The Google Book Settlement and European Authors

Note: this page is now partly out of date. If you are looking for information on the latest version of the settlement you should go to this page:

The Google Book Settlement: a survival aid for UK authors


The proposed Google Book Settlement represents an attempt to use the machinery of a private settlement in a civil law case to overturn fundamental principles of national and international copyright law in the interests of Google Inc., a wealthy corporation. It seeks to legitimate a massive appropriation of rights in American copyrights from authors worldwide. It aims to bind authors and their heirs and assigns in perpetuity, in many cases without their express agreement, to a non-negotiable contract of near-stupefying complexity containing numerous provisions detrimental to their rights. It intrudes into existing contracts (including contracts signed outside the US), in many cases assigning to the publisher rights that the author has never licensed.

Article 17.2 of the Charter of Fundamental Rights of the European Union states: 'Intellectual property shall be protected.'

It is my trust that the governing bodies of the EU and the government of the UK, together with other European governments, will act to uphold the rights of British and European authors and copyright-holders under international law, sustain them in the unhindered enjoyment and exercise of their copyrights, and protect their property in those copyrights from harm and the destruction of value.

Gillian Spraggs

Continue reading this paper online

or download it:

Portable document format (pdf) | Word document (doc)

For more about the Google Book Settlement see my online journal.

Some other responses to the Google Book Settlement


A. Summary

The proposed Google Book Settlement represents an attempt to use the machinery of a private settlement in a civil law case to overturn fundamental principles of national and international copyright law in the interests of Google Inc., a wealthy corporation. It seeks to legitimate a massive appropriation of rights in American copyrights from authors worldwide. It aims to bind authors and their heirs and assigns in perpetuity, in many cases without their express agreement, to a non-negotiable contract of near-stupefying complexity containing numerous provisions detrimental to their rights. It intrudes into existing contracts (including contracts signed outside the US), in many cases assigning to the publisher rights that the author has never licensed.

It is my hope that the government of the UK, in conjunction with the EU, will act to uphold the rights of British and European authors and copyright-holders under international law, sustain them in the unhindered enjoyment and exercise of their copyrights, and protect their property in those copyrights from harm and the destruction of value.

The history of the Google Book Settlement may be summed up as follows. In December 2004, Google began scanning very large numbers of in-copyright books obtained through a partnership arrangement with a number of institutional libraries. No permission was sought for this from the copyright-holders. The digitised books have been indexed for use in Google's Book Search service. In 2005 the Authors Guild and the Association of American Publishers began separate legal actions against Google alleging copyright violations. The parties arrived at a settlement in October 2008. Under the terms of the agreement, which has not yet been ratified by the court, Google would be permitted not only to continue scanning at will all books published anywhere in the world on or before 5 January 2009, but also to exploit the scans commercially in the United States, by methods that include posting advertisements alongside page views of books, and selling institutional and individual access to its corpus of digitised works. Other projected uses include selling e-books and print-on-demand copies. The case was brought as a class action. The settlement class is specified as 'all Persons that, as of the Notice Commencement Date, have a Copyright Interest in one or more Books or Inserts': this translates as virtually every author, literary estate, publisher or other rights-holder anywhere in the world. If the settlement agreement is accepted by the court, it will become legally binding under US law on all these persons. Only rights-holders who formally opt out of the settlement by 4 September 2009 will be exempt.

The settlement agreement is set out in a document of 141 pages. Another 212 pages are appended in 17 attachments. If it is accepted by the court, it will operate as, in effect, an immensely complicated, multi-clause contract, lasting for the duration of the copyright in each work. In many cases it will allocate to publishers income from the exercise of rights (electronic rights, American rights) that have not been licensed to them by the authors. It will also establish a new institution to be called the Book Rights Registry. All authors who remain opted-in to the settlement, whether by design or default, will have to register their works with the Registry. Otherwise they will unable to control, to the extent permitted by the settlement, the uses that Google makes of their works; nor will they receive any payment. This provision runs counter to a core principle of the Berne Convention, which states that the rights enjoyed by copyright-holders 'shall not be subject to any formality'. The revenue earned by the works of rights-holders who fail to register with the Book Rights Registry will be applied in the first instance to paying Registry's operating expenses. If any is left, much, and perhaps all, of this money will be distributed among the rights-holders who have registered. (There is provision for allocating a residual element to charitable causes.) The settlement has many other disturbing features, some of which are described in this paper.

Return to top

B. The Purpose of this Paper

This paper is intended to raise a number of serious concerns about the Google Book Settlement. It is aimed primarily at UK and European authors and legislators. About three months ago I began to be aware of the controversies which the settlement had stirred up among US copyright lawyers and other commentators, and the anxieties and anger being expressed by a large number of authors, small publishers and literary agents. I looked for a satisfactory account of the implications of the settlement for UK and European authors, and could not find one. I therefore began to investigate the matter for myself. I have read the settlement agreement and the attached documents (some parts many times), and studied the extensive commentary available on the web. This brought me to the conclusion (in which I am far from alone) that the agreement will be highly damaging to the interests of authors everywhere, and perhaps especially to non-US authors. In addition, it probably will not operate to the benefit of most of the smaller publishers (and may not do much, if anything, to benefit large publishers, either). I became increasingly disturbed, and convinced that I should write up my findings for circulation. The length and complexity of the settlement and the need to be as sure of my ground as possible, is one reason why it has taken me so long to produce this paper. Moreover, throughout July, and even over the past few days, important interviews and opinions have continued to be published on the web, answering some of my questions, and raising further issues that I needed to explore. This project has taken me more time than I ever dreamed when I started it. But so far as the Google Book Settlement is concerned, the point of no return is still a short distance ahead. There is time (but not much time) to take action.

