ABSTRACT: This essay is a critical examination of the claims of deconstruction and "textuality." The three sections of the argument concern: I. Derrida, in one of his most frequently cited deconstructions; II. the problems of Saussurean semantics and post-structuralist textuality; III. broadly deconstructive strategies to be found in musicological writings.
"[A]ny convincing challenge will have to do more than just rehearse what amounts to a litany of anti-deconstructionist id�es re�ues. It will need to show precisely where Derrida's arguments go wrong; where he misreads or misconstrues his philosophical source-texts; or where the claims of deconstruction themselves fall prey to a better, more adequate, historically informed, or cogent theoretical critique."I accept that these conditions are fair, and I am happy to abide by them in this paper. What I shall attempt here is a critique first of Derrida himself, secondly of textuality and interdisciplinarity, which simplify and distort Derrida's writings, and thirdly the use that certain musicologists have made of these arguments during the last decade, in writings that may today be generally dubbed the New Musicology (a label bound to disappear before long). If I were to limit my criticisms to the post-structuralist assumptions which pervade much recent work in musicology and the other humanities, I might be accused by some of attacking too easy a target. It is for this reason that much of the present paper confronts the most important source for such ideas, however simplified and debased they may have become in their dissemination. I will remind the reader at this point that the necessary restrictions on space preclude any exhaustive treatment of the issues here; according to the Socratic principles examined in the first part of the paper, I would encourage further dialogue within the arena of mto-talk.
 Let me begin with a cautionary tale. In Derrida's essay, "Plato's
Pharmacy," from the collection Dissemination, we find one of the
most frequently cited examples of a metaphysical text which contains the
seeds of its own deconstruction
 The Phaedrus is an examination of rhetoric, which argues that the art is pernicious insofar as it ignores reason and knowledge. Contrary to Derrida's reading, Plato does not present rhetoric and philosophy as a binary opposition, since he is condemning only some rhetoric, while generously conceding that, for example, the rhetorical school of Isocrates, one of Plato's own rivals, has much of worth to offer. So Plato is not interested in making a rigid and unyielding distinction between the logic of an argument, and the way in which it is presented. The passage Derrida discusses is centered upon an Egyptian myth which Socrates relates to Phaedrus, the superficial young orator, by way of an illustration of his arguments on writing. The god Theuth came to Thamus, the king of Egypt, offering a long list of his inventions as gifts. Thamus first listened to the account Theuth gave of each of his gifts, accepting some, while rejecting others, according to his judgement of the merits of each. When Theuth arrived at the gift of writing, he said, "Here is an accomplishment, my lord the king, which will improve both the wisdom and the memory of the Egyptians. I have discovered a remedy for memory and wisdom." Thamus rejects the gift, with a weighty reproof, which is worth quoting in full.
"Theuth, my paragon of inventors, the discoverer of an art is not the best judge of the good or harm which will accrue to those who practise it. So it is in this case; you, who are the father of writing, have out of fondness for your offspring attributed to it quite the opposite of its real function. Those who acquire it wil cease to exercise their memory and become forgetful; they will rely on writing to bring things to their remembrance by external signs instead of on their own internal resources. What you have discovered is a remedy for recollection, not for memory. And as for wisdom, your pupils will have the reputation for it without the reality: they will receive a quantity of information without proper instruction, and in consequence be thought very knowledgeable when they are for the most part quite ignorant. And because they are filled with the conceit of wisdom instead of real wisdom they will be a burden to society."(Thamus seems to have great insight into the note-taking habits of some students, who write furiously and memorise great quantities in the hope that these activities will serve as a substitute for understanding.) The word I rendered as "remedy" is in the original "ph�rmakon," which the lexicons define as alternatively a curative medicine, or a poison. Indeed, the two meanings are preserved in two modern Greek cognate words: ph�rmakon now means medicine alone, while the verb pharmak�no is to poison. This is the crux of our present argument.
 Now according to Derrida, this apparent ambiguity renders the
meaning of the word "undecidable." Let us examine this on two grounds.
First, within the present context, "ph�rmakon" clearly carries its
positive meaning, since Theuth is hardly going to persuade the king by
offering him a poison. Derrida realises this, but he wants to argue that
the other, opposed meaning insinuates itself upon the intended meaning.
