The zombies wander the land; the lone survivor tries desperately to find support and resources. This might describe any number of films, but it also acts as a metaphor for the contemporary editor or biographer dealing with the undead of literary estates. The creative artist is long deceased, and what lives on is the administration of the literary estate, mechanistic in its movements and predatory in its actions. The literary estate uses the letter of copyright to disadvantage the scholarly work of editors and biographers in ways that the spirit of copyright does not justify.

This situation particularly affects editors working on the texts of authors who have died since 1938. This period covers most of the Modernist canon and after - key authors such as T.S. Eliot, James Joyce and Virginia Woolf. The issue has also become more acute over the past 20 years as editors try to place as much material as possible (increasingly on screen rather than on paper) before the reader to explain and justify decisions and preferences. The ability to provide and link published and unpublished material on a website has created new opportunities to demonstrate the evolution of an author's thought and writings. But instead of welcoming the opportunities to stimulate interest in an author's work and thereby to increase the revenues derived from it, literary estates can block them out of an kneejerk conservatism towards an unfamiliar medium and a fear of loss of control of the heritage they are protecting.

Our academic perspective privileges the generation and transmission of new ideas and learning conducted in an atmosphere of liberty of action and freedom of expression. The viewpoint of the literary estates (and their publishers) is based on a desire to control use of copyright material and exploit it to the full commercially. Both positions are reflected in copyright law itself; yet only that of ownership is stressed and promoted.

Copyright (see box) allows authors to derive an income from their published creative work in their lifetimes and for a couple of generations afterwards. Their contracts with publishers often last for the duration of copyright, giving the publishers a monopolistic revenue from successful works over the same period. Yet copyright legislation includes a "fair dealing" (UK) or broader "fair use" (US) exemption. This allows us to quote in reviews, to analyse texts in articles, to provide gobbets in student textbooks by way of illustration of the points we make as educators - as long as criticism outweighs quotation, as the publishers of Coles Notes study guides discovered in a key case, Sillitoe v McGraw Hill Book Co (UK) (1983). Copyright, in other words, is not an absolute right but is generally qualified by a sense of what might be in the best interests of society, particularly of education and cultural development. Indeed, the notion that it is of fixed term - compared with other forms of ownership - and that the property then passes into the public domain represents the major index of how qualified a right copyright has been. Intellectual property must at some point become available to all so that it can work to the public good. This is not a legal principle that must be fought for but one that must be exploited without reservation.

É«ºÐÖ±²¥

Is there any difference between published and unpublished material? Not in principle, but the copyright period for unpublished material generally starts on its first publication rather than on the death of its author. And it is most often unpublished material in the form of manuscripts that informs the work of editors. In theory, this could be closed off indefinitely. Anything written by T.S. Eliot, for example, even though it is owned by a public body such as a university library or a museum where it can be consulted, could not be quoted without permission of the Eliot estate, which is controlled by his widow. To judge from evidence to date, such permission would be refused - although the path to approval might be eased if any subsequent publication were to be issued by Faber, Eliot's long-term publisher.

On the other hand, the Irish Dail in 2004 passed an amendment to its copyright legislation to allow the copying of manuscript notebooks bought by the National Library of Ireland for electronic display in a major Joyce exhibition. The amendment was introduced in the Senate by David Norris, a noted Joyce scholar, when it became clear that the estate, controlled by Stephen Joyce (grandson of James) was unlikely to grant permission. This demonstrated again that copyright was a pliable concept in its implementation, balancing the commercial interests of author (and estate) and the public good. The Dail agreed that the control conferred by copyright should not be used to prevent the public good. The notebooks became the digitised centrepiece of the exhibition; the British Library's Turning the Pages technology allowed visitors to Dublin to read them without causing wear and tear to the originals. All that was lacking was the ability of the Joyce scholar to quote from them.

É«ºÐÖ±²¥

The UK Parliament also amended copyright legislation when it granted perpetual copyright in Peter Pan to ensure continuing income to the Great Ormond Street Hospital for Children. The issue for editors and biographers, however, is not specifically the duration of copyright but the right to quote in a scholarly context. Duration is anyway constantly being extended to serve commercial interests; Congress did this for Mickey Mouse in the US, and Parliament, at the behest of the European Commission, may do it for Cliff Richard, and sound recordings by other artists, in the UK. Any fight over duration of literary copyright cannot be easily won against these commercial forces - although many are fighting rearguard actions.

