Foucault on the "Question of the Author": A Critical Exegesis Author(s): Adrian Wilson Source: The Modern Language Review,

Vol. 99, No. 2 (Apr., 2004), pp. 339-363 Published by: Modern Humanities Research Association Stable URL: http://www.jstor.org/stable/3738750 Accessed: 24/10/2010 17:37
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. 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/action/showPublisher?publisherCode=mhra. 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. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Modern Humanities Research Association is collaborating with JSTOR to digitize, preserve and extend access to The Modern Language Review.

http://www.jstor.org

FOUCAULT THE AUTHOR':

ON A

THE

'QUESTION

OF

CRITICAL

EXEGESIS

It might be said that the author-figure, whose death was announced in the late 1960s, came back to life in the 1990s, when there emerged a renewed debate in literary theory over the problem of authorship;1 and this prompted a reappraisal of those now classic essays in which Roland Barthes and Michel Foucault originally proclaimed?or seemed to proclaim?the author's demise.2 The present paper continues this reassessment Foucault's chief by examining to the author-figure's contribution 1969 funerary rites: his lecture of February entitled 'Qu'est-ce qu'un auteur?'.3 That lecture, translated into English in 1977 as 'What is an Author?', entered the canon of discussions of authorship and in at least three times.4 Yet has been selectively reprinted English throughout the 1980s it never received the close critical attention which it deserved, and on which its classic status should surely have entailed. Instead, commentators all sides variously endorsed and criticized what were taken to be Foucault's his argument. This curious conceptual claims, without actually scrutinizing of Sean Burke's elegant and silence was broken in 1992, with the publication The Death and Return of the Author?a wide-ranging compelling reappraisal of the anti-authorial works of Barthes, Foucault, and Derrida. By focusing on the Burke has revealed a remarkable rhetoric of these theoretical writings, range and troubles in their arguments. of both strengths Not the least of Burke's the question of Foucault's achievements has been to rephrase 'What is an Author?' itself, asking instead: 'What (and who) is an author?'5 As we shall see in due course, this reformulation proves to be apt indeed; and I hope to show that a strategy akin to Burke's yields still further fruits when applied anew to 'What is an Author?' For help with this paper I thank Mike Beaney, Sean Burke, John Christie, John Forrester, Marina Frasca Spada, Nick Jardine, Mark Jenner, Sharon Macdonald, and Roger White. Errors are my own responsibility. 1 The same applies to the accompanying figure of the literary work: see Margit Sutrop, 'The Death of the Literary Work', Philosophy and Literature, 18 (1994), 38-49. 2 Peter Lamarque, 'The Death of the Author: An Analytical Autopsy', British Journal of Aes? thetics, 30 (1990), 319-31; Sean Burke, The Death and Return of the Author: Criticism and Sub? jectivity in Barthes, Foucault and Derrida (Edinburgh: Edinburgh University Press, 1992; repr. 1998); Sutrop, 'The Death of the Literary Work'; Donald Keefer, 'Reports of the Death of the Author', Philosophy and Literature, 19 (1995), 78-84. See also the essays in What is an Author?, ed. by Maurice Biriotti and Nicola Miller (Manchester: Manchester University Press, 1993), and the papers reprinted in Authorship: From Plato to the Postmodern, ed. by Sean Burke (Edinburgh: Edinburgh University Press, 1995). 3 Page references in the text below are to Michel Foucault, 'What is an Author?', trans. by Donald F. Bouchard and Sherry Simon, in Language, Counter-Memory, Practice, ed. by Donald F. Bouchard (Oxford: Blackwell, 1977), pp. 113-38. For the relationship between this text and the original 1969 lecture see the editor's note, p. 113, and Burke, Death and Return, pp. 89-94. There were two French versions: see Dits et ecrits ig$4-ig88 par Michel Foucault, ed. by Daniel Defert and Francois Ewald, 4 vols (Paris: Gallimard, 1994), 1, 789-812. 4 Textual Strategies: Perspectives in Post-Structuralist Criticism, ed. by Josue V. Harari (London: Methuen, 1980), pp. 141-60; Theories of Authorship, ed. by John Caughie (London: Routledge, 1981; repr. 1986, 1988), pp. 282-91; Authorship, ed. by Burke, pp. 233-46. 5 Burke, Death and Return, pp. 78-89 (title ofthe section).

340 i. Foucault's

Foucault What

on the

'Question

of the Author'

is an Author?'

and its Contexts

The chief context of Foucault's 'Qu'est-ce qu'un auteur?' was Barthes's essay in 1968?a 'La mort de l'auteur', written in 1967 and published typically pithy in the words of its title, 'the death ofthe author'. Here, piece which announced, Barthes began by asking: after quoting a sentence from Balzac's Sarrasine, Who is speaking thus? Is it the hero of the story . . .? Is it the individual Balzac . . .? Is it Balzac the author . . .? Is it universal wisdom? Romantic psychology? We shall never know, for the good reason that writing [ecriture] is the destruction of every voice, of every origin. Writing is that neutral, that composite, that oblique space where our subject slips away, the [photographic] negative where every identity is lost, starting with the identity of the very body which writes.6 As this striking introduction made clear, the point of Barthes's argument was to replace the figure of the author (or rather, 'the Author', capitalized) with the and the necessity of this transfor? figure of ecriture.7 To evoke the possibility a little history of writing and authorship, a history mation, Barthes developed which fell into three phases: primordial fall, future redemp? grace, subsequent tion. In the original state of grace?preserved to this day in 'ethnographic had known itself for what it was; subsequently, societies'?writing writing was corrupted by the gradual birth of modern society, which installed the tyrannical a moment of redempfigure of the Author; finally, there was now supervening tive return, i.e. the 'destruction of the author' or 'death of the author', which would at long last restore writing to itself. But how could writing redeem itself from its authorial deformation? In the course ofthe modern age, Barthes explained, certain writers?first and foremost then after him Valery, Proust, and the Surrealists?had Mallarme, struggled to bring about this very emancipation; yet their valiant efforts had proved to be no more than a series of heroic failures. These attempts, then, amounted in the end to an unwitting collective to 'the sway of the Author'. Yet testimony which was making it possible, for the help was now at hand from linguistics, first time, to strip away the illusions of authorship. For linguistics had recently revealed the truth of language itself, namely: that the whole ofthe enunciation is an empty process, functioning perfectly without there being any need for it to be filled with the person ofthe interlocutors. Linguistically, the author is never more than the instance writing, just as / is nothing other than the instance saying /: language knows a 'subject', not a 'person', and this subject, empty outside of the very enunciation which defines it, sufhces to make language 'hold together', sufhces, that is to say, to exhaust it.8 To assimilate the lesson supplied was to dethrone the Author. by linguistics No longer would writing emanate, or be taken to emanate, from some parental 6 Roland Barthes, 'La mort de l'auteur' (1968), repr. in his Le Bruissement de la langue (Paris: Seuil, 1984), pp. 61-67 (p. 61); English translation adapted from that of Stephen Heath, 'The Death of the Author', in Barthes, Image Music Text, essays selected and translated by Stephen Heath (London: Fontana/Collins, 1977), pp. 142-48 (p. 142). 7 Hence the radical distinction between ecriture (writing as detached from 'all agency, all ac? tivity') and ecrire (writing as the action of an agent): see Anne Banfield, 'Ecriture, Narration and the Grammar of French', in Narrative from Malory to Motion Pictures, ed. by Jeremy Hawthorn (London: Arnold, 1985), pp. 1-22 (p. 13). 8 Barthes, 'La mort de l'auteur', pp. 63-64.

ADRIAN

WILSON

341

figure anterior to itself, i.e. from the Author or from the Author's 'hypostases'? society, history, psyche, liberty; instead, writing could now at last be repositioned back where it belonged, that is to say, inside language. This apocalyptic redemption of writing would entail killing not only the Author but also the Critic; the collusive pair Author-Critic would now be replaced by the new couplet of 'the and the sovereign modern scriptor' reader. The 'modern would be scriptor' a writer who is not an Author, whose being does not precede writing but on and delimited the contrary is constituted by writing itself.9 Correspondingly, this point, the Author's Barthes did not foreground although product was a whereas the 'modern was a associated not with book but with 'book', scriptor' a 'text'. But the fundamentally was to be the reader, who was figure redemptive already the true and only source ofthe otherwise mythical unity ofthe text, and whose constitutive role in the making of ecriture would now be revealed and 'The birth of the reader', Barthes concluded, 'must be at the cost of accepted. the death of the Author.' Such were the main lines of Barthes's 'La mort de l'auteur'. Without doubt that essay was one of the stimuli for Foucault's lecture 'Qu'est-ce qu'un auin the Foucault teur?', delivered early year (1969); yet following delicately avoided mentioning Barthes by name. Instead he framed his discussion as a which had been levelled at his own Les Mots et response to certain criticisms les choses?criticisms were partly justified.10 In that book, which, he admitted, in 1966, he had bypassed 'the question ofthe author'; concerned as published he was with 'discursive rather than with 'the familiar of a layers' categories a or an he had the names of authors book, work, author', carelessly 'employed had [. . .] in a naive and often crude fashion' (pp. 113, 115). This, he explained, of his enterprise. The nature of opened the way to various misunderstandings that enterprise would shortly be clarified by UArcheologie du savoir, which was in press. 'Nevertheless,' at that moment he went on, 'as 'a privileged moment of individualization in the history of ideas, knowledge, and literature, or in the history of philosophy and science, the question of the author demands a more direct response' And auteur?' that 'Qu'est-ce (p. 115).11 qu'un comprised response. Foucault to examine the author 'as a function of discourse', reproposed placing the conventional figure of 'the author' with what he called 'the authorfunction'?a role played by that concept which sought to capture the discursive One his that it is precisely the figure. might paraphrase argument by saying author-function which authorizes the very idea of 'an author'. Foucault devel? oped this novel conception chiefly with reference to the seemingly simple case of 'a book or a series of texts that bear a definite signature' (pp. 131, 136). Even at this 'level' the phenomenon of authorship hands an acquired in Foucault's 9 Notice incidentally the shifting valuation attached to the figure of the 'modern'. Modernity was initially associated with the falsehood of the author ('the Author is a modern figure', ete); but now, through the liberation supplied in recent times by linguistics, modernity is aligned with the recovery of writing's truth. 10 Michel Foucault, Les Mots et les choses: une archeologie des sciences humaines (Paris: Gallimard, 1966); trans. by A. Sheridan as The Order of Things (New York: Random House, 1970). 11 Cf. Michel Foucault, L Archeologie du savoir (Paris: Gallimard, 1969); trans. by A. Sheridan as The Archaeology of Knowledge (London: Tavistock, 1972).