Secondly then, might the word's use in the language settle the question of
meaning? I would argue that it does, and this is exactly what Derrida is
constrained to ignore due to his whole approach to language. For the
apparent double-meaning is not a binary opposition operating solely
within a closed linguistic system, but is linked rather to features of the
extra-linguistic material world. As Paracelsus had already said prior to
Plato, "it depends only upon the amount whether a poison is poisonous or
 So far we have one pivotal example of Derrida's supposed "close reading" proving quite otherwise. Pivotal because this passage supposedly lays bare the "inaugural moment in the 'logocentric' epoch whose effects reach down from Plato to the present day." It would be unlike Derrida, of course, to engage in a simple search for origins; what he is claiming is not so much that Plato is the creator of the "logocentric epoch" and the source of all its problems, but rather that the Phaedrus passage is symptomatic of a kind of necessity imposed upon us by language, whenever we engage in the search for philosophical truths. But what kind of necessity is this? What are the forces which according to Derrida, always already subvert the meaning of texts? By way of explanation, let us look at Derrida's next step in his discussion of the Phaedrus passage, which introduces the mechanism of intertextuality.
 Derrida claims that the forces of language bring a cognate word to bear upon the poison/cure discussion. He claims that this other word, which means either magician or scapegoat is the key to this discussion, and since words, according to Derrida's post-Saussurean semantics, gain their identity and meaning through all the other, absent terms of the language, it is of no matter that this word fails to appear in Plato's text. A magician, according to the lexicon, can be a user of potions, possibly to poison others, while Derrida notes, from a reading of Frazer's The Golden Bough, that in the earlier history of Athens, certain persons living within the polis would be ejected as a means of averting an impending disaster. He elaborates upon this notion of the scapegoat, claiming that the word is both within Plato's text, due to the presence of the supposedly related "ph�rmakon," but also absent, excluded, just as the scapegoat was excluded by the laws of the polis. The word is then drawn up into a discussion of the inside/outside opposition which Derrida again thinks is pervasive and inevitable within logocentrism.
 I would contend though, that Derrida is again mistaken in a way
which is symptomatic of his approach to language. You will notice that I
have so far avoided mentioning the Greek word for magician and
scapegoat. This is for a good reason. Since Attic Greek in its written
form had no accents, only the spoken language revealed which syllables
were stressed. Because of Derrida's charge that speech has been
privileged throughout the "logocentric epoch" this issue is of central
importance. What has escaped Derrida here, is the fact that magician and
scapegoat were not rendered by the same Greek word, but rather by two,
and the distinction is only apparent from speech
 The Phaedrus is in part Plato's defence of the Socratic dialectic method. In opposition to this, he places the rhetorician's speech, written out beforehand, which can be delivered by another on a different occasion, to an audience quite different from that originally intended. There is of course a certain irony in the fact that Plato is himself leaving a piece of writing as a hostage to fortune, in praise of Socrates who resisted the temptation to depart from spoken dialectics, and so preserved nothing in writing; and indeed this irony can hardly have escaped Plato. Socrates, in the dialogue, is making a complaint of a very practical nature: that speeches already written out cannot answer back, and cannot possibly adjust themselves to the prejudices or ignorance of all future audiences. There is no binary opposition here, though, because, as I have said, Socrates allows for a rhetoric, an art of speech writing which is carefully attuned to its audiences' needs, not to flatter, but to persuade. But of course the dialectical method remains the ideal for Socrates, since minds interact directly, and every fallacy, over-hasty conclusion or misunderstanding can in principle be addressed immediately by the other speakers. Socrates is also concerned that knowledge gained purely from written sources will not necessarily be accompanied by understanding; the reader may accept the conclusions of a detailed argument dogmatically, without following the argument. (This matches my own teaching experience, since, for example, a few students in my aesthetics courses have been horrified by the notion that the quality of the argument is to be valued above the conclusions reached, since they have been accustomed to the neat packaging of supposed facts, which can be memorised for examination purposes - in which case, understanding is a mere luxury.) This is the kind of practical problem Socrates is addressing, and to treat it according to a hermeneutics of suspicion that searches at all costs for a privileging of speech over writing seems to miss the point, without any substantial gain to compensate - this is in addition to the fact that the alleged close reading turns out to be gravely flawed in ways which directly undermine Derrida's deconstructive interpretation. To put a contemporary gloss on Socrates' argument, we might offer e-mail as a form of writing which would completely satisfy all his strictures, since the writer can answer back; the academic e-mail discussion list, at its best, would have pleased Socrates greatly, and might even offer an advantage which he would acknowledge: answers can be more carefully framed, since we can ponder over them at far greater length than would be tolerable in spoken dialogue.
