and Law REVIEW ARTICLE zyxw zyxw zyxwvu Copyright and the Death of the Author in Literature zyxwv zy Lionel Bently * Martha Woodmansee and Peter Jaszi,The Construction of Authorship: Textual Appropriation in Law and Literature, Durham, NC: Duke University Press, 1994, 462 pp, pb f16.95. David Saunders, Authorship and Copyright, London: Routledge, 1992, 269 pp, hb f35.00. A The Death of the Author In the essay, ‘What is an Author?’ Michel Foucault drew attention to the fact that the notion of the ‘author’ is socially constructed.’ Foucault claimed that the literary author was invented during the eighteenth century and isolated ‘ownership of the text’ as one of the characteristics of the relationship between the text and the author. Foucault urged us to imagine a culture where discourse would circulate without any need for an author, a world where it did not matter who was speaking.* Roland Barthes went one step further and declared the ‘death of the author.’3Barthes argued that, once published, the text is no longer under the control of the author and that the author is irrele~ant.~ Instead, Barthes asserted that the text is merely a product of other texts and can only be understood through those other texts. Individual authorship of works is to be replaced by interte~tuality.~ Although this radical questioning of the role of the author has not been zyxw universally accepted,6 it has proved extremely influential within literary *Lecturer in Law, King’s College, London. My thanks go to Anne Barron, Adam Tomkins and the Copyright Reading Group at Cambridge. Lambropoulos and Miller (eds), Twentieth Century Literary Iheory: An Introductory Anthology (Albany: SUNY, 1987) pp 124- 142. Moreover, Foucault described the emergence of the modern concept of authorship as ‘the privileged moment of individualization in the history of ideas, knowledge, literature, philosophy and the sciences.’ ibid p 139. Heath (ed), Image, Music, Text (London: Fontana, 1977) pp 142- 148. ibid p 142. zyxwv ‘Any text is a new tissue of past citations. Bits of code, formulae, rhythmic models, fragments of social languages, etc pass into the text and are redistributed within it, for there is always language before and around the text. Intertextuality, the condition of any text whatsoever, cannot, of course, be reduced to a problem of sources or influences; the intertext is a general field of anonymous formulae whose origin can scarcely ever be located; of unconscious or automatic quotations, given without quotation marks’: Barthes, ‘Theory of the Text’ (trans MacLeod) in Young (ed),Untying the Text: A Poststructuralist Reader (London: Routledge, 1981) 31, p 39. Masten, ‘Beaumont and/or Fletcher: Collaboration and the Interpretation of Renaissance Drama’ in Woodmansee and Jaszi (eds), 7he Construction ofAuthorship (Durham, NC: Duke UP, 1994) 361, p 37 1. (‘Like bibliography, much of the more self-consciously interpretive “literary criticism” continues to rely implicitly on the assumption that texts are the products of a singular and sovereign authorial consciousness.’)For a modem defence of authorship, see Hirsch, ‘In Defense of the Author’ in Validity in Interpretation (New Haven: Yale University Press, 1967) pp 1 - 23. In fact, it has been 63 The Modern Law Review Limited 1994 (MLR 57:6. November). Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF and 238 Main Street, Cambridge, M A 02142, USA. 973 The Modem Law Review zy zyxwvuts zy zyxw scholarship and clearly has potential significance for law in general, and copyright law in particular. After all, copyright law is a legal institution which declares itself as designed to recognise the ‘rights’ of authors - indeed, the French equivalent of [Vol. 57 copyright is called ‘droit d’uuteur.’ Copyright law is a system to which the notion of the author appears to be central - in defining the right owner, in defining the zyxwv work, in defining infringement. The critique of authorship in literature thus raises a number of questions for copyright law: what is the relationship between law and zyxwvu literature? Is the legal conception of authorship related to that which has operated in literary theory? If so, must law recognise the death of the author? Has it done so? Even if copyright need not recognise that death, could it or should it do so? Different and often conflicting answers to these questions are offered by David Saunders’ Authorship and Copyright (hereafter AC) and many of the essays in The Construction of Authorship (hereafter TCA) .I B The Historical Connection Between Copyright and Authorship The claim that the concept of authorship in literature is intimately related to that which operates in law is principally an historical claim that copyright law, romantic authorship and the overpowering significance of the author were ‘born together.’8 That is, the link established in law between an author and a work, and the romantic conceptualisation of the work as the organic emanation from an individual a ~ t h o r , emerged ~ simultaneously at the end of the eighteenth century.’O The consequence of this, it is claimed (by Rose, for example), is that the literary critique of authorship threatens the intellectual foundations of copyright law. If the legal walls establishing ownership of the text were built on the same intellectual foundations as romantic authorship, and those premises turn out zyxwvut to be sand rather than rock, copyright will sooner or later come tumbling down. Recently, these historical claims have received some support from the researches observed that, no matter what Barthes or Foucault may have wished, their texts are still attributed to them and probably their estates still reap royalties on their copyrights. See, for example, Miller, Authors (Oxford: Oxford University Press, 1989) p 173. Equally, the notion of authorship has come increasingly to dominate other discourses, such as film, where the idea of the film director as author (‘auteurism’) has taken a firm hold. Naremore, ‘Authorship and the Cultural Politics of Film Criticism’ (1990) Film Quarterly 20: ‘Even though the generation of ’68 produced some of the most valuable and brilliantly iconoclastic writing in the history of film, they never really dispensed with authorship,’ cited by D’Lugo. ‘Authorship and the Concept of National Cinema in Spain’ in Woodmansee and Jaszi (eds), n 6 above, p 327. 7 These essays were delivered at a conference at Case Western Reserve University in 1991 and previously published in (1992) lO(2) Cardozo Art & Entertainment W. 8 Jeffrey Masten is distrustful of the metaphor of birth because it ‘naturalizes and makes inevitable an event - or rather, set of events - that were, as I will suggest, contingent and by no means biological, transcultural, or even uniformly occurring across discourses and genres within a given culture’ (n 6 above, p 363). Jane Gaines argues that there are structural similarities between legal and literary discourse and that ‘[tlhe two discourses inform each other because they share the same cultural root buried deep in the seventeenth century’: Contested Culrure: 7he Image, the Voice and rhe Law zyxwvuts (London: British Film Institute, 1992) p 23. 9 Woodmansee, ‘On the Author Effect: Recovery Collectivity’ in Woodmansee and Jaszi, n 6 above, pp 27- 28, argues that ‘[bloth Anglo-American “copyright” and Continental “authors’ right” achieved their modem form in this critical ferment, and today a piece of writing or other creative product may claim legal protection only insofar as it is determined to be a unique, original product of the intellection of a unique individual (or identifiable individuals).’ 