342

Foucault

on the

'Question

of the Author'

but he went on to show that still further considerations unexpected complexity; applied in the cases of Marx and Freud, who were not just authors of works but 'initiators of discursive practices'. Indeed, it emerged that the concept of the 'author-function' would require some further elaboration to embrace such '"fundamental" for Foucault that 'the enigmatic link be? authors', explained tween an author and his works'?the a premiss of the author-function?took distinctive form with respect to psychoanalysis and Marxism.12 Nevertheless, he indicated that the 'author-function' concept applied not only to the author of 'an ordinary text' but also to 'initiators of discursive practices' such as Marx and Freud. Ultimately, the authority of even Marx and Freud was therefore, derived from the author-function?so Foucault was suggesting, even though he abstained from demonstrating this concretely.13 Thus, in harmony with the of Les Mots et les choses and of the forthcoming L Archeologie du arguments the of authors concealed the real source of author? savoir, apparent sovereignty itself. Correspondingly Foucault too, albeit in a different ity, namely discourse culture. To this way from Barthes, was seeking to herald a new, post-authorial end he opened and closed his discussion with a quotation from Beckett: 'What matter who's speaking?' No longer should we bend our ear to the supposedly voice of the named, individual author; instead, we should attend to personal the anonymous ofthe collective discours (pp. 115-16, murmuring 138). Hence the transmutation Foucault's title. The of the author performed by very figure was turned from a 'who' into a 'what'?though strangely enough, the rhetorical which presaged a future of glorious anonymity came from a named question author, Beckett. On the face of it, the argument of Foucault's 'Qu'est-ce qu'un auteur?' seems closely akin to that of Barthes's 'La mort de l'auteur'.14 Certainly the two pieces shared several paradoxical gestures: the fact that the author's death was itself the authorial of Barthes and of Fou? an author ed event, requiring signatures of certain chosen authors such as Mallarme and cault; the selective privileging from the death sentence; the seeming Beckett, who were apparently exempted as to whether Barthes and Foucault were signing a death warrant, ambiguity or preaching at a funeral. Yet these resemblances carrying out an assassination, are misleading, for as we shall soon begin to see, Foucault took considerable to distance himself from Barthes?not least both the tradi? by criticizing pains tional concept ofthe literary 'work' (which Barthes had effectively left intact), and the new concept of ecriture (which Barthes had installed in place of 'the of its own, in at least Author').15 And in fact Foucault's essay had a significance three respects. 12 With respect to the use of the word 'works' in the phrase 'the enigmatic link between an author and his works', cf. below, pp. 354?55. 13 See further below, p. 348. 14 This is how these two essays are depicted in Banfield's otherwise illuminating 'Ecriture, Narration and the Grammar of French',passim, esp. pp. 14, 19?21. The arguments of the two are distinguished, though only in passing, by Lamarque, 'The Death ofthe Author', p. 319, and by Donald E. Pease, Author', in Authorship, ed. by Burke, pp. 263-76 (p. 272). A complementary case is that of Burke, Death and Return, who treats the two separately (pp. 22-27, 78-89), at the cost of suppressing the dialogue between them (this in line with his general strategy, defended p. 178 n. 30). 15 See Section 2 below (pp. 344-48).

ADRIAN

WILSON

343

In the first place, whereas Barthes had sought to criticize and to supersede the author-figure, Foucault worked instead to problematize that figure, i.e. to make 'the author' the site of an enquiry. And there is reason to believe that he theories of thus exerted a significant influence upon literary and philosophical at least in the Anglophone world. Already, in the early 1960s, Wayne authorship, Booth had introduced the concept of the 'implied author', but the latter figure In contrast Foucault was conceived as an authorial construction. posited the of the as a construct the and reader; interpretative space which he author-figure conceptions thereby opened has since been peopled by a series of constructivist Alexander Nehamas's of the author?first author', concept of the 'postulated Currie's theory ofthe 'fictional author', and latterly Jorge Grathen Gregory These conceptions of authorship, author'. cia's figure of the 'interpretative in Gracia's formulation, which have attained a new level of sophistication only Foucault was extendbecame thinkable thanks to Foucault's essay.16 Secondly, literature to the domain of non-fictional from imaginative ing the problem he implied in his opening remarks, where (as we have seen) he de? writing?as moment of individualization fined 'the question of the author' as 'a privileged or in the history of philo? and literature, in the history of ideas, knowledge, This move was rather less explicit than the first: indeed, sophy and science'. between such domains, for the most part Foucault oddly elided the distinction the between Homer and Galileo.17 from the arts to sciences, gliding efTortlessly ofthe author question was also potentially fecund? his extension Nevertheless, on the sciences have only recently commentators begun to take up although 'What is an Author?' which Foucault thus created.18 Thirdly, the opportunity the new figure of 'the text" which was al? part in constituting played a significant in the 1970s and at the time, and which was to gain ascendancy ready emerging his that Not did Foucault lecture; more figure throughout 1980s. deploy only work' and of ecriture the of 'the his of helped critique concepts particularly, the figure of 'the to propel Barthes himself into taking up more systematically as we have seen, Barthes had counterposed text'. In 'La mort de l'auteur', to the 'work'), but only in passing and the 'text' against the 'book' (equivalent the contrast between these.19 But in 1971 he devoted a new without thematizing in doing so piece, 'De l'ceuvre au texte', to just this distinction?and polemical which Foucault had raised in 1969.20 Barthes took on board the very criticisms of 'the Text' should displace the traditional now proposed that the category 16 Booth's concept was used allusively by Foucault ('What is an Author?', p. 129). See Wayne C. Booth, The Rhetoric of Fiction (Chicago: University of Chicago Press, 1961); Alexander Nehamas, 'The Postulated Author: Critical Monism as a Regulative Ideal', Critical Inquiry, 8 (1981), 13149 (p. 136); id., 'What an Author is', Journal of Philosophy, 83 (1986), 685-91; id., 'Writer, Text, Work, Author', in Literature and the Question of Philosophy, ed. by Anthony J. Cascardi (Baltimore and London: Johns Hopkins University Press, 1987), pp. 265-91; Gregory Currie, The Nature of Fiction (Cambridge: Cambridge University Press, 1990), pp. 79-83, 117 n.; Jorge J. E. Gracia, 'Texts and their Interpretation', Review of Metaphysics, 43 (1990), 495-542; id., 'Can There be Texts without Historical Authors?' American Philosophical Quarterly, 81 (1994), 245-53. Cf. further n. 79 below. 17 Cf. p. 341 above. For an exception see p. 349 below. 18 Steve Woolgar, 'What is a Scientific Author?', in What is an Author?, ed. by Biriotti and Miller, pp. 175-90. 19 See above, p. 341. 20 Roland Barthes, 'De l'ceuvre au texte' (1971), repr. in Le Bruissement de la langue, pp. 69-

344

Foucault

on the

'Question

of the Author'

with the of 'the work'; and this new figure of 'the Text', concept dignified capital letter, effectively replaced the figure of ecriture which he had deployed in his earlier essay. It appears, then, that Foucault's 'Qu'est-ce qu'un auteur?' was the oblique link between those two Barthesian 'La mort de 1'auteur' and 'De classics, Foeuvre au texte'. Indeed, Barthes had every reason for revising his claims in the light of Foucault's in the prefatory section of that lecture, lecture?for after referring as we have seen to his own Les Mots et les choses, Foucault had demolished the argument of Barthes's 'La mort de Fauteur'. This initial phase of Foucault's discussion merits attention not only because it opened the space for his own argument, but also as a remarkable in its rhetorical achievement own right.