 So here we have an example of writing which co-incides with Socrates' discursive ideal. We should also remember that the examples of writing Socrates was considering were themselves to be communicated in speech, since rhetoric was the topic under consideration. We ought further to recall that silent, inward reading is a relatively modern phenomenon, which was rare enough to invite comment only five centuries ago; even with an audience of one - the reader alone - written texts were still associated with the speaking voice. The concert platform, by the way, displays its own ethic on written music: the consensus since Liszt has demanded that a recitalist memorise the score on the understanding that only then has the music been internalized.
 Derrida is also misleading in his account of the supposedly canonical
reading of the Phaedrus which his deconstructive reading is
supposed to overthrow. The scholarly consensus, according to Derrida, is
that the Phaedrus is one of Plato's poorest products, either immature, or
written in his dotage, and in either case not worthy of the same attention
as most of his other works. This, however, is a travesty of contemporary
Platonic scholarship, which has established conclusively that the
Phaedrus is a late work, and it is admired for its literary
qualities, not condemned. Glancing at the translation most widely
available in the U.K., by Walter Hamilton
 By way of summary then, Derrida, in his treatment of the Phaedrus, first misrepresents the modern scholarly consensus on the text; secondly, he declares that his own close reading will overthrow this alleged canonical interpretation, by respecting the details of Plato's text, but on closer inspection this reading is found to be less than close, and subject to external purposes, leading to a reading which the details of the text cannot sustain, such as various imposed binary oppositions; thirdly, he bases his deconstruction of the text on the semantic undecidability of a word which is perfectly decidable on non-linguistic grounds; fourthly, he imports another word into the text, justifying this strategy on the basis of a dubious necessity arising from his semantics of diffe'rance; fifthly, this absent word, supposedly a key to the passage, turns out to be two words, spoiling the putative relationship, and ironically proving symptomatic of a certain blindness in Derrida's approach.
 Saussure explored the phonemic level of language in these last
lectures, and concluded that phonemes belonged to a system of
oppositions, allowing us to distinguish cat from mat. His daring
speculative leap was to suggest that the semantic level of language was
also governed by a system of oppositions, and it was this idea which made
the structuralist enterprise in the social sciences and humanities possible,
giving us L�vi-Strauss's structuralist anthropology and Nattiez's
structuralist musicology. Saussure was suggesting that words gained their
meaning purely from an intra-linguistic system of differences, each
concept gaining its identity from all that it was not, that is from all the
other concepts. In the hands of Saussure's successors, such as Jakobson,
this speculation began to appear unsustainable. Jakobson split the
phonemic atom in the 1930s, showing that each phoneme consisted of a
number of simultaneous material distinctive features; the science of
phonetics can describe these features in terms of their acoustics, or in
terms of the physiological changes in the speaker's buccal cavity. This
analysis of the phoneme removed the motivation for the oppositional
model in linguistics and the structuralist enterprise in general, but its
implications were not immediately felt
 Derrida, in Grammatology, ignored the Chomskian
revolution and took Saussure's lectures of 1917 to be the last word on
language and meaning. Aside from the shortcomings mentioned above,
this also meant the acceptance of the potentially most misleading of
Saussure's methodological choices: the bracketing of the language- and
mind-independent world, in order to present language as a self-sufficient
system. But once the external world is removed from the account, for
methodological reasons, it is hard to see at what point it can logically be
re-introduced. And so we arrive at a kind of linguistic idealism which
treats the world as a creation of language. While Derrida himself would
not seem to endorse this doctrine, it is implicit in much post-modernist
writing in the humanities today. Saussure, in his talk of the arbitrary
connection between the signifier and the signified in no way introduced a
revolutionary idea: he is saying simply that the sound of a word has no
necessary connection with its associated concept - this much should be
evident to any child when she first notices that there exist languages other
than her own. But once this association has been made, the connection
between the word and the objects in the extra-linguistic world to which it
may refer is not at all arbitrary. Most literary theorists' accounts of
Saussure are hopelessly confused on this point