Rufus Pollock of the University of Cambridge has compiled an independent balance of the benefits arising to society from quicker access against the cost to individuals of losing their rights. He has worked out that the optimal length of copyright is about 14 years. But this is an average. The case of works such as Joyce's Ulysses having a clear and continuing commercial return confounds this figure. It will still be profitable at the end of its current period of copyright in 2011 - 89 years after first publication in book form.

This novel has indeed been previously the object of attempts to extend its term of copyright. The 1984 Hans Gabler edition commissioned by the Joyce estate in collaboration with publishers such as Garland and Penguin anticipated the impending expiry of copyright in 1991. This work represented the coincidence of editorial and commercial (estate and publishers) interests - on the one hand, to create a more accurate text of Ulysses likely to differ greatly from the very corrupt original; and, on the other, to extend an exclusive revenue from the novel until at least 2034. Although other - known to be flawed - editions would in the course of events enter the public domain, this, the "definitive" edition, would continue to be the intellectual property of estate and publishers.

Cambridge University Press had set a slightly earlier precedent in its reissuing of the edited works of D.H. Lawrence as new works using unpublished material and thereby attracting a fresh copyright. Indeed, the tactic had been anticipated by Mark Twain in 1906. Twain's works, under the US legislation at the time, were in copyright for only 42 years after first publication. He wished to ensure continuing income from them for himself and, on his death, his family. He proposed to add chunks of his unpublished autobiography to books such as Tom Sawyer and Huckleberry Finn. This added value would provide not only new copyrights but also the ability to compete commercially with the out-of-copyright original works. Which of us would recommend to our students a cheap, older edition of a Shakespeare play if a more "accurate" up-to-date but more expensive edition were available?

The skirmishes, glorified as the "Joyce Wars", over just how definitive was the Gabler 1984 edition of Ulysses obscured the wider commercial intent of estate and publishers (Gabler being innocent in this respect). On the other side of the clashes from Gabler (and, in one minor engagement in The New York Review of Books, the author of this article) was John Kidd. His own proposed, even more "definitive", edition of Ulysses to be published by Norton never appeared. This may have been due, at least in part, to the refusal of the Joyce estate to permit publication of unpublished and published material to be used in Kidd's new editorial construct.

What Garland, Penguin and the Joyce estate did not anticipate was the extension of copyright. In 1995, the copyright period in the UK became, as the result of European Union harmonisation, the author's life plus 70 years as opposed to life plus 50. This brought Joyce's other works and editions back into copyright until 2011. The UK mitigated the commercial effects of this for publishers who had in good faith produced works that had been out of copyright and were then brought back into it through implementation of the EU directive. The UK Government introduced the principle of a compulsory licence to apply to such works, as a service to the publishing industry rather than to editors and academics. This allowed publishers to continue to produce "re-copyrighted" works while recognising that ownership lay with the estates. Publishers would pay the estates for the right to produce such works. Arbitration of what constituted a reasonable payment would be made in case of disagreement by a copyright tribunal. This has allowed, among other examples, the continuing availability to the public of the Oxford University Press World's Classics reprint of the 1922 Shakespeare and Company first edition of Ulysses.

This safeguarding of commercial interests became an accidental reassertion of public rights when Danis Rose decided to create Ulysses: A Reader's Edition. This was published by Lilliput in Ireland and Macmillan in the UK in 1997. Whatever doubt there may have been about his editorial methods, Rose was exercising a form of scholarly freedom to produce a new version of Ulysses by using both published and unpublished material controlled by, even if not in the possession of, the Joyce estate. However, that brought him into conflict with the commercial interests of the estate and those publishers licensed by it to produce the novel and to profit from its consistent sales, particularly in the academy, year on year. The Joyce estate took Rose and his publishers to court.

Judgment was passed in 2001. It was mixed. The same system of compulsory licence that had allowed the continuing sale of the World's Classics reprint permitted Rose and his publishers to use material from editions of the novel published during Joyce's lifetime with reasonable financial recompense to the estate. However, any use without the estate's permission of manuscript materials, published after Joyce's death or unpublished, was an infringement of its copyright. Rose then published the new Reader's Edition containing only the allowable material. However, this can only have been a compromise of his original editorial principles and entailed in his introduction a rationalisation of what were in effect editorial decisions imposed by a legal judgment.

É«ºÐÖ±²¥

The lack of a compulsory licence and copyright tribunal system in Ireland had been highlighted in the Joyce estate's behaviour in respect of Cork University Press and the proposed inclusion of Joyce extracts in David Pierce's Irish Writing in the Twentieth Century: A Reader (2000), aimed at students and the rare general reader. The estate demanded a large fee of ?7,000, later rising to ?7,500, for text that would have made up no more than 2 per cent of the total book. When negotiations broke down, Cork University Press decided to use the Rose edition that was then being contested in the UK courts. The estate won an injunction in the High Court of Ireland to prevent publication. The book did go on sale, but with pages 323 to 346 excised. Cork University Press was threatened with closure, and its publisher resigned. Under the UK system, this would not have happened because the original fee would have been referred to the copyright tribunal for adjudication of what constituted a reasonable amount.