2. Counter-History It will be recalled that Barthes had approached the author problem by sketching a history of authorship. in contrast, Foucault, began by making it clear that while he was well equipped to produce a history of his own, he would not here be undertaking that task: For the purposes of this paper, I will set aside a sociohistorical analysis ofthe author as an individual and the numerous questions that deserve attention in this context: how in a culture such as ours; the status we have given the the author was individualized author, for instance, when we began our research into authenticity and attribution; the systems of valorization in which he was included; or the moment when the stories of heroes gave way to an author's biography; the conditions that fostered the formulation of the fundamental critical category of 'the man and his work'. For the time being, I wish to restrict myself to the singular relationship that holds between an author and a text, the manner in which a text apparently points to this figure who is outside and precedesit. (p. 115) As Foucault's discussion it seemed as if he was indeed prefatory proceeded, the historical tactic which Barthes had eschewing adopted. Yet in fact his intro? remarks were devoted to ductory precisely rebutting the history which Barthes had put forward; and within their seemingly non-historical form Foucault what we may call a counter-history, i.e. a radical rework? subtly constructed ing of the story Barthes had told.21 In Barthes's story, writers such as Mallarme had failed to dethrone the usurpit required the assistance of linguistics, ing figure ofthe Author; and accordingly and of course the courage of Barthes himself, to redeem writing from its tragic fall. But in sharp contrast, Foucault argued that literature itself had already of the author', i.e. 'the total brought about what he called the 'disappearance

77; English translation by Stephen Heath, 'From Work to Text', in Barthes, Image Music Text, PP- I55_?4- F?r a critical discussion of this essay see Sutrop, 'The Death of the Literary Work'. 21 The following observations have been prompted by the remarks of Donald E. Pease ('Author', p. 272), who has rightly drawn attention to the disparity between Foucault's argument and that of Barthes. On the present reading, however, it would appear that Pease's account is inaccurate in taking Foucault's use of the present tense in his title as an index of this disparity.

ADRIAN

WILSON of the writer'.22 Indeed,

345 this was

effacement of the individual characteristics the point of his opening allusion to Beckett:

Beckett supplies a direction: 'What matter who's speaking, someone said, what matter who's speaking'. In an indifference such as this we must recognize one of the funda? mental ethical principles of contemporary writing. (pp. 115-16) this theme, Foucault turned to what he called 'the kinship be? Developing tween writing and death'?a 'inverts the age-old kinship which, he explained, of Greek narrative or epic, which was designed to guarantee the conception of a hero' (pp. 116-17). This original, protective function of nar? immortality rative was not confined to the Greeks, for in a similar way 'Arab stories, and The Arabian Nights in particular, had as their motivation [. . .] this strategy death' (p. 117). In both Greek and Arab culture, then, narrative for defeating had begun as 'a protection against death'; but in 'our culture' this relationship its own author. To illustrate this has been inverted, for writing now annihilates claim, Foucault used the triad Flaubert-Proust-Kafka: Writing is now linked to sacrifice and to the sacrifice of life itself; it is a voluntary obliteration of the self that does not require representation in books because it takes place in the everyday existence of the writer. Where a work had the duty of creating immortality, it now attains the right to kill, to become the murderer of its author. Flaubert, Proust, and Kafka are obvious examples of this reversal. (p. 117) This picture has not only pressed further Foucault's but has counter-history, Barthes's use of the figure of authorial death; for by assigning also outflanked has deprived of that figure to literature Barthes's itself, Foucault argument The 'murderer' its putative originality. of the author is not Barthes but 'our instanced The extinction culture', by the writing of Flaubert-Proust-Kafka. of the author, then, far from being an event of the future which requires the aid in Barthes's of linguistics little history), has already been (as it was depicted achieved by the hand of literature itself. Foucault has constructed a very different temporality from Correspondingly, in doing so, he has created some glaring gaps. that deployed by Barthes?though On the one hand, the figure ofthe present has been radically redefined: the conjoined figures of Beckett, Flaubert, Proust, Kafka have together defined a single which Foucault calls 'our culture' (p. 117), or 'the writing of cultural moment, our day' (p. 116). To link Beckett with Flaubert in this way is precisely to negate Barthes's to the progress from Mallarme picture of an imperfect, incomplete present; for Flaubert was Mallarme's near-contemporary, just as Beckett is the of Barthes and of Foucault. Similarly Proust has been reassigned: contemporary him as one of those who had striven without success to Barthes had positioned achieve the 'death of the author', but Foucault includes him within the roll of Yet on authors who have actually brought about the author's 'disappearance'. with a double uncertainty. the other hand, this redefined present is attended In the first place, one individual is curiously absent from Foucault's picture, will require him himself. Surely Foucault's counter-history namely Mallarme for Mallarme was a crucial figure in Barthes's history, to reposition Mallarme, 22 Foucault, 'What is an Author?', pp. 117 ('effacement', 'death or disappearance'), 119, 120, 121 ('disappearance'); cf. n. 26 below.

346

Foucault

on the

'Question

of the Author'

which would that movement serving as he had to initiate, however imperfectly, with Barthes's culminate own argument; has so yet on this matter Foucault far been silent, for Mallarme's name is absent from his pantheon. Secondly, his counter-history is signally incomplete, carrying a profound void at its very heart. For Foucault has opened up a massive gap between the Graeco-Arabie in which writing warded off death or its implications, and 'our cul? moment, across that gap he has posited an ture', in which writing is itself annihilation; inversion; yet he has offered no hint as to how or when this inversion took place. In short, where Barthes had offered a narrative, Foucault has merely posited a contrast: his counter-history, of motion within structural having no principle the origin of 'the writing of our day'. it, has left unexplained Foucault now proceeded Leaving these problems implicit and in suspense, to draw the practical moral which flowed from his counter-history. Since imag? inative literature had already accomplished the 'disappearance or death of the it followed that the 'task of criticism' was not to bring about this author', event?as Barthes had of course been claiming23?but, on the contrary, to catch that is, to 'explore' the 'consequences' up with what literature had achieved: of the author's disappearance, to 'appreciate' the 'importance of this event', of it (pp. 117, 119). However, to 'take full measure' Foucault went on, this of was obstructed certain idioms necessary enterprise being by contemporary criticism. One such idiom was the traditional category of 'the work', and more the paradoxical retention of that category in recent structuralist cri? particularly ticism: 'if some have found it convenient to bypass the individuality ofthe writer or his status as an author to concentrate on a work, they have failed to appreciate the equally problematic nature of the word "work" and the unity it designates' 'thesis' was the much more recent 'notion of (p. 119).24 Another unhelpful 'as currently employed', had 'merely transposed the ecriture'] for this concept, of an author to a transcendental In a characteristics anonymity'.25 empirical nicely ironic and reflexive touch, which was surely not lost on his auditors at refrained from naming the the College de France in February 1969, Foucault associated both with structuralist author who was pre-eminently contemporary of ecriture': that is, Roland Barthes. criticism and with the 'conception of his prefatory But Foucault's master stroke came in the final paragraph discussion: This conception of ecriture sustains the privileges of the author through the safeguard ofthe a priori; the play of representations that formed a particular image ofthe author is extended within a grey neutrality. The disappearance of the author?since Mallarme, an event of our time?is held in check by the transcendental.26 Is it not necessary to draw a line between those who believe that we can continue to situate our present discontinuities 23 However, Barthes's stance entailed that he himself could no longer be a critic, since as we have seen he depicted 'the Critic' as collusive partner of 'the Author'. This ambiguity, like that associated with modernity, was left unresolved in 'La mort de l'auteur'. Cf. p. 341 above. 24 On this point see Section 4 below (pp. 354-56). 25 Banfield seems to read this passage as directed not against Barthes but instead against sub? sequent Anglophone misunderstanding of Barthes's conception: 'Ecriture, Narration and the Grammar of French', pp. 19-21. 26 A point of detail is that Foucault used the phrase 'disappearance of the author', in place of Barthes's 'death of the author'?though initially he combined the two, writing of 'death or disappearance'. See n. 22 above.

ADRIAN

WILSON

347

within the historical and transcendental tradition of the nineteenth century and those who are making a great effort to liberate themselves, once and for all, from this conceptual framework? (pp. 119-20, emphasis added) whom Foucault had been holding back Thus the pivotal figure of Mallarme, until this moment, has now been played as the trump card: it was Mallarme who had brought about the 'disappearance of the author' in the first place. And the other this completes and confirms Foucault's counter-history, resolving for the 'inversion' ofthe relationship issue which he had left in silent suspense: a historical location and a between writing and death has now been assigned himself.27 Admittedly, this counter-history cause, in the person of Mallarme narrative and the Graeco-Arabic has left a potentially troubling gap between of Western as it does most of the trajectory Mallarmean inversion, eliding from view its this has been concealed but literature; by dispersal problem who of Mallarme, through the text. What matters is the radical re-evaluation Barthes within the present, within 'our time'. Conversely, has been positioned and transcendental himself has been consigned to the past, to 'the historical tradition ofthe nineteenth century'; his concept of ecriture, far from heralding it in a still the author problem by reinscribing a new dawn, has compounded on its head: has been turned form. In short, Barthes's more mystified history in the role of whom Barthes to east Mallarme, inadequate precursor, sought for the had in fact been far ahead of him in the first place. The final indignity his apocalyptic Barthes is that Foucault has preserved unfortunate tone, while not Barthes, from him the banner of the future. For it is Foucault, snatching once and for all,' from the who is 'making a great effort to liberate [himself], of 'the nineteenth framework' century'. 'conceptual to develop Barthes could now proceed thus Having swept away, Foucault of what he called 'the question of the author'. He began his own treatment the author's name, in order to set up his central thesis: by problematizing the existence, and that 'the function of an author is to characterize circulation, within a society' (p. 124). This claim served to of certain discourses operation 'features' of the heart of his lecture, i.e. a sketch of four characteristic introduce defined with reference to the case of 'books or texts with the 'author-function', of to a 'schematic' discussion Next (pp. 131-36) Foucault proceeded authors'. of raised 'the initiation discursive the more complex practices' problems by with the paternal form of authorship associated (p. 136), i.e. the distinctive passage (pp. 136? figures of Marx and Freud. Finally, in a brief concluding with a series of wider themes: the analysis linked his argument 38), Foucault of 'the privileges of the subject' (p. 137); and the the question of discourse; for of which he the future, an anonymity evoked discourse envisaged anonymity his from Beckett: 'What matter who's earlier speaking?' by recalling quotation (P. 138). I shall be concerned in particular with In examining Foucault's argument, his discussion, two themes which permeated yet which he never considered directly: the figure of the text and the individual identity of the author. It will be 27 Notice, in addition, that Foucault has inverted the form of Barthes's history: Barthes had proceeded in chronological order, but Foucault has moved from Beckett backwards to Mallarme.