 Derrida takes as his starting point Saussure's semantics of pure difference, which, as I said, leaves out the referential function of language. At this point, since Derrida thinks it should be taken seriously, I would like to ask just what a semantics of pure difference would mean. Take, for example the word "sheep" in English: its French equivalent is "mouton"; but whereas we have a second word for the animal when slaughtered for consumption, namely "mutton," the French must manage with the one word for both concepts. Now, in a linguistic economy of pure difference, we are to understand that whereas we anglophones, thanks to our language, can carve up the world conceptually in such a way as to distinguish live from dead sheep, the francophone, on the other hand can make no such distinction. I would suggest though, that it is far from immaterial to most Frenchmen whether the dining table supports a live or a dead sheep, just as French farmers seem aware of the distinction when they set alight British lorries carrying live sheep. Such are the absurdities of a theory which treats language as a completely autonomous system. But this unworkable semantics is what Derrida insists on employing as his starting point. From there, he generates his favourite paradox by showing that meaning is perpetually deffered by the economy of difference - there is no self-present meaning when a sign owes its identity to the rest of the system.
 I would like to consider now the textualist strategies of interdisciplinarity which emerge from this position, and which pervade much recent work in the humanities. The Saussurean exclusion of any language-independent reality removes the referential function from language, and without reference, truth claims must also be abandoned. This blurs all distinctions between literary writing, which makes no direct assertions about the world beyond the text, and any writing which purports to extend our knowledge of extra-textual matters, whether particle physics or music theory. Distinctions between verbal and non-verbal artifacts can also be removed, now that language has been dispossessed of its referential functions. Thus the inability of music to form propositions is irrelevant: it is no more, and no less discursive, on this view, than language. All human artifacts can therefore be treated within a generalised theory of "text" without respect to the different media involved. Interdisciplinarity is the attempt to realise this project: since all is text, the boundaries between the disciplines is illusory. The different arts are thus to be treated in the same manner, and the differences between the arts and factual discourses are considered illusory.
 This has clear consequences for music theory, since the dissolving of
disciplinary boundaries in favour of generalised text removes all
justifications for specialist technical knowledge in this field or any other.
There is no ground for complaint therefore when the interdisciplinary
outsider encroaches upon the territory we thought was ours, without any
training in music theory, perhaps without even the ability to read music
notation, for no technical prerequisites can be stipulated if there are no
disciplinary boundaries left. Instead of music theory's approach to music
as an art with its own vast repertoire of techniques, based on its peculiar
material, sound, the interdisciplinary approach reduces musical works to
cultural documents. As the psychologist Maslov once said, if your only
tool is a hammer, you tend to see everything as a nail. Music theorists
should therefore think again before they join in the celebration of
interdisciplinarity, since the price of the ticket is ultimately the
renunciation of technique. While these techniques are individually open
to interrogation - witness the growing literature tracing the genaeology
of Schenker's thought in the work of William Pastille
 This, however, is to look towards the future. Inconsistency can sometimes have its advantages, and the consequences of misguided theories are not always realised. It is also to pass well beyond Derrida into the murkier realms of contemporary academia. What I propose to look at now are some attempts by musicologists to deconstruct musical texts using similar strategies to those found in Derrida. These strategies include the close reading of musical texts in order to find marginal elements which subvert the hierarchies supposedly present in the score. This accordingly involves the isolation of weighted binary oppositions, with a privileged and a suppressed term, which can be mapped onto binary oppositions of political import, tracing encoded patterns of complicity and subversion in the music. Related to this is the strategy of discovering a transgressive element in the score which cannot be contained within the frame set by canonical readings of the work, but disrupts and places in question the opposition between the inner and the outer which sustains the ideology of the autonomous musical work. As with Derrida, I shall be examing these transgressive close readings to see whether they have textual warrant.