10 Barthes also notes that the author is a modern figure: n 3 above, pp 142- 143. 974 0 The Modem Law Review Limited 1994  zyxwvuts November 19941 zyxw zy zyxwvutsr zyx Copyright and the Death of the Author of historians such as Mark Rose and their conclusions underpin many of the essays in The Construction of Authorship.“ In ‘The Author in Court: Pope v CurZl’ (TCA, pp 21 1-229), Mark Rose describes how the English Statute of Anne, passed in 1710, which had not been zyxwv intended to be for the benefit of authors, came to be used by them.12The Statute of Anne was a trade regulation device reinstating order to the book trade that had been thrown into confusion as a result of the failure to renew the seventeenth- century printing licensing 1 a ~ s . The I ~ Act was concerned with ‘books’ and their ‘proprietors’ (ie the Stationers), not authors and their works. Rose reviews Pope’s use of the Statute of Anne to prevent the publication of letters sent by him to Swift. Rose argues that the legal holding - that the property in the letter passed to its recipient but the literary property was retained by the author - represents a critical moment in the development of intellectual property law. Whereas the Statute of Anne conceived of property in books as physical objects, in Pope v Curlf, Lord Hardwicke recognised the author’s right as an intangible right in the ‘work,’ as distinct from the book. According to Rose, the idea that the author is creator of the text was developed further in the second half of the eighteenth century as part of the sustained legal debate as to whether authors could claim a common law natural right to property in the literature they produce.14 This debate, like the Statute of Anne, was promoted by the London publishers, who sought to extend their statutory monopolies, limited at most to twenty-eight years, by gaining recognition of a perpetual right under the common law. The debate generated a collection of tracts which attempted to substantiate this claim to a common law literary property by reference to the origins and justifications of property and the nature of a~thorship.’~ Lockean discourse of property, which speaks of a natural right of property in the products of labour, was blended with the acceptance of a literary work as the zyxwv 11 For similar historical accounts, see Woodmansee, ‘The Genius and the Copyright: Economic and Legal Conditions of the Emergence of the “Author”’ (1984) 17 Eighteenth Century Srudies 425; Woodmansee, The Author, Ari and the Market: Rereading the History of Aesthetics (New York: Columbia University Press, 1993); Carla Hesse, ‘Enlightenment Epistemology and the Law of Authorship in Revolutionary France, 1777- 1793’ (1990) 30 Representations 109; Roger Chartier, L’Ordres des Livres: Lecteurs, Auteurs, Bibliotheques en Europe entre XIV el XVIII SiPcle (Aix-en- Provence, Paris: Alinea, 1992). His essay ‘Figures of the Author’ (trans Lydia Cochrane) appears in Sherman and Strowel (eds), Of Authors and Origins (Oxford: Oxford University Press, 1994). 12 Act for the Encouragement of Learning (1710) 8 Anne ch 19. 13 For the century and a half prior to the lapse of those regulations, the book trade had been in the control of the guild of Stationers, who had developed their own system of allocating publishing rights amongst themselves. In effect, the Statute of Anne amounted to a reluctant acceptance of this monopoly: Lyman Ray Patterson, Copyright in Historical Perspective (Nashville: Vanderbilt University Press, 1968). It may be that these propositions are overstated and the idea of authorship was already of some zyxwvut importance. The 1710 Act refers to ‘authors’ and makes the continuation of the copyright term from 14 zyxwvutsrqp to 28 years dependent upon the author’s survival. Feather’s claim, in ‘From Rights in Copies to Copyright’ in Woodmansee and Jaszi (eds), n 6 above, p 208 that ‘[tlhe so-called Copyright Act of 1710 mentions neither copyright nor authors’ is wrong. Additionally, the Stationer’s use of the claims of authors in inducing Parliament to pass the Statute indicates that authorship also had some rhetorical power: see Feather, ‘The Book Trade in Politics,’ 8 Publishing History 19, p 45. 14 The debate is also described by Saunders, Authorship and Copyright (London: Routledge, 1992) pp 57-74. 15 Rose, ‘The Author as Proprietor: Donaldson v Becket and the Genealogy of Modern Authorship’ (1988) 23 Representations 5 1, reprinted in Sherman and Strowel (eds), n 11 above; Rose, Authors and Owners: The Invention of Copyright (London and Cambridge, Mass: Harvard University Press, 1993). See also De Grazia, ‘Sanctioning Voice: Quotation Marks, the Abolition of Torture and the Fifth Amendment’ in Woodmansee and Jaszi (eds), n 6 above, p 298 (‘copyright legislation privileging the author emerged at the same time as quotation marks privileging the utterer’). 0 The Modern Law Review Limited 1994 975 The Modern Law Review zy zyxwvuts zy product of an author’s labour to produce a reinterpretation of existing copyright rules as a statutory recognition (rather than generation) of an author’s common law right. Fears concerning the consequential effects of such a right on the public were [Vol. 57 answered by confining the proprietary right to those elements of the work in which the author’s personality is individualised, namely the expression, leaving the underlying ideas free for public use and criticisrn.l6 While the House of Lords ultimately ruled against this common law right, the widespread debate laid a grounding into which the romantic conception of authorship could be reimported from Germany through the likes of Samuel Taylor Coleridge. In this way, Rose confirms Foucault’s hypothesis that the modern conception of authorship was grounded in and intimately linked with claims to literary proprietorship. C The Survival of Authorship in Modern Copyright Assuming a strong historical connection between literary authorship and literary property, a number of attempts have been made to show that the critique of romantic authorship signified by the notion of the death of the author implies a necessary rethinking of the role of authorship in copyright law. If the two concepts of authorship and literary property emerged at the same time, based on common conceptions of individualism, personality and creativity, then it might be reasonable to expect the concepts to disappear at the same time. Indeed, it has been argued that the death of the author in literary theory has already been paralleled by the demise of copyright and its replacement with trade zyxw marks law.I7 Evidence for this tendency is said to be found in the proliferation of cases complaining that a person’s image, look, personality or voice have been misappropriated. The boundaries of copyright law, built on the concept of authorship, no longer correspond to our ideas of what should be protected and the more flexible actions in privacy, personality and passing off have been employed instead. Indeed, the failings of copyright have resulted in a strengthening of the actions which have been developed to fill the gap - which, in turn, make copyright law increasingly insignificant. This argument that ‘copyright is dead’ is, however, unconvincing. Although it is true that copyright law has failed to be the prime legal mechanism for the expression of the needs or interests of those involved in character or personality merchandising, it is difficult to see how this failing represents the death of copyright law. Indeed, the history of copyright is the history of its expansions into new domains - photography, sound recordings, films, computer programs. l9 Only from a very limited viewpoint can its failure to expand into the domain of 16 Rose’s use of Hargrave’s An Argument in Defence of Literary Property (1774) is somewhat problematic. The other tracts Rose cites appear to base their claims to literary property on Lockean theory, but Hargrave’s is unusual in emphasising that the text bears the imprint of the author’s personality. It is therefore something of an overstatement to treat Hargrave as representative of late eighteenth-century legal thought. 