348

Foucault

on the

'Question

of the Author'

since each of these convenient to proceed analytically rather than sequentially, lec? at several different sites within Foucault's topics appeared and reappeared ture. Nevertheless, the structure of his exposition will be taken into account, for as we shall see, that structure itself played a significant rhetorical role. One section of Foucault's lecture will be left aside here, his discussion of Marx and of discursive That passage has been omitted Freud, the 'initiators practices'. from consideration for two reasons. In the first place, it was at a tangent to Foucault's main argument?for as has already been mentioned,28 Foucault did not explain how the authority and originality which characterized Marx and Freud was to be assigned to the 'author-function'. this Secondly, part of Fou? cault's essay has already been treated in exemplary fashion by Burke, who has its fundamental brought out forcefully aporia: that is, the fact that to install Marx and Freud as 'initiators of discursive was to undermine the practices' of discourse itself. Thus this particular posited sovereignty passage in Fou? cault's 'What is an Author?' Burke's wider argument: strikingly exemplifies that 'the principle of the author most powerfully reasserts itself when it is that of 'the the author is never more alive than when absent', thought concept which follows, while complemen? thought dead'.29 And indeed the discussion which are entirely compatible with his tary to Burke's, will lead in directions conclusions.

3. The Author In his

and the Text

the limits of 'What is an remarks, when demarcating introductory Foucault tied 'author' to 'text'. As we have and Author?', strictly reciprocally with 'the singular relationship that holds seen, his essay would be concerned an author and a text, the manner in which a text apparently between points to this figure who is outside and precedes it' (p. 115). And yet after some fur? ther preliminary observations in Section 5), he (which I shall be considering broke this link, implicitly the meaning of 'text'. 'In our culture', he redefining observed at the end of his prefatory discussion, the name of an author is a variable that accompanies only certain texts to the exclusion of others: a private letter may have a signatory, but it does not have an author; a contract can have an underwriter, but not an author; and, similarly, an anonymous poster attached to a wall may have a writer, but he cannot be an author. (p. 124, emphasis added) then, a text no longer implied an author; By this stage of Foucault's exposition, rather, a text amounted simply to anything written or printed. Correspondthis was the point?the 'author-function' was associated not with ingly?and the text as such, but rather with some texts. The core of the essay?now to 'those books or texts with duly limited authors' of an exposition of four specific 'characteristics (p. 124)?consisted of the "author-function"' these four 'characteris(p. 130). After explaining 28 See above, p. 342. 29 Burke, Death and Return,

pp. 6, 7; cf. p. 362 below.

ADRIAN tics' in turn, Foucault summarized numbered and listed them):3?

WILSON them as follows (for convenience,

349 I have

is tied to the legal and institutional systems that circum[i] the 'author-function' scribe, determine and articulate the realm of discourse; [2] it does not operate in a uniform manner in all discourses, at all times, and in any given culture; [3] it is defined not by the spontaneous attribution ofa text to its creator, but through a series of precise and complex procedures; [4] it does not refer, purely and simply, to an actual individual in so far as it simulta? neously gives rise to a variety of egos and to a series of subjective positions that individuals of any class may come to occupy. (pp. 130-31) was Foucault's overview of the four 'characteristics of the author-func? I shall now examine the respective which these four points passages in order to bring out both Foucault's and the summarized, explicit argument shifting uses to which he put the figure of the 'text'. Of these four 'characteristics' the first two were not so much descriptive as to show that the author-function was concircumstantial, designed historically is connected tingent and mutable. The point of [1] was that the author-function with transgression, with punishment, and with property, and in particular with the iegal codification' of authorship which took place around 1800 (pp. 12425).3I Under [2] Foucault argued that the figure ofthe author was imposed at different historical texts' on the one hand and upon periods upon 'scientific of or fiction' on the other As this latter point im? 'text[s] poetry (pp. 125-27). the historical observations Foucault was making in [1] and [2] were not plied, with the succession of epistemes which he had depicted in Les Mots connected et les choses, nor for that matter with the literary counter-history he had offered earlier in his lecture. On the contrary, as he had made explicit at the outset,32 Foucault was not concerned to construct even so much as the sketch of a history. Rather, the two claims he was making here were (as he put it) 'transhistorical', were serving a merely illustrative and his historical allusions purpose.33 Fou? in the present age: the role cault's real concern was with the author-function was simply to establish first the legal associations of his historical of examples the author-function (i.e. [1]) and then its contingent quality (i.e. [2]). Far more then, were [3] and [4], for these depicted the concrete and practical important, working of the author-function. In developing the figure of the text, [1] and [2], Foucault was still deploying and using this to refer to (in the words of my own earlier gloss) anything written or printed. Yet by the time he came to offer the summary quoted above, the figure ofthe text had become curiously marginal, appearing only in [3] and even 30 Under [3] the published English translation begins 'it is not defined by'; I have replaced this with 'it is defined not by'. 31 Nesbit's otherwise incisive analysis is thus inaccurate in claiming that 'Foucault did not call in the law to supply answers to his essay question, "What is an Author?"; oddly, he did not even mention it.' See Molly Nesbit, 'What Was an Author?', Yale French Studies, 73 (1987), 229?57 (p. 240), partially repr. in Authorship, ed. by Burke, pp. 247-62 (p. 255). 32 Cf. p. 344 above, at the start of Section 2. 33 This is confirmed by the fact that Foucault inserted, in a qualifying parenthesis (pp. 126-27), some brief remarks (shrewd and interesting in themselves) about the varieties of both literary and scientific authorship in the present day. Such tion'.

350

Foucault

on the

'Question

of the Author'

there in an almost liminal role. How this had happened will become clear as we examine the way that he had elaborated points [3] and [4]. The key moment of Foucault's discussion was point [3]. He began with a negative claim: 'The third point concerning this "author-function" is that it is not formed spontaof a discourse to an individual' neously through the simple attribution (p. 127). It should be observed in passing that this negation?whose will significance in Foucault's soon emerge?was later summary, where (inter alia), rephrased was replaced with 'text'. The implied-yet-unarticulated 'discourse' equivalence between discourse and text raises troubles of its own,34 but as we shall see, this was but one of the difficulties Foucault's use of the figure of the surrounding text. Foucault continued: It results from a complex operation whose purpose is to construct the rational entity we call an author. Undoubtedly, this construction is assigned a 'realistic' dimension as we speak of an individual's 'profundity' or 'creative' power, his intentions or the original inspiration manifested in writing. Nevertheless, these aspects of an individual, which we designate as an author (or which comprise an individual as an author) are projections, in terms always more or less psychological, of our way of handling texts: in the comparisons we make, the traits we extract as pertinent, the continuities we assign, or the exclusions we practise. In addition, all these operations vary according to the period and the form of discourse concerned. A 'philosopher' and a 'poet' are not constructed in the same manner [. . .] (p. 127) Here we have reached the heart of Foucault's the figure of the argument: a "realistic" dimension', is an interpretative author, for all that it is 'assigned texts'. That is to say, 'the construct, which arises from 'our way of handling author' of a text is categorically distinct from the historical individual who wrote that text, for all that the two bear?or seem to bear?the same name. We can now appreciate the force of Foucault's opening negation: it was designed to of the 'creator' a text the 'individual' to whom a discoursedistinguish (that is, from the corresponding This distinction or-text is attributed) author-figure. was of course implicit in Barthes's 'La mort de l'auteur'; but where Barthes had seen merely an obstacle to be overthrown, Foucault an rightly perceived of problem. And he thereby opened the way to a new understanding explanatory the meaning of authorship. the terms in which Foucault expressed Admittedly, this point were far from clear. It remained whether 'the' entirely ambiguous of the author-function was one phenomenon or several;35 the very concept account of that concept 'author-function' was never defined; and Foucault's But in fact, the distinction he turned out, as we shall see, to be incoherent.36 was drawing did not depend on the 'author-function' concept; indeed, it will 34 In its usual Foucauldian reference, 'discourse' is transtextual: individual texts are merely its instances, or the sites of its manifestation. If 'discourse' is equated with 'text', the separate category of 'discourse' collapses. Nevertheless, the implied equivalence between the two was repeated under [4] ('What is an Author?', p. 129): cf. p. 353 below. 35 This ambiguity was twofold, applying both over time and as between different 'discourses'. It became most explicit in its historical aspect, and specifically within [3], where Foucault observed that 'all these operations vary according to the period and the form of discourse concerned' and then immediately added that 'there are, nevertheless, transhistorical constants in the rules that govern the construction of an author' ('What is an Author?', p. 127; cf. p. 352 below). 36 See p. 359 below.

