the kind of significance that is borne by the metaphors in question is projected by performers and is integrated into the presentation of musical material, at least some of the time: any sensitive music-lover knows real-life examples. This, however, is not to say that when a performance does incorporate a performative interpretation, everything in the performance forms part of that interpretation; because alongside the performative interpretation the performance may include sundry superfluities. Finally, because what we have arrived at is in part an account of the value of performative interpretation, it goes some way towards explaining why we should care about these philosophical questions.33 Paul Thom, Division of Arts, Southern Cross University, PO Box 157, Lismore 2480 NSW Australia. Email: [email protected]

31

In this respect the concept of performative interpretation that I am defending differs from Levinson’s. According to him, every performance at least corresponds to a possible interpretation, even if it is not an interpretation that the performers intend or that is original to the performers. A performer’s performative interpretation, for Levinson, just is ‘the way he or she plays the work in question, which though it might conduce to certain analytical conceptions of the work rather than others, is compatible with a number of them, or with no conception at all’ (Levinson, ‘Performative vs. Critical Interpretation in Music’, pp. 47, 54–55).

32

Schuller, The Swing Era, p. 530. An earlier version of this paper was presented to a symposium on musical interpretation held at the Curtis Institute, Philadelphia, on 2 February 2002. I thank Jacques Catudal, Michael Krausz, Joseph Margolis, and Gunther Schuller for helping me to clarify my thoughts on that occasion.

33