 At any rate, McClary chooses the Brandenburg Concerto movement to demonstrate how her politics of tonality is played out in a piece of music from this era. Her reading is of course transgressive of canonical readings, which fall within the norms of formalism - hence the "blasphemy of talking politics," as she puts it. She characterises the concerto as a form which "addresses the tensions between the dynamic individual and the stable society," the ritornelli representing stability and the concertino representing dynamism. In the case of this movement, the ritornello ill fits this description, due to its elusive phrase-structure and its strange melodic contours (in measures 3,4), but to point this out would burden the account McClary wants to give with complications. The transgressive element she picks out is the role of the harpsichord, which almost succeeds in enacting a revolution, emerging from its traditional humble task to overthrow not only the ripieno, but the two conventional solo instruments as well. She focusses upon the famous cadenza, in which the harpsichord "unleashes elements of chaos, irrationality, and noise until finally it blurs almost entirely the sense of key, meter, and form upon which eighteenth-century style depends." The harpsichord only allows itself to be reabsorbed into the final ritornello "when the alternative seems to be madness." This analysis faces some grave problems. Firstly, the harpsichordist was traditionally in a position of leadership; the humble role McClary would assign it must leave this out of consideration. Secondly, the harpsichord behaves as a normal member of the concertino group from the first entry, so that McClary must resort to supressing the harpsichord's contribution to stability for most of the movement, just as she suppressed the instability of the ritornello. Thirdly, and this is most important, the account McClary gives of the cadenza badly misrepresents the score: the harpsichord emerges to the fore gradually, not violently, and of the sixty-five measures, the first forty-one are more placid and undisturbed than any earlier passage. This section ends with six bars of dominant pedal, from which emerges a resumption of the demisemiquaver arpeggios and scales from earlier in the movement. The demisemiquavers proceed for six-and-a-half measures before the dominant pedal returns. Of these measures, only two and a half contain any chromaticism, and they take the form of a chromatic descent in diminished-7th chords, such as we find elsewhere in Bach (the Bb Partita, for example). Even during this short passage, the rhythm is very clear: the left hand takes the first note of every quaver beat, while the right hand follows with three demisemiquavers. Twelve measures of unambiguous dominant pedal return after this, and the harpsichord cadenza ends with as much decorum as it had begun.
 Another essay of McClary's analyses the first movement of Brahms's
Third Symphony, a paradigm of Hanslickian absolute music according to
the canonical interpretations which McClary wishes to challenge
 Finally, let us look at an analysis by James Hepokoski of a scene
from Il trovatore: the gypsies' "Anvil Chorus" which alternates
with Azucena's narrative, "Stride la vampa"
 The pivotal use made of Azucena's strophic aria in this system of
oppositions appears over-ambitious, since a strophic form was used
conventionally for narratives, regardless of any French, bourgeois or
other associations this form supposedly carried. Hepokoski's
interpretation finally comes unstuck on general issue of the fixed forms
thesis that he depends upon in order to extract his initial binary
opposition: this thesis altogether lacks support in the contemporary
literature. The one commentator who mentioned forms - Basevi - used
the word to refer to the constituent parts of the larger forms, spanning a
scena, which Hepokoski means by the word. Basevi, as Roger Parker has
 These three examples are not offered as paradigms of Derridean deconstruction, but as representative examples of the use of deconstructive strategies in recent musicology. But even if McClary and Hepokoski were meticulous in following Derrida's practices, they would still face shipwreck upon the issue of Saussurean semantics, which Derrida shares with the cruder linguistic idealism of so many in the humanities today. As I have argued, the search for binary oppositions within texts is without motivation once Saussure's semantics is understood to be unworkable. And since Saussure was attempting to found a scientific project, he submitted himself, by implication, to scientific method, which is fallibilist rather than dogmatic; to respect Saussure's historical role is very different, therefore, from nurturing misplaced dogmatic loyalty in a theory which was disproven long ago. I have many other examples of putative musicological deconstructions which fail to live up to their claims, but since the danger of causing offence by attacking someone in the audience increases with the number of examples used, I thought it would be more diplomatic to reserve these for any private discussion which might arise later. In conclusion though, if the god Theuth, of Plato's myth, should approach you bearing the gift of deconstruction, ponder his claims carefully, lest you receive poison rather than cure. Musicology today needs a cure for its overweening theoretical ills, but that cure is unlikely to be found in the incantations and potions of deconstruction's magicians (ph�rmakos). If, on the contrary, you think I have made a scapegoat (pharmak�s) of deconstruction, you are welcome to place your protest on mto-talk.