ADRIAN

WILSON

351

be more easily appreciated if we set that concept aside and rephrase the point in somewhat different terms. The outlines of Foucault's distinction become clearer if we introduce the the historical individual who wrote the separate term 'the writer' to designate if distinct from the author to whom we assign that text?and given text?as wre focus upon a specific example. I shall take the case of the For this purpose writer John Locke, who produced a number of works including Two Treatises and An Essay concerning Human Understanding, and who died of Government in 1704. The difference 'writer' and 'author', in this particular between case, can be indicated?partially and schematically?by the accompanying table. Work Two Treatises Essay Writer John Locke John Locke Author Locke Locke Identity of author Political philosopher Philosopher of knowledge

not one author but Observe, to begin with, that the name 'Locke' designates two, each with a definite identity?an identity which arises from the use to which we put the respective texts. One of these 'Lockes' is a political philosopher, who of knowledge, wrote the Two Treatises of Government] the other is a philosopher who produced the Essay concerning Human Understanding. Clearly it follows from this alone that each of these 'Lockes' is distinct from the writer John Locke. Correspondingly, writer and author do not have quite the same name: the writer was named John Locke, whereas both the 'political philosopher' of knowledge' and the 'philosopher are known simply as 'Locke'. Curiously? and as we shall later see, significantly?Foucault did not make this point; but of authors, albeit with certain exceptions, it is in fact characteristic that they are known by their surnames alone.37 Again, John Locke wrote many other ceuvres of both 'Locke'-the-politicalworks, which fall outside the respective and 'Locke'-the-philosopher-of-knowledge.38 Further, John Locke philosopher died in 1704, whereas both of our two 'Lockes' are alive today?for we routinely assert that 'Locke argues', 'Locke claims', and the like.39 In short, the author is indeed distinct from the writer, just as Foucault was claiming.40 In fact the dif? ference between them corresponds distinction exactly to Michael Oakeshott's between the 'historical pasf, i.e. the past that has passed, and what Oakeshott calls the 'practical past", i.e. the past as present in our culture.41 Writers are bodily, mortal beings, who lived and died in the historical past. Authors, on the contrary, are living figures who inhabit the practical past; although they too may turn out to be mortal?for instance, it has been well observed that 'Addison and Steele are dead'?their death is not a bodily event but a cultural occurrence, the 37 Cf. p. 358 below. 38 See Ian Harris, The Mind of John Locke (Cambridge: Cambridge University Press, 1994). 39 See Adrian Wilson, 'What is a Text?', Studies in History and Philosophy of Science (in press). 40 In my 'What is a Text?' this distinction is conveyed in different terms, using 'author' for 'writer' and 'virtual author' for 'author'. I have returned here to Foucault's usage, chiefly in an attempt to be as fair as possible to his account. 41 See Michael Oakeshott, On History and Other Essays (Oxford: Blackwell, 1983), pp. 1-44 (particularly pp. 16-19, 34~44)> IO?; and Wilson, 'What is a Text?'

352 mortality canonical

Foucault

on the

'Question

of the Author'

of particular canonical texts.42 Conversely, it is precisely the life of the text (such as the Two Treatises) which gives life to the author (in this The characteristics of that constructed case, Locke-the-political-philosopher). 'the author' then?to return to Foucault's formulation?from 'our arise, figure of texts'. But what is that this we must also notice way handling phrasing of Foucault's has transformed his picture ofthe relation between text and author; for it detaches the figure of the author from the figure of the text. In the previous Foucault had depicted the authorstages of his argument, figure as being tied in one way or another to a text. But in the passage we are considering, this link between text and author has been broken; a text in itself does not 'point to' an 'author' (Foucault's first formulation), nor does an author's name 'accompany' a text (his second designation); rather, the authorfigure arises from 'our way of handling texts'. Thus the figure of the author no longer inheres in a text; rather, it is superimposed upon it. Or to put this the other way around, 'texts' have now been depicted as innocent raw materials, to which we apply those interpretative which construct the authorprocedures And this altered designation function. as Foucault went on, in his persisted of point [3], to argue that 'the rules that govern the construction elaboration of an author' show certain 'transhistorical constants' this (p. 127). To illustrate claim he invoked the example of literary criticism?suggesting that the concept embodies a distinct set of 'critical 'author', as used in contemporary criticism, modalities' which derive from early Christian exegetical theories such as those set out by St Jerome in De viris illustribus. as coherence and Such principles which served for as criteria for to Jerome consistency, assigning authorship in which the of devices texts, recur today (Foucault observed) repertoire by 'the author serves to modern criticism ties texts to their authors: for instance, neutralize the contradictions that are found in a series of texts' (pp. 127-29).43 In this formulation, the connection between text and author is purely exterior; in the text itself. Thus Foucault's the figure of the author has no grounding of his of fulcrum had the curious ef? argument?has exposition point [3]?the fect of eliding the issue with which he began, namely the bond between text that holds between an author and a text, and author, 'the singular relationship the manner in which a text apparently points to this figure who is outside and precedes it' (p. 115). Foucault's was Indeed, point [4], to which he devoted the next paragraph, He began by observing concerned with precisely this problem. and rebutting the very implication just noticed: However, it would be false to consider the function of the author as a pure and simple reconstruction after the fact of a text given as passive material, since a text always bears a number of signs that refer to the author. Well known to grammarians, these textual signs are personal pronouns, adverbs of time and place, and the conjugation of verbs. (p. 129) From these opening sentences of the paragraph it would seem that Foucault 42 Brian McCrea, Addison and Steele are Dead: The English Department, its Canon, and the Professionalization of Literary Criticism (Newark, NJ: University of Delaware Press, 1990). 43 The four attributes specified by Jerome were quality, coherence, consistency, and chronological compatibility.

ADRIAN

WILSON

353

was now reverting to his original notion?that 'a text apparently points' to an authorial figure?and relatedly, that he was restoring the bond between text and that bond author which had just been dissolved in the course of [3]. Specifically, was now apparently by the textual 'signs that refer to the author'? supplied Yet as it turned as 'shifters'. which Foucault adapting Jakobson, designated, to Foucault's do not in fact these 'shifters' out, (according exposition) play quite at once went on to argue that in 'texts with an author' this role. For Foucault he More particularly, multidirectional the 'shifters' are essentially (pp. 129-30). reveals what this "author-function"' that supports claimed that 'all discourse he called a 'plurality of egos', which play different authorial roles. (To illustrate this point he used the examples of 'a novel narrated in the first person' (p. 129), and 'a mathematical treatise' interpret ed with the aid of Booth's conception,44 which he these uses to little later the concrete take a I shall put up (p. 130); and from 'the division that it is specifically And he suggested instances.45) arises. On this distance' between these different 'egos' that the author-function to the extent that a text 'apparently points to' an authorial figure, interpretation, it does so at most obliquely. the same paragraph had effected a further subtle transformation. Meanwhile 'text' was inherently initial formulation, It will be recalled that in Foucault's in but that this was swiftly displaced tied to 'author', by a second picture, which 'the name of an author is a variable that accompanies only certain texts of others' (pp. 115, 124).46 Now, under point [4], Foucault to the exclusion for as we have just seen, was seemingly reverting to his original construction; he asserted here that 'a text always bears a number of signs that refer to the author' (emphasis added). Yet in fact the matter was more complex than this. between anew the distinction to introduce at once proceeded For Foucault one'?but he now drew a different 'texts with an author' and 'those without of the authorline between these, thereby introducing yet a third formulation text relation. The opening ofthe paragraph, which was quoted above, continued thus: But it is important to note that these elements have a different bearing on texts with an author and on those without one. In the latter, these 'shifters' refer to a real speaker and to an actual deictic situation, with certain exceptions such as the case of indirect speech in the first person. When discourse is linked to an author, however, the role of 'shifters' is more complex and variable [. . .] (p. 129) with speech are now connected without-an-author Thus texts (or discourse)47 How are we to map onto this and with a deictic act, i.e. an act of demonstration. which were given in the examples of texts-without-an-author third formulation 'an anonymous Foucault's second formulation?'a private letter', 'a contract', nor can we reconcile is possible, poster attached to a wall' ? No such mapping the assertion that 'a text always bears a number of signs that refer to the author' is unattainable or the third. Consistency with either the second formulation 44 Foucault, 'What is an Author?', p. 129. Here Foucault referred to Booth's The Rhetoric of Fiction only allusively; the citation was supplied by the editor. Cf. above, at n. 16. 45 Cf. p. 358 below. 46 See p. 348 above. 47 Cf. p. 350 above.

354

Foucault

on the

'Question

of the Author'

used the word 'text' in three different ways: here, for Foucault has successively its meaning has slipped from the authored, through the written, to the uttered. two linked rhetorical What has happened here? We have been witnessing moves. In the first place, 'author' and 'text' have been treated asymmetrically: in the very act of bringing the author into focus, Foucault has pushed the figure of the text outside the circle of his interrogation. the sliding usage Secondly, of 'text'?taking in along the way a half-suppressed with 'discourse' synonymy and, as we shall shortly see, with the literary 'work' as well?has deprived that term of any consistent meaning. And together these moves have conferred as upon 'texts' a state of innocence, even as 'the author' has been depicted The of the the as author, carefully depicted essentially fallen. figure product of our interpretative contrasts practices, strangely with the figure of the text, which has come to acquire the character of a quasi-natural object, a simple and equally strangely, the assigning of authorship has given. Correspondingly, at no point been depicted as having any effect on the status of texts themselves: that is, texts are curiously unaffected an authorial by the act of constructing figure. In short, the figure of 'the text' has slipped away, passing unnoticed beyond the analytical horizon.