17 Gaines, n 8 above, p 25. See also Lury, Cultural Righrs: Technology, Lxgality and Personality (London: Routledge, 1993). zyxwvut 18 In the UK, eg, by Mirage Sntdios v Counter-Feat Clorhing [I9911 FSR 145; in US, see Coombes, ‘Author/izing the Celebrity: Publicity Rights, Postmodern Politics and Unauthorized Genders’ in Woodmansee and Jaszi (eds), n 6 above, pp 101 - 131. 19 Recognised respectively by the Fine Arts Copyright Act 1862 (photographs), Copyright Act 1911 (sound recordings). Copyright Act 1956 (films) and Copyright (Amendment) Act 1985 (computer programs). 976 0 The Modern Law Review Limited 1994  zyxwvuts November 19941 zy zyxwvutsr Copyright and the Death of rhe Author protecting personality rights be seen as signalling copyright’s (or the author-in- copyright’s) demise. Copyright remains a stronger and preferable form of protection for creators and proprietors in the considerable area to which it extends. In fact, as is clear from the essays in The Construction ofAuthorship, there is plenty of evidence that copyright law continues to employ the rhetoric and conceptual underpinnings of authorship, in both the judicial and legislative arenas. The United States Supreme Court’s decision in Feist v Rural Telephones,20to the effect that a telephone directory is not a work of authorship, has been treated as demonstrating the power of romantic preconceptions which continue to inform judicial interpretation of the copyright statute. In that decision, Justice O’Connor declared that it was a constitutional requirement that a work must have some zyxwv creativity to be protected by copyright and that an alphabetically arranged list of names did not bear the stamp of such creativity. For the Supreme Court, then, authorship in law required some expression of personality rather than mere sweat of the brow.21 At a legislative level, the continuing prominence of romantic conceptions of authorship can be seen in the recent recognition, in both the United States and the United Kingdom, of ‘moral rights’ - that is, personal rights of authors and artists to be named in relation to the work and to control alterations of the work.22As Jaszi comments: ‘The instance of moral rights is but one example of how Romantic conception of authorship is displaying a literally unprecedented measure of ideological autonomy in legal context. Recent copyright decisions show that even as scholars of literary studies elaborate a far-reaching critique of the received Romantic concept of authorship, American lawyers are reaching out to embrace the dull range of its implications.’23 zyxwvu The poststructuralist critique of authorship appears so far to have had no significant influence on copyright law which has continued to employ romantic images of authorship, at least in some contexts. This immunity of copyright law’s notion of authorship to the radical destabilisation of the same notion in the literary field seems less surprising, given the historical insights of some of the essays in The Construction of Authorship and the more general insights proffered by David Saunders’ Authorship and Copyright.24 Some of the contributions to The Construction of Authorship indicate that the historical connection between authorship in law and literature, suggested by Foucault and maintained by Rose, is strongly in need of q~alification.~~ In truth, it 20 113 L Ed 358; 111 S Ct 1282 (1991). 21 ‘What is important about Feist, for our purposes, is the gap that it discloses between the legal and the literary debate over the notion of the author.’ Price and Pollack, ‘The Author in Copyright: Notes for the Literary Critic’ in Woodmansee and Jaszi (eds), n 6 above, p 441. 22 Copyright, Designs and Patents Act 1988, Ch IV (UK); Visual Artists Rights Act 1990 (104 Stat 5128) (US). 23 Jaszi, ‘On the Author Effect: ContemporaryCopyright and Collective Creativity’ in Woodmanseeand Jaszi (eds), n 6 above, p 35. Kernan, 7he Death ofLirerature (New Haven: Yale University Press, 1990) p 950. Gaines, n 8 above, argues that the author is dying in copyright and that moral rights are ‘symptoms of the displacement of the author.’ 24 Saunders has many other goals. In particular, he argues that both literary historians and poststructuralist theorists have been unduly preoccupied by authorship and subject-centred accounts. zyxw For a more balanced summary, see Review (1993) Entertainment LR 59, and Vanden Bossche (1993) 36 Victorian Studies 487 (claiming that Saunders misrepresents other historical accounts). 25 Indeed, as the French literary historian and bibliographer Roger Chartier has pointed out, Foucault’s zyxwvutsrq historical account, while incomplete, was more sophisticated than is frequently suggested: ‘in no way does he [ie Foucault] postulate an exclusive and determinant connection between literary property and the author function’: see Sherman and Strowel (eds), n 1 1 above. 0 The Modern Law Review Limited 1994 977 f i e Modern Law Review zy zyxwvuts zy seems, the author-function has operated in different contexts at different times and in different ways, all of which have been layered on top of one another. The emergence of the proprietary author at the end of the eighteenth century may thus [Vol. 57 represent the growth of the powerful, modern, romantic conception of authorship, but the histories fail to establish conclusively any causative, necessary or determining link between the legal and the literary. All there is is a complimentary and reinforcing connection, a parallel development. The first important qualification of Rose’s history is to be found in the observation that concepts of authorship had long played some role (if not a critical one) within literary discourse. Masten, for example, has discovered that the increasing use of the ascription ‘anonymous’ around 1676 ‘begins to signal the author-ization of a text, the importance that someone, anyone, is sp‘eaking. The author’s emergence is marked by the notice of its absence’ (TCA, p 362). Furthermore, bibliographers have found that in the two centuries prior to the eighteenth century, it became increasingly common for books to contain the works of single rather than several authors. In this period, ancient texts were more frequently attributed to a single name than works in the vernacular where the privilege of being named was reserved for only a few great literary figures. Moreover, according to Foucault himself, prior to the eighteenth century, attribution of authorship had been orthodox in relation to scientific texts but not literary discourse, and the period merely saw a reversal of that orthodoxy. A second qualification lies in the fact that authorship also operated as a category within law prior to the ‘literary property debate’ of 1760- 1775. In the two zyx centuries before that, in England and elsewhere, it had been required that texts were attributed to authors as part of the penal control of literature for the purposes of censorship.26Foucault’s version of history recognised that the author-function in relation to ownership was preceded by the author as the subject of punishment. A further qualification of the Foucaultian thesis on authorship is that even ideas of literary proprietorship can be found earlier than the eighteenth century. As Lindenbaum’s discussion of ‘Milton’s Contract’ (TCA, pp 175 - 190) illustrates, writers were able to use their personal property rights in unpublished manuscripts to exact a fee from publishing stationers. More significantly, Feather’s essay, ‘From Rights in Copies to Copyright’ (TCA, pp 191-209), examines the concept of the author as proprietor of the text in the regulation of the book trade prior to the eighteenth century. Feather finds some evidence of recognition of an author’s right - first, in the grant of some printing patents to authors; and, secondly, through the development of practices of paying authors for their ‘copies.’ He concludes that, while it would be ‘perverse’ to claim that authors’ rights were widely recognised in pre-revolutionary England, ‘it would be more accurate . . . to suggest that they were dimly perceived’ (TCA, p 208). These qualifications of the assertion of a twin birth of copyright and authorship are important not because they suggest there is no relationship, but because they add an element of contingency and complexity to the history. The works suggest, ~ ~~ 26 Ross describes how authorship and publication were used as instruments of social control in the regulatory practices that were derived from royal proprietorship and directed at the threat of sedition: zyxwvuts ‘Authority and Authenticity: Scribbling Authors and the Genius of Print in Eighteenth Century England’ in Woodmansee and Jaszi (eds), n 6 above, p 242. Chartier confirms the view that the author was the fundamental mechanism for the designation of books and ‘an essential weapon in the struggle waged against the spread and distribution of texts which were thought to be heterodox’ in sixteenth- century France, but that liability was in no way greater than publisher, bookseller, hawker or owner: see Sherman and Strowel (eds), n 11 above, p 19. 978 0 The Modem Law Review Limited 1994 November 19941 zyxwvuts zyx Copyright and the Death of the Author at least, that it has been possible to conceive of authorship prior to copyright and the proprietary author prior to romanticism. Authorship in copyright is not, even in its historical foundations, simply equatable with authorship in literature and, therefore, a critique of literary authorship need not necessarily strike at the roots of zy copyright law. The prehistories of authorship and copyright make much less surprising the failure of copyright to automatically respond to developments in literary discourse. This increased sophistication, however, does not explain how those developments specifically based on romantic conceptions of authorship are capable of surviving the critique and there can be little doubt that, since 1800, cultural zyxwvu assumptions about authorship have informed the development of copyright law .27 The influence of romantic conceptions of the relationship between an author and his work can be seen to have operated in the actions of le islators and judges in extending the duration of the copyright owner’s monopolyja the narrowing of the fair use defence,29 as well as the extension of the copyright owner’s rights to cover derivative works (such as translations) and other sources of remuneration (such as performances).30 Furthermore, romantic conceptions of creativity have operated to define the domain of creative works and thus to limit the scope of subject matter protectable by copyright. Bernard Edelman, for example, has argued that the limitation of droit d’uuteur protection in France to ‘toutes les oeuvres de beaux arts’ posed problems for legal protection of photography in the mid-nineteenth century.3* It was only when economic pressures brought about reconceptualisation of the photographer as a creator and photography as art that French copyright law admitted photography to the sacred pantheon of copyright works.32In a similar vein, Marjut Salokannel has exposed the way in which it was 27 zyxw Streeter, ‘Broadcast Copyright and the Bureaucratization of Property’ in Woodmansee and Jaszi (eds), n 6 above, p 304. (‘The conceptual system of copyright relies heavily on this construct. Although the individuality of the author seems obscured by the commercial concerns of Anglo-American copyright law, the categories associated with this law, such as originality and the distinction between an idea and its expression, are. derived from the romantic image of authorship as an act of original creation whose uniqueness springs from and is defined in terms of the irreducible individuality of the writer.’) 28 1814 Copyright Act (extending the period to 28 years or life of the author, whichever was the longer) and Literary Property Act 1842 (42 years or the life of the author plus seven years). In fact, zyxwvut zyxwvu Wordsworth played a significantpart in supporting Sergeant Talfourd’s attempts to extend the duration zyxw of copyright which culminated in the 1842 Act. See Feather, A History of British Publishing (London: Routledge, 1988) p 171; Woodmansee and Jaszi, ‘Introduction’ in Woodmansee and Jaszi (eds), n 6 above, pp 4-5. 29 By the mid-nineteenth century the court no longer looked to see whether the defendant had produced a new work but looked at what he had taken: what a derivative user added was, by and large, irrelevant. Compare Sayre v Moore (1785) in Cary v Longman (1801) 1 Fast 358.35911; 102 ER 138, 13911;with West v Francis 5 B&Ald 737, 106 ER 1361; Bramwell v Halcomb (1836) 2 My & Cr 737, 40 ER 1110. 30 In the United States, the case of Duly v Palmer (1868) 6 Fed Cas 1132 (CCSDNY) - recognising a right to perform dramatic compositions under the 1856 Act (1 1 Stat 138) - has been called ‘the first great intellectual leap, auguring copyright’s break from the confines of copies and the eventual statutory expansion of derivative rights. ’ Goldstein, ‘Derivative Rights and Derivative Works in Copyright’ (1982) 30 J Copyright Soc’y USA 209, 213. Saunders would not deny that such factors were influential, n 14 above, pp 144- 145, 148. 31 Edelman, Ownership of the Image: Elementsfor a Marxist Theory of Law (trans Kingdom, London: Routledge, 1979); Tagg, m e Burden of Representation (London: Macmillan, 1988) ch 4. No such limitation of the domain of copyright has been made explicit in the United Kingdom, though commentaries reveal that an association between copyright and ‘literature and the fine arts’ has operated since the 1830s when copyright came to define itself as a domain distinct from patents and designs, and under the influence of international treaty negotiations adopted a form similar to continental copyright. For an illustration, see Talfourd’s attempted codification in 1837. 