2. Jacques Derrida, "Plato's
Pharmacy,"Dissemination (1972), trans. Barbara Johnson
Return to text
3. Christopher Norris, Derrida
(London, 1987), 28.
Return to text
4. Plato, Phaedrus, 274ff.
Return to text
5. Plato, Phaedrus, trans. W. Hamilton
(London, 1987), 96.
Return to text
6. This etymological confusion must surely have
struck many readers. The first reference to the matter in print is, I
believe, Alexander Nehemas, review of Norris, ibid.,The
Philosophical Review, Vol. 100/2 (1991), 303-6.
Return to text
7. cf. Nehemas, ibid.,
Return to text
8. Plato, trans. Hamilton, op. cit.,
Return to text
9. Saussure, Cours de linguistique
g�n�rale (Paris, 1916).
Return to text
10. Derrida, Grammatology (1967),
trans. G. Spivak (Baltimore, 1974), 27-73.
Return to text
11. For an excellent summary of this issue, see
Leonard Jackson, The Poverty of Structuralism (New York,
Return to text
12. Jackson also takes De Man to task for his use of
the metonomy/metaphor opposition, which is dependent upon the same
syntagm/paradigm opposition long since rejected by linguistics,
Return to text
13. Jean-Jacques Nattiez, Fondements d'une
semiologie de la musique (Paris, 1975).
Return to text
14. Roland Barthes, Syst�me de la
mode (Paris, 1967).
Return to text
15. Roland Barthes, S/Z (Paris,
Return to text
16. Jean-Jacques Nattiez, Music and
Discourse, trans. C. Abbate (Princeton, 1990).
Return to text
17. For example in Tony Bennett, Formalism
and Marxism (London, 1979), 4-5; and in Terry Eagleton,
Literary Theory (Oxford, 1983), 96,7 - at least Eagleton is
correct in his account of the signifier/signified distinction. Similar
confusions emerge in accounts of Saussure to be found in Catherine
Belsey, Critical Practice (London, 1980); Terence Hawkes,
Structuralism and Semiotics; Julia Kristeva does not escape these
confusions, as the pages of The Kristeva Reader (Oxford, 1986),
ed. Toril Moi, readily witness. Kristeva is at least consistent, in making
explicit her unconcern for, and even rejection of linguistics qua
science, in "The Ethics of Linguistics," Desire in Language
Return to text
18. Most recently in "Schenker's Value Judgments,"
MTO Vol. 1.6 (1995).
Return to text
19. Alan Street, "Superior Myths, Dogmatic
Allegories," Music Analysis Vol. 8 (1989).
Return to text
20. Matthew Brown and Douglas Dempster, "The Scientific
Image of Music Theory," Journal of Music Theory Vol. 33 (1989),
Return to text
21. Susan McClary, "The Blasphemy of Talking Politics
During Bach Year," Music and Society: The Politics of Composition,
Performance, and Reception,ed. Richard Leppert and Susan McClary
(New York: Cambridge University Press, 1987).
Return to text
22. Susan McClary, "Narrative Agendas in 'Absolute'
Music: Identity and Difference in Brahms's Third Symphony," Musicology and
Difference, ed. R. Solie (Berkeley: University of California
Press, 1993), 326-344.
Return to text
23. James Hepokoski, "Il trovatore as
Sociocultural Dramma: Generic Mixtures and Clashing
World-Views," unpublished conference paper, Belfast International
Congress of Verdi Studies (1993).
Return to text
24. Roger Parker, "Insolite forme,"
unpublished conference paper, Belfast International Congress of
Verdi Studies (1993).
Return to text
Return to Beginning
 Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:
This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.
 Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.
This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.