4. The Text and the Work And yet, for all this, Foucault did assign to the realm of the textual a condition to what I have called, in paraphrase, the fallen state of 'the corresponding author'. That condition he associated with the specifically concept of the work, which as we have seen he criticized in the course of his demolition of Barthes.48 It has been understood that the task of criticism is not to reestablish the ties between an author and his work49 or to reconstitute an author's thought and experience through his works and, further, that criticism should concern itself with the structures of a work, its architectonic forms, which are studied for their intrinsic and internal relationships. Yet, what of a context that questions the concept of a work? What, in short, is the strange unit designated by the term, work? What is necessary to its composition, if a work is not something written by a person called an 'author'? [. . .] If an individual is not an author, what are we to make of those things he has written or said, left among his papers or communicated to others? Is this not properly a work? (pp. 118-19) Here Foucault was point ing out, quite correctly, that Barthes's critique of 'the author' entailed the need for a critical reappraisal of the category of 'the work'?and as we have seen, Barthes was indeed to take this point in his 'De 1'ceuvre au texte' of 1971.5? Foucault went on to illustrate his case with a concrete example; remarkably enough, he chose for this purpose the 'works' of his own hero Nietzsche: Assuming that we are dealing with an author, is everything he wrote and said, everything he left behind, to be included in his work? This problem is both theoretical and practical. 48 See p. 346 above. 49 The phrases 'an author and his works' and 'the man and his work' (the latter in quotation marks) had also been introduced a little earlier, when they were instantly displaced by the phrase 'an author and a text' ('What is an Author?', p. 115, quoted above, at pp. 342, 344 respectively). 50 See pp. 343-44 above.

ADRIAN

WILSON

355

If we wish to publish the complete works of Nietzsche, for example, where do we draw the line? Certainly, everything must be published, but can we agree on what 'everything' means? We will, of course, include everything that Nietzsche himself published, along with drafts of his works, his plans for aphorisms, his marginal notations and corrections. But what if, in a notebook filled with aphorisms, we find a reference, a reminder of an appointment, an address, or a laundry bill, should this be included in his works? Why not? These practical considerations are endless once we consider how a work can be extracted from the millions of traces left by an individual after his death. Plainly, we lack a theory to encompass the questions generated by a work [. . .] (p. 119, emphasis added)51 Let us observe in passing that Foucault here raised an issue of signal impor? come to receive increasing attention: that is, the tance which has subsequently construct point that the concept of 'the work' or 'the works' is a retrospective one, linked of course to the constitution and, for the most part, a posthumous remarks is that 'the of a textual canon.52 But the striking feature of Foucault's is given exactly the same status as 'the work' here, in his prefatory discussion, in the body of his lecture. Just as 'the author' is a author' will be assigned 'function' constructed (not simply an existent or once-existent person), so 'the work' is a constructed given); just as entity (not simply a natural or empirical that 'the 'author' remains an open question' Foucault had begun by observing the ques? (p. 113), so here he pointed out that 'we lack a theory to encompass or could tions generated by a work'. The figure of 'the work', then, comprises, from Foucault's that is to which was the term missing argument: very comprise, in the correlate of 'the author'. the two it textual Indeed, passages say, supplies with where he used the word 'work' again, Foucault treated this as synonymous and Foucault played as 'text'.53 Elsewhere, however, 'the work' was suppressed this bond, we have seen with 'text'?first tying it to 'author', then loosening then seemingly by tying it once again, only to dissolve it finally and decisively the 'questions the textual into the uttered. Correlatively, generated merging as rhetorical they had been posed precisely by a work' were left unanswered: their role in as a counter to Barthes, and this exhausted specifically questions, Foucault's exposition. In sum, Foucault's of the author' was achieved at insight into the 'question blindness with respect to what we may call 'the question the price ofa systematic 51 A similar point had been made in the previous year (1968) by Genette, in an essay entitled '"Stendhal"' (with quotation marks). See Gerard Genette, Figures of Literary Discourse, trans. by Alan Sheridan (New York: Columbia University Press, 1982), pp. 147-82. On Nietzsche's significance for Foucault see Burke, Death and Return, p. 96 (and cf. pp. 112-13). 52 See e.g. Frank Kermode, 'Institutional Control of Interpretation', in his Essays on Fiction igyi-82 (London: Routledge, 1983), pp. 168-84; Terry Eagleton, The Function of Criticism from 'The Spectator' to Post-Modernism (London: Verso, 1984); Jerome J. McGann, 'The Text, the Poem, and the Problem of Historical Method', in his The Beauty of Inflections: Literary Investigations in Historical Method and Theory (Oxford: Clarendon Press, 1985), pp. 111-32; Dale Spender, Mothers of the Novel: 100 Good Women Writers before Jane Austen (London: Routledge, 1986); McCrea, Addison and Steele are Dead; Gracia, 'Texts and their Interpretation'; John Guillory, Cultural Capital: The Problem of Literary Canon Formation (Chicago: University of Chicago Press, 1993)53 These were the comments on 'modern literary criticism' in relation to St Jerome ('What is an Author?', pp. 127?28), and the discussion of Marx and Freud (pp. 131-36; cf. above, at the end of Section 2). The same equivalence had also emerged earlier, at p. 115. On the relation between 'text' and 'work' cf. McGann, 'The Text, the Poem, and the Problem of Historical Method'.

356

Foucault

on the

'Question

of the Author'

of the text'.54 This imbalance reflected a threefold rhetorical the asymmetry: of'text' terminological split between 'work' and 'text', allied tothe ambiguities itself; the structural dispersal of 'work' and 'text' between frame and argument; the answered 'the author' and the and the gulf between question concerning unanswered 'the work'. question concerning

5. The Author's

Name,

the Author's

Individuality

We have just seen that the aporetic quality ofthe figure ofthe text in Foucault's lecture arose not only from his shifting usage, but also from the dispersal of 'text' and 'work' within his essay. The complementary theme of the author's name reveals a similar dispersal; for this, like 'the work', was wholly encased within Foucault's and was thereby separated from the authordiscussion, prefatory function itself. And as we shall see, this arrangement again had a rhetorical task to perform?in this case, helping to suppress the individuality of the authorfigure. Foucault the theme of the author's name as a pair of questions: presented 'What is the name of an author? How does it function?' (p. 121). The rhetorical in its own right, for structure of his response to these questions is interesting from I will attempt to he began with a disclaimer a ('Far solution, offering indicate some ofthe difHculties related to these questions'), and yet concluded with a series of positive theses (p. 123): an author's name 'serves as a means of it 'remains at the contours of texts'; its function is precisely the classification'; 'author-function' the existence, and itself, namely, 'to characterize circulation, certain discourses a of within One operation society'. might perhaps paraphrase this cluster of arguments as by saying that the name of the author functions a form of linkage between texts and discourses?though any such gloss would in view of the uncertain be hazardous, and overlapping reference of 'text' and in 'What is an Author?' 'discourse' It is also worth noting that it was at the of this passage that Foucault introduced conclusion his second formulation of the relation between 'text' and 'author'?the notion that 'the name of an author is a variable that accompanies of others' only certain texts to the exclusion But these tele of the derived their force from Foucault's (p. 124).55 passage he made along the way. particular handling of the observations In the first place, the name of an author Those observations were twofold. of proper names in general, to wit, that such names 'poses all the problems' 'oscillate between the poles of description and designation, and [. . .] they are not totally determined or either their functions' by descriptive designative an author's name leads to a of additional series difficulties 121).56 Secondly, (p. of its own, since 'the link between a proper name and the individual being named and the link between an author's name and that which it names are not 54 Cf. Paul de Man, Blindness and Insight: Essays in the Rhetoric of Contemporary Criticism (first published as a collection 1971; 2nd edn, revised, Minneapolis: University of Minnesota Press, !983), p. 16 and passim; and Wilson, 'What is a Text?' 55 Quoted above, p. 348. 56 Reading 'by either' for the translators' 'either by'.

ADRIAN

WILSON

357 name

and do not function in the same way'. That is, 'an author's isomorphous is not precisely a proper name among others' (p. 122). For example: It is altogether different to maintain that Pierre Dupont does not exist and that or Hermes Trismegistus have never existed. While the first negation merely that there is no one by the name of Pierre Dupont, the second indicates that individuals have been referred to by one name or that the real author possessed the traits traditionally associated with Homer or Hermes. (p. 122)

Homer implies several none of

We might indeed pause to meditate both upon the fact that 'the author' is assigned a personal name, and upon the double significance?descriptive and designative?of personal names in general. What this twofold significance constructs, surely, is the who, i.e. the concept of personal identity, an identity in the personal name which is inseparable from the name itself. It is precisely attain that fusion which is otherwise that signifier and signified elusive yet with the is always being sought. The proper name, then, is doubly freighted notion of identity: in that to which it refers, and in the mode of its reference.57 And to the extent that, as Foucault observed, 'the name of an author poses all the problems related to the category of the proper name', the author's name, That is, an author is always and necessarily too, carries this same significance. a named individual?as Foucault had made clear when defining the 'question of the author' in the first place.58 in mind, let us observe Foucault's With these considerations treatment of the author's name. We have seen that he depicted the author's name as both the personal name and differing from it. These two aspects were resembling the initially in balance; but as this part of his prefatory discussion proceeded, weight of his picture moved steadily towards a stress on the difference between authorial names and personal names. And eventually, at the culmination of the he detached the one from the other: passage, actually We can conclude that, unlike a proper name, which moves from the interior of a discourse to the real person outside who produced it, the name of the author remains at the from the other, defining their form, and characterizing contours of texts?separatingone their mode of existence. (p. 123, emphasis added) of course, was to clear the space for Foucault's point of this contrast, distinction between writer and author. But (in my own paraphrase) subsequent his rhetoric here has effaced the zone of overlap between an author's name and a proper name; and this move merely replaces one mystery by another. Every? day usage conceals the difference between the name of the author and ordinary of Foucault's this analysis is that it has brought proper names; the strength to light. But the price that has been paid is to suppress difference the resemblance between the two?a resemblance which is in fact the very condition of the 'author-function', for it is precisely as the bearer ofa name that 'the author' the cultural role which Foucault is attempting to disclose. So too performs from his Foucault has now, by the end of his prefatory passage, eliminated 'the link between an author's name and that which it names' problem-space The 57 Cf. Jacques Derrida, Of Grammatology, trans. by Gayatri Chakravorty Spivak (Baltimore: Johns Hopkins University Press, 1974), pp. 107-18. 58 See pp. 341, 344 above.