32 Edelman’s analysis of the history is linked with his account of how law and other ‘ideologies’ are necessarily connected. In contrast, if Saunders accepted that romanticism had played a role in these 0 The Modern Law Review Limited 1994 979 The Modern Law Review zy zyxwvuts zyx necessary to redefine cinema around a single individual author - the director as ‘auteur,’ so that the film could be seen as a creative product - as art - and be granted copyright p r ~ t e c t i o n . ~ ~ [Vol. 57 zyxw In Authorship and Copyright, Saunders blends historical review with the theoretical insights of systems theory to provide a more thoroughgoing explanation of why copyright law has proved immune to poststructuralist questioning of a u t h o r ~ h i p .Saunders ~~ observes from the histories of the development of copyright law in the United Kingdom, United States, France and Germany that law is ‘an independent and variable phenomenon of culture’ (AC, p 6), the product of a vast array of legal and cultural influences (AC, pp 11, 40, 94) which are not reducible to consciousness, economic, language, etc. Thus, he says that the ‘book describes the historical variability, complexity and technicality of law, legal systems and customary practices concerning literary and artistic property’ (AC, p 246, n 17). For example, when the legislature enacts a copyright law, many different influences are operating - some legal and some from outside law. Similarly, when a judge interprets the copyright law, he or she is likely to be much more concerned with legal coherence and continuity - with the presentation of the law as a logical whole - than with literary theory.35Given the conclusion that legal forms are a result of a complex interaction of legal and non-legal influences, Saunders argues that there is no necessary relationship between law and culture. That is not to say that cultural discourse never has an impact upon law,36but rather that where culture has influenced law, this influence has been coincidental. More specifically, the aesthetic persona has less directed copyright than overlapped with it (AC, p 212). In fact, where culture influences law it does so (and can only do so) in legal terms. This independence of the legal from the literary notion of authorship can easily be seen in the way in which the concept of authorship operates within Anglo- American copyright law. More specifically, the concept of authorship is sometimes present and sometimes absent within copyright discourse. According to Streeter, copyright demonstrates a ‘mixture of indifference and obsession’ with authorship (TCA, p 305).37While copyright may be built on an image of creative authorship, copyright law uses that image as a point of attachment - a point at which to ascribe a property right and by which that right can be determined. But the essence of that ascription is that it is a divestible or alienable right. In law, zyxwvutsr zy developments, it would not be because the law had to take account of cultural assumptions but because it chose to. 33 ‘Film Authorship in the Changing Audiovisual Environment’ in Sherman and Strowel (eds), n 11 above. 34 The use of systems theory is more implicit than explicit, though Saunders cites Luhmann’s 7he Diferenriarion of Sociery (trans Holmes and Lamore) (New York: Columbia University Press, 1982) n 14 above, p 6 . 35 This is illustrated by two essays in 7he Consirucrion ofAurhorship: Price and Pollack emphasise the significance of the analogy between patent law and copyright law in the development of copyright zyxwvut (Price and Pollack, n 22 above, p 443). while de Grazia emphasises the significance of literary proprietorship in the Supreme Court’s decision in the defamation case between the psychoanalyst Paul Masson and 7he New Yorker (n 15 above, p 289). Another significant influence on copyright’s zyxwvut development is the growing prominence afforded by the law to the idea of restitution and unjust enrichment. See Gordon, ‘On Owning Information: Intellectual Property and the Restitutionary Impulse’ (1992) 78 Villanova LR 153. 36 Saunders argues that law has become steadily ‘aestheticised’ (n 14 above, p 188) by way of ‘a re- orienting of certain areas of law towards the magnetic image and ideal of aesthetic personality’ (ibid pp 190, 210). 31 n 30 above, p 305. 980 0 The Modem Law Review Limited 1994 November 19941 zyxwvutsrq zyx Copyright and the Death of the Author authorship is a point of origination of a property right which, thereafter, like other property rights, will circulate in the market, ending up in the control of the person who can exploit it most p r ~ f i t a b l y Since . ~ ~ copyright serves paradoxically to vest authors with property only to enable them to divest that property, the author is a notion which needs only to be sustainable for an instant. This means that copyright law is able to imply and invent authors where there is zyxwvu no corresponding (cultural or other) ‘reality.’ Despite the argument that the zyxwvu Supreme Court decision in Feist is a recognition of romantic authorship, it also exemplifies the simultaneous independence of legal conceptions of authorship. For, despite the rhetoric of Justice O’Connor, the test of originality recognised was one of ‘minimal ~ r e a t i v i t y . ’There ~ ~ is no requirement that the work be of any artistic quality.@ As Pollack and Price note, few items are below this level of originality (TCA, p 455).41Similarly, in the United Kingdom, the critical stamp of authorship - originality - can be found in a verbatim report of a speech,42or the ‘automatic writings’ of a spiritual medium.43 More radically still, the Copyright Act 1988 recognises investments of capital and administrative organisation as constituting authorship of films and sound recordings.44The Act even ‘invents’ authors for computer generated works, where no human author exists. At the same time as the law can invent authors where romantic literary theory would deny them, law can deny authorship where literary theory might recognise it. Thus, copyright law denies authorship to the contributor of ideas45 and, in cases of collaborative works, frequently refuses to recognise contributors as authors in an attempt to simplify ownership.& Because a single property owner means that assignments and licences of copyright are easier and cheaper to effect, copyright law prefers to minimise the number of authorial contributions it is prepared to acknowledge rather than reflect the ‘realities’ of collaborative authorship. To simplify ownership in this way may privilege certain contributions over others, but it provides a property nexus around which contractual zyxwvuts 38 ‘It can be argued that copyright as a whole serves the interests of publishers and distributors more closely than it serves the interests of either authors or users of copyrighted works’ (n 30 above, p 306). 39 Ginsburg has argued that Anglo-American copyright law has always protected works of commercial value as well as works of creativity: see ‘Creation and Commercial Value: Copyright Protection of Works of Information’ (1990) 90 Columbia LR 1865. Indeed, she also demonstratesthat early French copyright law also protected such ‘works of sweat’: see Ginsburg, ‘A Tale of Two Copyrights: Literary Property in Revolutionary France and America’ (1990) Tulane LR 991, reprinted in Sherman and Strowel (eds), n 11 above, ch 7. 40 Here law’s instinct of self-preservation causes it to deny itself the power to discriminate between works of high and low quality, and instead to identify ‘original literary works’ objectively by determining whether the work was produced by the author rather than being wholly copied: see Price and Pollack, n 22 above, p 453 (‘trying to determine who is an author has the general tendency to implicate the aesthetic test, one that has been so strongly eschewed by American law’). 41 Price and Pollack, n 22 above, p 455 n 57. 42 Wulrer v Lane [ 19001 AC 539. 