What if the principle of compulsory licence and fee arbitration were applied more widely? This would represent a new form of "fair use" or "fair dealing" in which materials, including those as yet unpublished, would be made available but if used commercially would result in revenue to the estate. This would result in more competition in the market and the generation of new ideas and new perspectives on the past. Both these outcomes are part of a greater public good and not of a narrow academic ambition. It would take not only changes to legislation in the EU but also to international copyright treaties. It would be worth the battle. But it is the editors and scholars who will have to lobby initially to see our rights protected and extended in this area. If we pursue and win this struggle, we can deal with the dead, and the undead, equitably and reasonably.

Property rights and wrongs: the nature of copyright

É«ºÐÖ±²¥

Copyright was put on a statutory footing in the UK in the 18th century as a means of providing a living for those who worked by their brain as opposed to their hands. It was regarded as a form of "intangible property", albeit of limited term. It covered expression of ideas but not the ideas themselves.

Adam Smith argued that creative artists (and lawyers) were worthy of a higher return than other workers because of the longer investment in their training.

The later European view of copyright regarded a published work as the author's offspring as much as his property, endowing him with inalienable moral as well as tradeable commercial rights.

The Anglo-American tradition in copyright, which is based firmly in the notion of property and income, resisted this concept. In the UK, only two general moral rights, of identity and integrity, crept in with the 1995 European Union harmonisation legislation; other rights, more restrictive about how the material could be used, were rejected. In the US there are no moral rights, only property rights, for authors.

The biography of Lucia Joyce written by Carol Loeb Shloss was refused the use of quotations and documents that threw light on the relationship between James Joyce and his daughter, not as an act of prurience but in an attempt to understand more fully the author's works. The Joyce estate defended its right to "protect" the Joyce family against intrusions into past, private lives through the use of copyright.

However, copyright legislation disallows only quotation of material, not the transmission of ideas through paraphrases of the material. Thus, what the biography lacked was the quoted evidence to substantiate its paraphrased assertions, something that reviewers were not slow to point out on its first publication.

The use of copyright like this to protect privacy, rather than to benefit commercially from property, is in US terms a "misuse of copyright", an argument that Shloss pursued in her successful case against the Joyce estate in the US in 2007. In this case, too, copyright was seen to be a qualified right, not an absolute one: it exists to reward creative endeavour but also to create further opportunities for innovation in the public good, such as new editions or new biographies.

Derivative works: fair use or infringement?

Derivative works represent a newly contentious area, particularly in defining the level of innovation or fresh creation. Warner Bros, Scholastic and J.K. Rowling are pursuing a case in New York against the publisher of a proposed Harry Potter Lexicon.

The Lexicon is based on an extant fan website that Rowling welcomed because, although it comprised lists of her spells and curses and other material drawn from her Harry Potter series of novels, it sought no commercial gain and may even have played a part in stimulating interest in her books.

However, once that same material - copied but rearranged and in places rewritten - was to be packaged and sold as a book, not only in the US but in other countries where the publisher had sold translation or reprint rights, the resulting work constituted, as most observers read the case, not fair use - where's the added creativity or innovation? - but rather an infringement of copyright and a denial of income from her creation that Rowling was entitled to.

There are other companions and encyclopaedias, for example to the work of Philip Pullman and Terry Pratchett, but these tend to be produced by the original publishers, or even co-authored by the original author, and therefore do not inhibit the commercial exploitation of the original copyright source by its creator.

Other derivative works that have been challenged using copyright have tended to be much more creative responses to existing works rather than so nakedly parasitical.

Lo's Diary comprises a retelling of Vladimir Nabokov's Lolita from the point of view of the girl, not the man. In this case, an out-of-court settlement was reached with the Nabokov estate, as also happened in the case of The Wind Done Gone, published by Houghton Mifflin, and Margaret Mitchell's estate, which was defending the interests of Gone with the Wind.

É«ºÐÖ±²¥

These settlements underline the commercial nature of copyright and the interests of estates. Many commentators believe that the well-publicised hesitations of Dimitri Nabokov to destroy, according to his father's wishes, the unfinished, unpublished work The Original of Laura are no more than an attempt to create interest in the book ahead of its eventual publication.