358

Foucault

on the

'Question

of the Author'

issues concerning the (p. 122, quoted above), and with it, all the associated concept of identity. As the sentence just quoted makes clear, the name of the author no longer either designates or describes; instead, it is to be characterized in two its the sense of discursive effects. That is, of Foucault's its function, by is the name of an author? How does it function?' original questions?'What It is precisely this the second now survives.59 (p. 121, quoted above)?only nexus which will dominate the remainder of Foucault's functional paper: in? deed, this is the very point ofthe term 'the author-function'. of the author's name perhaps This self-limitation of Foucault's treatment in his whole discus? a certain curious and ironic oversight makes intelligible the author's name from ordinary per? sion. Despite his concern to distinguish the author from the writer, Foucault nowhere sonal names, and to demarcate associated with authorial names: the fact remarked upon the chief convention known by their surauthors are commonly that (as I have already remarked) himself used ('Mallarme', Foucault names alone.6o That convention?which him in turn?has several and by which I have here been designating 'Beckett'), until women authors It tends to be features. recently, gendered (for interesting it inverts were usually assigned given names?Jane Austen, Emily Dickinson); which prevails in daily life (the better we 'know' the the order of familiarity to a author, the more we suppress the author's given name); it is associated, and exceptions (for instance, degree, with death; it shows various modulations male poets such as W. B. Yeats, T S. consider the initialized twentieth-century is the of a possible enquiry which would site e. e. Here, then, Eliot, cummings). with the very themes that Foucault was opening up, and which be concerned did not would support the central distinction he was drawing. Yet Foucault it. And we can take this as a much less thematize even notice this phenomenon, the question 'What is the name of an symptom ofthe fact that he had suppressed his attention to the question 'How does it function?' More? author?', restricting Foucault later over, having here banished the author's name from consideration, He had of course begun from the went on to abolish the author's individuality. and he retained this principle that the is an individual; premiss author-figure But the first three 'characteristics ofthe "author-function"'.61 as he elaborated of the various in the course of developing is, in his discussion point [4]?that the eliminated textual 'signs that refer to the author', or 'shifters'?Foucault of the author, as we shall now see. individuality to the This remarkable transmutation by two steps, corresponding proceeded 'a novel narrated in the first that Foucault considered: two concrete examples treatise' (pp. 129-30).62 Using the first example, person', and 'a mathematical in fact point not to Foucault argued, as we have seen, that textual 'shifters' that it is from the a single individual but rather to a 'plurality of egos'?and 'arises'. between these 'egos' that the author-function 'division and difference' of egos', then, serves as the occasion for the authorThe textual 'plurality 59 Note that designation and description themselves were described as 'functions', but these were functions of a different kind?pertaining to signification rather than to efhcacy. 60 See p. 351 above. 61 See e.g. the quotation at p. 350 above. 62 Cf. p. 353 above.

ADRIAN

WILSON

359

is based. By this as the material function, upon which the author-function while the reference of textual 'shifters', means Foucault divided and pluralized the and of authorthe nevertheless integrity retaining individuality implicitly the responsibility quietly transferred figure. But the next step of his exposition itself. He for this 'division' from the textual 'shifters' to the 'author-function' of 'a mathematical to the further example now proceeded treatise', showing his gloss depicted the that here too a 'plurality of egos' could be detected?but Hence 'plurality of egos' in such a text as the product of the author-function! of point [4]: which he reported in his summary the conclusion does not refer, purely and simply, to an actual individual in so far [the author-function] as it simultaneously gives rise to a variety of egos and to a series of subjective positions that individuals of any class may come to occupy. (pp. 130-31, emphasis added)63 that of the latter clause?'a series of subjective While the meaning positions of any class may come to occupy'?is individuals (to this reader, at least) entirely does unclear, the general bearing ofthe point is not in doubt: the author-function ofthe figure ofthe author?the not refer to an individual at all. The individuality lecture?has now been erased. And this completes very premiss of Foucault's had begun by suppressing from consideration the the move which Foucault meaning of the author's name. what Foucault To recapitulate: captured was the constructed nature of 'the the personal being and author'; what his analysis buried was the individuality, This was, of course, the inversion of what happens identity, ofthe author-figure. was since Foucault's in conventional enough, argument usage?appropriately is is that this But what that striking suppression very usage. pitched against was achieved of the author's personal being, far from being argued explicitly, Foucault's Furthermore, concept ofthe 'authorby rhetorical sleight-of-hand. in the course of his own exposition, into incoherence. function' degenerated, the author-function as the product of the At first, under [3], he had portrayed ofthe critic (pp. 127-28). Then, within [4], he depicted reader, and particularly from the 'division and as arising from the text?specifically the author-function with 'a novel narrated in of egos' associated between the 'plurality distance' Yet as we have just seen, this posited relation the first person' (pp. 129-30). inverted?for under the further example of 'a mathematical was immediately treatise', Foucault argued that the 'plurality of egos' arose not from the textual In short, the 'author-function' 'shifters' but instead from the author-function. as as cause and as effect, arising from the text and as imposed variously appeared which it. The confusion everyday usage effects between writer and author, upon in was transferred had been at such pains to remove, and which Foucault the 'author-function' form to the concept of the 'author-function': concealed itself. of the author-figure came to inherit the original ambiguities These aporiai at the heart of Foucault's argument suggest that his attempt to from a 'who' into a 'what' was doomed by its very nature. turn the author-figure was unable to sustain in practice the vision he was seeking Foucault Certainly for in fact named authors were invoked at the strategic sites of his to construct; Cf. Foucault, The Archaeology of Knowledge, p. 95.

360

Foucault

on the

'Question

of the Author'

The author-function itself was assigned an author, in the person exposition.64 of St Jerome. So too, as we saw earlier on, the author's had 'disappearance' its author: Mallarme. And the 'direction' of Foucault's the enquiry, question 'What matter who's speaking?', of dis? was 'supplied' not by the murmuring course but by a named author of the present day: Beckett.65 Thus my analysis Sean Burke's tactical rewriting of Foucault's that strongly supports question, is to say, the shift from 'what is an author?' to 'what (and who) is an author?'66 Burke's reformulation may be glossed as arguing that the &;/mZ-question which Foucault posed turns out to entail, and necessarily which so, the w/zo-question he suppressed;67 and this is the result to which Foucault himself has unwittingly led us. The author-figure is indeed a construct, just as Foucault argued. But thesis which he developed contrary to the accompanying by covert means, the is constructed as a personal being. author-figure specifically

6. Conclusion The signal achievement of Foucault's 'What is an Author?' was to reveal that the a construct associated construct: with figure of the 'author' is an interpretative canonical works, notionally identified with the writer of such works, but none the less categorically distinct from that writer. On the other hand, Foucault's to a considerable was masked discovery degree by the very terms in which he the author-figure's articulated it?above his all, by eliminating personal being. Yet as I shall go on to suggest, the attendant aporiai were not only of Foucault's own making, but also stemmed from difficulties inherent in the issues he was raising. Foucault's and to raise attempt to write out the author-figure's individuality, in its stead the impersonal arose of course from concept of 'the author-function', the larger concern which dominated his writings of the late 1960s: namely, to of In 'What is an Author?' this establish the putative discourse. sovereignty effort was a radical failure, at several levels. We have seen that the suppression of the author-figure's and personal name, individuality, being rested upon a series of covert rhetorical manoeuvres; that this suppression was self-defeating, since in fact it is precisely as a person that the author-figure is constructed; that the 'author-function' and that Foucault concept had no consistent meaning; himself was incapable of adhering to the stance which he was concerned to define. Further, Foucault's intended erasure ofthe author-figure's individuality was paradoxically at odds with his own wider purposes. For the larger question 64 pp. 115-16, 138, 127; see Sections 1-3 above. It is possible that the writer Michel Foucault was aware of these ironies; but we may say that the author Foucault was/is not, since his text did not/does not draw attention to them. 65 Indeed it might be said that the triad Jerome-Mallarme-Beckett together comprise a unity, a conjoint emblem of authorship or of its history. Perhaps this triad corresponds to the triad pastpresent-future; or perhaps Mallarme and Beckett together define the present, and the future is by yet another author, namely Foucault himself (cf. p. 347 above). supplied 66 Cf. p. 339 above. 67 This is, of course, by no means the only possible interpretation of Burke's question. For instance, we might take Burke as asking: to whom does Foucault reserve the privilege of authorship ? The latter reading would be in line with Burke's focus on the Marx-Freud section of Foucault's lecture (see p. 348 above).