43 Cumins v Bond [1927] 1 Ch 167; Leah v Two WorM Publishing [1951] Ch 393. 44 Streeter describes how the ‘relatively authorless medium of television is constituted in part by a set of legal practices that nominally rest on a romantic notion of literary authorship’ (n 30 above, p 305). 45 Kenrick v Lawrence (1890) 25 QBD 99. 46 Wiseman v Wiedenfeld[ 19851 FSR 525. In the context of property law, this has been referred to as the agglomerative tendency: see Donahue, ‘The Future of the Concept of Property’ in Pennock and zyxwvutsrqponm Chapman (eds), NOMOS XXII: Property (New York: New York University Press, 1980) pp 28-68. Similar simplifying practices have also operated in the book trade: see Masten, n 6 above, p 364 (citing Bentley, The Profession of Dramatist in Shakespeare’s Time (Princeton, NJ: Princeton University Press, 1971) p 199). @ The Modern Law Review Limited 1994 98 1 The Modern Law Review zy zyxwvuts zy zyxw arrangements can be made recognising the value of those other contribution^.^^ This independence of copyright law from literature is less obvious in civil law [Vol. 57 zyxwv jurisprudence. In particular, the French law of ‘droitd’auteur’ appears to be much more closely aligned with literary conceptions of authorship than Anglo-American copyright law. In fact, Saunders sees the French law as a ‘limiting case’ against which to test his thesis that law and literature are independent systems with no necessary internal correspondence (AC, pp 75, 80, 194). However, Saunders chooses a number of examples where French law has granted protection where there is no ‘creative author’ - in particular to computer programmers - to demonstrate that the limits of ‘droitd ’auteur’ are not intrinsically linked with those of literary authorship (AC, pp 198-199). Where they have been so linked, Saunders argues, that is because French law has chosen to do so. In light of the observation that law is not reducible to culture, Saunders argues that the poststructuralist critique of authorship has no necessary implication for copyright law. Even if Barthes were ‘right’ and the author is dead, law does not have to accept this ‘truth.’ The goals and functions of copyright law are different from those of literary theory (AC, p 223), and just because Barthes says that the author is dead does not mean that the publishers suddenly stop administering their copyrights or paying authors’ royalties. To believe that it would have had such an effect was ‘a sign of naivety or aesthetic arrogance’ (AC, p 233). Barthes’ conclusions may come to influence copyright law, but if they do come to be incorporated into copyright, Saunders’ point is that they will be incorporated as legal principles, as ‘law’s truth.’ The fact that there is a gap between the legal concept of authorship and the understanding of authorship in literary circles simply does not matter (AC, p 223). Saunders’ observations concerning the distinctiveness of the literary and legal fields help us to avoid the real problems that would be faced if the poststructuralist critique had to be incorporated into or accommodated by law. In so far as Barthes’ claim is an extension of Saussurean linguistics - that the meaning of texts derives from a system of ‘signifiers’ and ‘signifieds’ rather than from the author - it is difficult to see exactly what this would mean for copyright law. This is because the relationship between copyright law and ‘meaning’ is extremely troublesome. In some ways copyright is not about meaning at all, so that the radical critique would fail to bite. Literary copyright is limited to the articular arrangement of words in P the text and colourable variations thereupon .4 The ‘idea-expression dichotomy’ 47 Furthermore, a variety of conceptions of authorship operate within copyright law, even if these different conceptions are frequently presented as unitary and coherent. For example, within existing UK law, a film director is treated as if he were an author for the purposes of ‘moral rights,’ but not for the purposes of deciding who is the ‘first owner’ of copyright in the film: ss 80( 1) and 9(2)(a) of the z Copyright, Designs and Patents Act 1988.It will be necessary to modify this position in the light of EU Directives requiring that the principal director be recognised as one of the authors of a film for the purposes of determining the duration of protection and ownership of rental and lending rights. As regards US law, see also Jaszi, n 24 above, p 49 n 69, arguing that copyright uses different conceptions of authorship in considering acquisition and infringement. 48 Although the definition of literary work includes texts which provide information and instruction, it confers protection also on those which merely provide pleasure: see Hollinrake v Truswell [I89413 Ch 420.However, in Erxon Corp v &on Insurance [1981] 2 All ER 495,[1981] 3 All ER 241,the Court zyx of Appeal denied copyright protection to a single invented word on the grounds that it had no meaning. zyxwvut The perceived meaning of a text plays some role in determining the boundaries of the property right. In deciding whether the appropriation is substantial, the courts will look at the ‘significance’ of what has been copied. Whether the quality part has been taken may then depend upon the ‘meaning’ of that part. In the United States, the question of whether a use of copyright work is ‘fair’ depends in part on whether it is ‘transformative,’ and a use is said to be ‘transformative’ if it changes the ‘meaning’ of what has been appropriated: see Campbell v Acuf-Rose Music Inc (1994)127 L Ed 2d 500, 515. 982 0 The Modem Law Review Limited 1994  zyx zyxwvutszy zyxwvutsr November 19941 ‘expression’ and leaves ideas freely available to the Copyright and the Death of the Author - ‘a tour de force of ideological mediation’ - confines the property in a work to In as much as the poststructuralist critique alerts us to the inevitability of intertextuality, that is of the penetration of one text by others, copyright law already acknowledges the needs and rights of others to draw on copyright work. The ‘idea-expression’ dichotomy operates to permit, excuse and sanction the reuse of ideas which inevitably seep from work to work,50 and defences of fair use and fair dealing allow the reproduction or appropriation of the text i t ~ e l f . ~ ’ D The Future of Authorship and Copyright Rather than viewing law as a reflection of literature, Saunders prefers to see copyright as constituted more by specific institutional practices established in particular technological environments. According to such an account, new technologies pose much more of a threat to the sustainability of copyright law than do insights from literary theory. These technologies operate both to create new subject matter needing protection and to provide new ways of replicating or distributing existing subject matter.52Each deals its own blow to the integrity of copyright. As a result of the acceptance of new subject matter, such as sound recording and films, it has been argued that copyright as ‘author’s rights’ is now dead (although the institution of copyright remains intact) .53 While recognition of protection for ‘entrepreneurial works’ has involved changes in copyright and these entrepreneurial works are now of greater economic significance than traditional authorial works, authorial works still constitute a significant (and sustainable) part of copyright law. A more serious challenge to copyright is felt to result from new modes of distribution - such as digitalisation, Internet and ‘information superhighways.’ These technologies change the ‘form’ of works, so that the boundaries of the properties can no longer be defined by anachronistic ideas of print and texts.54 These new technologies of distribution also threaten copyright because they make it easier to infringe and more difficult to police infringement. In effect, distribution of works is relocated from the public domain of the market place, where transactions are visible and easily regulated, to the private world of the home and zyxwvutsr 49 De Grazia, n 15 above, p 300 (citing Boyle, ‘A Theory of Law and Information: Copyright, Spleens, Blackmail and Insider Trading’ (1992) 80 California LR 1413). 50 Litman, ‘The Public Domain’ (1990) 39 Emory U 965. See more generally Yen, ‘The Interdisciplinary Future of Copyright Theory’ in Woodmansee and Jaszi (eds), n 6 above, pp 159- 173. zyxwvut 51 For a startling example, see Swan, ‘Touching Words: Helen Keller, Plagiarism, Authorship’ in Woodmansee and Jaszi (eds), n 6 above, pp 57- 100. 52 For example, the Fine Art Copyright Act 1862 granted protection to photographs and protection to zyxwvutsr existing copyright works from being reproduced by photographic means. Equally, digitalisation zyxwvutsr presents opportunities for new methods of appropriation and new works claiming protection: see eg Sanjek, “‘Don’t Have to DJ No More”: Sampling and the “Autonomous” Creator’ in Woodmansee and Jaszi (eds), n 6 above, pp 343-360. 53 Turkewitz, ‘Authors’ Rights are Dead’ (1990) 38 J Copyright Soc’y USA 41. 54 Such changes will require that the text be reconceived and that new ways of identifying the boundary between what is mine and what is yours be established. The reformulation of the ways in which works are identified, their boundaries ascertained and remunerations allocated are likely to rely increasingly on statistical approximations, while users are much more likely to be charged by reference to ‘time’ rather than numbers of pages. 0 The Modern Law Review Limited 1994 983  zy zyxwvuts zyxwv Ihe Modem Law Review [Vol. 57 office.55 However important all these threats are for the future of copyright, for Saunders they are interesting only in so far as they see crisis of copyright in technological changes, rather than changes in the literary notion of authorship. Although technology may require the development of new practices if copyright is to be sustained, it is technology - not the death of the author - that poses the chief threat to copyright. David Saunders’ observation that copyright law and literature are distinct domains is a useful antidote to those who would have us believe that copyright and literature mirror each other. What Saunders does not do, which he might have considered, is to go further and examine when, why and by what methods copyright law has come to adopt concepts drawn from cultural discourse and (in particular) how those concepts operate in law. For example, Saunders offers no zyx explanation as to why, suddenly, the Supreme Court of the United States has required that works display minimal creativity before they can be p r ~ t e c t e d . ~ ~ Saunders does, however, suggest that the ‘aesthetic persona as a positivity is now in the ascendant’ (AC, p 23) and that it is exerting an increasingly strong influence upon law. That is, though there is no necessary reason why copyright law must accommodate aesthetics, the increasing pervasiveness of the ideal of aesthetic personality is leading to a progressive aestheticisation of law (AC, pp 186- 2 11). Nevertheless, the argument that legal discourse is a result of many complex zy pressures (AC, p 212) provides us with no practical or theoretical insights into when the legal system will take notice of such cultural deve10pment.s.~’ As Woodmansee remarks, ‘the problem of how these two levels of discourse - the legal-economic and the aesthetic - interact is one that historians of criticism have barely explored. ’58 Saunders does little to develop this exploration. zyxwvutsrqp Saunders also fails to indicate how copyright law could, or indeed whether it should, accommodate the poststructuralist critique of authorship. In contrast, a 55 De Grazia, n 15 above, pp 301 - 302 (‘Photography, tapes, videos and xerography have blurred, if not dissolved, proprietary boundaries, allowing for the ready appropriation of materials . . . In the context of such technologies, the strict upholding of quotation marks might appear quaint and outmoded, an anxious gesture against an onrushing future’). In such environments, instrumentalist techniques of regulation tend to be both practically ineffective and politically incompatible with values such as privacy. More specifically, identifying use of a protected work for which a person would be liable will become as problematic as those faced in relation to, for example, reprography or home-taping. It seems likely that new modes of regulation and especially self-policing may prove necessary. Production costs could be recouped through the grant of blanket licences, one-off charges to those who enter works into a distribution system, coupled with possible extraction charges for users. It might well be that the technologies which provide new modes of distribution can also produce new techniques of identifying and monitoring uses of works, through, for example, electronic coded tags or coding of complete works or through auditing the computer’s memory. 56 Saunders explains the recent recognition of moral rights in the US merely by resort to obligations of international law imposed by Article 6 bis, Berne Convention to which the US recently became a signatory: n 14 above, pp 196-206. 210. Alternatively, moral rights might be justifiable without any need to accept ‘romantic authorship’: see Ginsburg, ‘Moral Rights in a Common Law System’ (1990) Entertainment LR 121. 57 In contrast with Saunders, Gaines’ Conrested Culture tries to provide some analysis of when, why and how law acknowledges literature. Using insights drawn from the work of Gramsci and Althusser, Gaines attempts to meet ‘the theoretical challenge of relative autonomy,’ namely, how to represent zyxwvuts political, social, economic, legal and cultural forms as connected and yet disconnected. Noting that ‘ideology works through us, often with our own enthusiastic cooperation,’ Gaines argues that connections between law and culture can be found explicitly where law refers to custom, and implicitly where law is ‘mixed with pithy sayings, homely analogies, personal judgments and frank characterisations,’ that is, ‘common sense’ drawn directly out of shared knowledge in the culture: see n 8 above. 58 Woodmansee (1984) n 1 1 above, p 440. 984 (3 The Modem Law Review Limited 1994  zyxwvuts November 19941 zy zyxwvutsrq zyxw Copyright and the Death of the Author number of essays in The Construction of Authorship appear to advocate that copyright law should be more sceptical about the role of authorship. These normative claims are based on the perception that the preoccupation of copyright zy with a romantic conception of authorship is unsatisfactory. It is unsatisfactory, first, because it does not correspond to or accord with ‘reality.’59While a number of the essays in The Construction of Authorship explore collaborative writings in history,60others suggest that collaboration is still a very common form of writing practice. For example, Andrea Lunsford and Lisa Ede, in ‘Collaborative Authorship and the Teaching of Writing’ (TCA, p 418), found that ‘much or most of the writing produced in professional settings in America is done collaboratively , and that, in fact, much of what we call creative writing is collaborative as well, though it almost always flies under the banner of single authorship.’ However, according to Jaszi and Woodmansee, copyright law - based on romantic authorship - presumes a solitary author. For Jaszi and Woodmansee, copyright law should ‘correspond’ with ‘the realities of contemporary polyvocal writing practice - which increasingly is collective, corporate and collaborative’ (TCA, p 38), rather than shoehorn different writing practices into a single inflexible legal conception of authorship. In ignoring these realities, copyright law is not only dishonest but marginalises or denies these practices. For Jaszi and Woodmansee, however, a further reason as to why copyright law should pay more attention to literary discourse is that romantic preoccupations operate to exclude many deserving works from protection.6’ For example, ideas of individual creativity result in denials of ‘protection to folklore and items of cultural heritage that are valued chiefly for their fidelity to tradition rather than their deviations from it’ (TCA, p 1 1).62Furthermore, the requirement of fixation denies protection to improvised works and works of oral tradition.63 Another reason to ‘reestablish communication between the two disciplines’ (TCA, p 28) is that technological developments will make the romantic conception of authorship 59 Jaszi, n 24 above, p 50. 60 Masten, n 6 above; Thomas, ‘Reading and Writing the Renaissance Commonplace Book’ in Woodmansee and Jaszi (eds), n 6 above, pp 401 -415; Gere, ‘Common Properties of Pleasure: Texts in Nineteenth Century Women’s Clubs’ in Woodmansee and Jaszi (eds), ibid pp 383-399. 61 A further reason why copyright law might helpfully reconsider ideas such as ‘authorship,’ ‘originality’ and the ‘idea-expression dichotomy’ is so that it can better accommodate postmodern artistic practices which deliberately set out to challenge notions such as authorship and originality by appropriating from others in the construction of their works. See Carlin, ‘Culture Vulture: Artistic Appropriation and Intellectual Property Law’ (1988) 13 Columbia-VLA J Law and the Arts 103, and criticism of the ‘unexamined assumption that copyright law must adapt itself to new modes of authorship’ by Saunders, n 14 above, pp 227-229. Another motive for dropping the author from copyright may lie in the very fact that Barthes’ claims to the demise of authorship do not seem to have had the impact which might be thought desirable. Romantic ideas of authorship remain pervasive in all fields of the social sciences and it sometimes feels as if we will never be able to escape the tyranny that this exerts over scholarship. This persistence may be strengthened and reinforced by copyright law and copyright rhetoric. 62 Jaszi and Woodmansee, ‘Introduction’ in Woodmansee and Jaszi ( 4 s ) . n 6 above, p 11. Sherman’s zy essay, ‘From the Non-Original to the Aboriginal: A History’ in Sherman and Strowel, n 11 above, ch 6, alerts us to the way in which copyright’s concept of ‘originality’ has been used as a political tool to deny copyright recognition to aboriginal art. Aboriginal art, rather than being treated as created, was by and large treated as ancient and ‘ab-original’ - always existing. The works were treated as zyxwv ethnographic museum pieces, not art gallery exhibits. Further, the art works were conceived not as original but as the product of tradition. 63 The requirement that a work be recorded is, however, more common in copyright systems than droit zyxwvutsrqp d’aureur regimes. It is difficult to see, then, why or how it is a consequence of romantic conceptions of authorship: see Gendreau, ‘The Criteria of Fixation in Copyright Law’ (1994) 159 Revue Internationale dc Droit d ’Auteur 1 10. 0 The Modern Law Review Limited 1994 985 The Modern Law Review zy zyxwvuts zy an even less appropriate model than it is now. In particular, the worldwide linking of computer terminals, known commonly as Internet, offers ever greater opportunities for interactive and collaborative writing. ‘Copyright’s recursive [Vol. 57 insistence on forcing all writing into the Procrustean doctrinal model, shaped by zyxwvu the individualistic, Romantic concept of authorship’ will have, Jaszi argues, ‘real, adverse consequences’ for electronic technology (TCA, p 55). While a number of the essays in 7he Construction ofAuthorship suggest that it would be desirable to remove romantic conceptions of authorship from copyright law, none of the essays suggest what the legal landscape would look like without authors .64 In fact, the criticism that copyright’s emphasis on ‘solitary authorship’ ignores collaborative writing practices appears not to be motivated by a desire to zyxwvut abandon ‘authorship’ completely but merely by a desire that copyright employ a different conception of authorship. As such, it uses Foucault’s observations that the author is constructed, but does not go so far as to advocate abandoning totally the concept of authorship. Instead, what is advocated is a more pluralistic concept of literary production which can accommodate a wide variety of writing practices. These conclusions leave one with a sense of disappointment. The audacious beginning - the use of the poststructuralist critique as a point of inspiration from which to rethink copyright law - results merely in an appeal for a more sophisticated legal acceptance of joint authorship. A more radical alternative would be to recast copyright law in materialist terms, recognising authorship merely as the investment of labour power and entitling the contributor not to ‘proprietorship’ but to rem~neration.~~ Couched in such terms, a copyright system might, as Woodmansee and Jaszi wish, be more accommodating of collaborative contributions. Aided by bureaucracy, statistics and technology, such contributions of labour power may be more accurately defined within structures which could transform author’s rights into remunerative rather than property rights. 64 A small cadre of liberal economists would see a world without authors as a world without copyright law in which the market could control the circulation of texts and readers their meanings. However, even in a world without authors, some kind of incentive to produce and disseminate texts, some kind of system of attribution and identification, some kind of ‘order,’ might well be thought desirable. 65 Frow has attempted to recast copyright in such materialist rather than romantic terms. He sees copyright law as concerned with the investment of labour and copyright as a recognition of labour expended by, amongst others, ‘writers.’ Such a reconceptualisation would not necessarily involve radical revision of thc law, since authorship is, in the United Kingdom copyright law at any rate, zyxwvutsr defined only in terms of origin. Frow notes that, despite the romantic implications, copyright’s concept of originality involves only a causal relationship between an author and the material form in which the work is embodied. This, he suggests, can be recharacterised as an investment of labour power. By so doing, he claims that ‘the ideology of free creativity could be displaced, but in terms that are derived internally (if critically) from the existing structure of copyright law’: Frow, ‘Repetition and Limitation: Computer Software and Copyright Law’ (1988) 29 Screen 4. 986 6) The Modern Law Review Limited 1994