ADRIAN

WILSON

361

he wanted to raise was 'the privileges of the subject' (p. 137); yet he had eliminated which the author-figure shares with the figure just those attributes of the subject.68 But perhaps the supreme to the concept paradox pertained of 'discourse'; and here my reading joins hands with that of Burke. For what Burke's analysis reveals is that in the sovereignty which Foucault assigned to of discours was neither more nor less than the the it, hypostasis figure of the author. That is to say, what Foucault himself wrote of Barthes's ecriture?that this concept 'has merely transposed the empirical characteristics of an author to with equal force, and with a transcendental anonymity' (pp. 109-10)?applies discours itself.69 signal irony, to Foucault's The key to this twofold is the fact that both ecriture and hypostatization of writing to speech. Just as Barthes set up discours rested on the assimilation a written text with the question 'Who is speaking thus?', ecriture by approaching of discours by applying to the written at invoked the sovereignty so Foucault 'What matter who's speaking?'70 In the case of ecriture this large the question not only is seemingly since as Ann Banfield has observed, manoeuvre puzzling, in opposition to speech' but also the division between the is ecriture 'conceived written and the spoken is enhanced grammar of the French by the particular if attend to the rhetorical is resolved we But this apparent paradox language.71 Barthes's of the written as work which was being performed apprehension by of every voice, is 'the destruction ecriture On Barthes's designation, spoken. inherent in this picture is that in contradiction every origin'.72 The irreducible order to play such a role, ecriture must itself acquire the status of voice and of the written under origin; and this was just what was achieved by apprehending the sign of the spoken. In the most literal sense, then, ecriture was the answer to the question 'who is speaking?' The very terms of Barthes's own rhetoric reveal Foucault accurately observed?the concealed that the figure of ecriture was?as of the author-figure. hypostatization the written to the In the case of discours, the rhetorical stroke of assimilating L Archeologie du savoir For as Foucault's spoken was rather less paradoxical. was the enonce, made clear, the unit from which discours was to be reconstructed this concept itself merged writing into speech. Moreover, the statement?and that is to say, 'surthat what he called 'remanence', Foucault further explained This inherent vival in time', is 'of the nature of the statement'.73 property or enonce was defined as the in time' meant that the statement of 'survival of course entails that the written is apprehended written-in-presence?which mode of fusing writing and speech was more But if Foucault's as speaking. than that of Barthes, the effects of this move were similar in each transparent dis? case: what we have seen of Barthes's ecriture is equally true of Foucault's announces 'what matter who's speaking?' unwittingly cours. The very question Thus and that it is speaking which 'matters'. is 'speaking' both that someone 68 See the the between discussion of and for a relationship penetrating summary on p. 347 above, the two see Burke, Death and Return, pp. 104-15. 69 Cf. p. 346 above. 70 See pp. 340, 342 above. 71 Banfield, 'Ecriture, Narration and the Grammar of French', p. 2. 72 Barthes, 'La mort de l'auteur', p. 61. 73 See Foucault, The Archaeology of Knowledge, p. 124.

362

Foucault

on the

'Question

of the Author'

the properties?voice and origin, discours, like ecriture, is assigned precisely have been so insistently re? agency and authority, presence and power?which moved from the figure of the author. In short, Foucault's rhetoric, like that of thesis: 'the principle of the author most Barthes, bears out Burke's principal reasserts itself when it is thought absent'; 'the concept ofthe author powerfully is never more alive than when thought dead'.74 There is also a further respect in w7hich the present exploration has harmonized with Burke's argument. One of Burke's central insights is that Foucault, like Barthes, apprehended such themes as writing, authorship, and language, within a larger vision of past, present, future, and their mutual artidiscourse culation. That is to say, both Barthes (in his writings of c. 1970) and Foucault his each in his own way, an eschatology75 (in 'archaeological' phase) deployed, called a historical and, linked with this, what I have elsewhere metaphysic.76 And this is just what we saw at work in the Barthes-Foucault over exchange the death (Barthes) or disappearance of the author. The gesture (Foucault) with which Barthes consigned to the past was matched, Mallarme in riposte, assimilation of Mallarme to the present; in each case, what we by Foucault's make of Mallarme is inextricably with what we make of ourselves. conjoined The sense of time at work in such a vision is by no means simply a matter of For instance, Foucault could assign Mallarme to 'our sequence. chronological time' while also invoking 'our present discontinuities' and implicitly relegating Barthes to 'the historical and transcendental of the nineteenth tradition cen? in time backwards tury': here, within a single gesture, the present is extended is defined as a moment of rupture (discontinuities), and is depicted (Mallarme), as burdened with the weight of a lingering past (Barthes). So too the culture of the nineteenth and as villainous century appears both as heroic (Mallarme) ('the historical and transcendental Thus in such a tradition'). metaphysic, past and sites in an evaluative present are not points in a sequential array but evocative matrix.77 Moreover, as Burke has shown, Foucault's vision of the sovereignty of discourse was intimately bound up with this larger metaphysic.78 as has already been mentioned, the troubles which entangled Nevertheless, Foucault's into the reflect not insight author-figure just his own particular meta? but also the inherently physical commitments refractory nature ofthe issue with which he was dealing. For it remains no easy matter to articulate and to clarify the central thesis which Foucault was advancing, i.e. the constructed nature of the author-figure. The difnculty arises not only because the issue of authorship extends across a large and complex for instance, both fic? field?embracing, tion and non-fiction, both descriptive and normative also, and concepts79?but 74 Burke, Death and Return, pp. 6, 7, quoted above, p. 348. 75 Foucault, The Archaeology of Knowledge, pp. 22-27, 9?, and/><w?m. 76 Adrian Wilson, 'A Critical Portrait of Social History', in Rethinking Social History: English Society and its Interpretation, i$yo-ig2o, ed. by Adrian Wilson (Manchester: Manchester Uni? versity Press, 1993), PP- 9-58 (pp. 34-36)77 For analogous gestures on the part of Barthes see nn. 9, 23 above. 78 Burke, Death and Return, ch. 2. 79 (a) Foucault's 'author-function' applied indifferently to fictional and non-fictional authors and was descriptive in intent. (b) The 'postulated-author' concept of Nehamas and the 'fietionalauthor' concept of Currie pertain to fiction and have a normative aspect. For some discussion of

ADRIAN

WILSON

363

more fundamentally, because we are here necessarily grappling with the bound? aries imposed by our own conceptual All such figurations?not figurations. just 'author' and 'work' but also, for instance, 'text' and 'document', 'source' and recalcitrant to elucidation, 'evidence', 'past' and 'present'?prove extremely because we them precisely normally get along by employing unreflectively. This is well illustrated the figure of the 'text'. As by the confusion attending we have seen, 'text' proves to have been a densely aporetic term in 'What is an Author?' But this problem is by no means confined to Foucault's use of that the meaning of the figure figure. On the contrary, as I have shown elsewhere, of the text has remained obscure and elusive ever since it began to be applied? in a shift to which Foucault's 'What is an Author?' itself contributed?to the written-at-large.8? such as 'text' and 'author', then, are bound up with Working figurations what Heidegger called the Vorhabe of understanding, that is, the 'fore-having' which assigns-in-advance to the objects of understanding a particular mode of role, serving as the ground being.81 And it is because they play this founding that such figurations prove so resistant to upon which interpretation proceeds, of a The most rhetorical and scrutiny. interesting promise aporetic approach is not just theorizations that such a strategy may help to illuminate such as those of Barthes and Foucault but also, and more fundamentally, the conventional themselves. figurations University of Leeds Adrian Wilson

Nehamas's concept see Robert Stecker, Apparent, Implied and Postulated Authors', Philosophy and Literature, n (1987), 258-71. (c) The concept ofthe 'virtual author' (Wilson, What isa Text?'; cf. nn. 39, 40 above) relates to non-fictional authorship and is descriptive. (d) The same is true of the more complex conception of Gracia, who distinguishes between the 'composite author' (a figure which incorporates the contribution of editors), the 'pseudo-historical author' (the picture which we construct of the historical author), and the 'interpretative author' (the author whom we posit in the act of interpreting a text). See the essays cited in n. 16 above, and also Jorge J. E. Gracia, A Theory of Textuality: The Logic and Epistemology (Albany, NY: State University of New York Press, 1995); id., Texts: Ontological Status, Identity, Author, Experience (Albany, NY: State University of NewyYork Press, 1996). 80 Cf. pp. 343-44 above. On the ambiguity of 'text' in Barthes's writing see Annette Lavers, Roland Barthes: Structuralism and After (London: Methuen, 1982), pp. 176-77, and Sutrop, 'The Death of the Literary Work', p. 43; on its obscurity in general see my 'What is a Text?' 81 Martin Heidegger, Being and Time (German original 1927; trans. by John Macquarrie and Edward Robinson, Oxford: Blackwell, 1962), pp. 191-95, particularly p. 193. Cf. Wilson, What is a Text?' For a different gloss on Vorhabe cf. Hubert L. Dreyfus, 'Holism and Hermeneutics', Review of Metaphysics, 34 (1980), 3-23 (pp. 10-11), and id., B eing-in-the-World: A Commentary upon Heidegger's 'Being and Time', Division I (Cambridge, MA: MIT Press, 1991), pp. 199-202.