It should be noted that I am a literary critic and cultural historian, not a lawyer. This paper is not written from a legal point of view and should not be mistaken for legal advice. Moreover, it is not, and cannot be, a substitute for reading the agreement itself. The full set of documents may be accessed on the Google Book Settlement website.

Return to top

C. History and Background

Google Book Search

Google Book Search allows users to search the full text of a large and still growing corpus of digitised books. The books in the corpus come from two sources: the Partner Program, under which publishers and authors submit their books for inclusion in the service as a form of promotion, and the Google Library Project, under which the books are supplied for scanning and digitisation by a number of partner libraries.

Some of these libraries, notably the Bodleian Library at Oxford and Harvard University Library, have only permitted Google Inc. to digitise books from their collections that are out of copyright and in the public domain, but a number of US institutional libraries have provided for scanning very large numbers of books that are still in copyright. No attempt has been made hitherto to seek permission for this from the owners of the rights in these books.

At present, Google Book Search allows its users full view and downloads of the digital files of many (not all) books that are in the public domain, 'limited preview', or in some cases full view, of books digitised under the Partner Programme by arrangement with the publishers, and 'snippet views' of fragments of text from many (not all) of the in-copyright works that have been digitised under the Library Project. Some books have been digitised under both projects and turn up through the Book Search in both views.

A confirmation that works digitised under the Library Project may be identified by the fact that they are displayed in snippet view was given by Alexander Macgillivray, head of the legal team who negotiated the settlement for Google Inc., in a lecture at Harvard on 21 July.

Google Inc. has always argued that its digitising of in-copyright works and provision of 'snippet views' of them through Google Book Search constitutes 'fair use' under US copyright law. It continues to hold to that view.

The Law Suits

The Library Project was launched in December 2004. In June 2005 the Association of American Publishers (AAP) called on Google Inc. to stop scanning works copyrighted to its members for six months, pending discussions.

In September 2005 the Authors Guild (a professional organisation representing about 8 thousand US authors) began litigation against Google Inc. alleging that the unauthorised reproduction of works not in the public domain constituted 'massive copyright infringement' and noting that the company had announced plans to display digitised works on its website alongside paid advertisements.

In October 2005 five major US publishers, co-ordinated by the AAP, began litigation against Google Inc. alleging that the Google Library Project was a commercial project and that the company was ignoring the rights of copyright holders 'in favor of its] own economic self interest'. They stated that though Google was aware that these publishers wished them to seek permission before including their copyrighted books in the Google Library Project, the company, claiming 'fair use', maintained that it did not need to do this. The publishers asserted that they had been forced to bring the action 'to protect and prevent ongoing and imminent harm to the copyrights in their books'.

In October 2008 AAP, the Authors Guild and Google Inc. announced a legal settlement, the 'Google Book Settlement'. As briefly described above (see A. Summary), the settlement agreement is long and highly complex, and seeks to bind by default nearly every author, publisher, or other copyright-holder anywhere in the world.

It received preliminary approval by the court in November, but since that time opposition to the deal has been growing. The original opt-out date was 5 May 2009, with a Final Fairness Hearing by the court scheduled for 11 June, but following representations by a number of authors and authors' estates, a four-month extension was granted by the judge. The new opt-out date is 4 September 2009, with a Final Fairness Hearing on 7 October.

The Scope of Google's Digitisation Project

At the end of October 2008, Google announced that it had then scanned and digitised about 7 million books, and that it considered that it was 'just getting started'. In 2004 the stated target was 15 million books, but since that time it has been revised upwards.

A news report on 3 November cites an official source at Google for the statement that more than a million of the books digitised had come from publisher partners, with the rest coming from libraries. Of the books digitised from libraries, over a million are public domain, and between 4 and 5 million are works in copyright, most of them described as 'out of print'.

Since that point Google has continued to scan and digitise books in the collections of its partner libraries. On 21 July 2009 Alexander Macgillivray, head of the legal team who negotiated the settlement for Google Inc., announced for Google, 'We have scanned over 10 million books … more than 1.5 million are in the public domain; more than 1.5 million are in the partner program … and we are continuing to scan at pace'.

A proportion of these books are in the public domain, but the great majority are still in copyright. A large number - how many, it is not presently clear - are by authors who are not US nationals. In this connection it may be noted that Google is touting greatly increased access to books in foreign languages as one of the anticipated benefits of the settlement for users of Google Book Search in the US.

Some, probably a great many, of the books that Google has digitised have never been published or distributed in the United States. However, Google has been able to digitise them from copies obtained by US institutional libraries.

Return to top

D. Why the Google Book Settlement Affects European Authors and Rights-Holders

1. Under US class action law, the Google Book Settlement, if accepted as fair by the court, binds all members of the class on whose behalf it was brought. In this instance, the settlement class is defined as 'all Persons that, as of the Notice Commencement Date, have a Copyright Interest in one or more Books or Inserts'. The Notice Commencement Date was set by the court as 5 January 2009.

2. In the publicity given to the settlement in the press and on the web the phrase 'US copyright interest' has been widely used in describing the scope of the settlement. This initially led many non-US authors and rights-holders, including myself, to assume that the settlement would only affect their rights in works published in the United States. However, this is not the case.

If this settlement is allowed to go through it will affect the rights in all books published on or before 5 January 2009 in any country that is a signatory to the Berne Convention.

See Berne Convention 5.1: 'Authors shall enjoy, in respect of works for which they are protected under this Convention, in countries of the Union other than the country of origin, the rights which their respective laws do now or may hereafter grant to their nationals, as well as the rights specially granted by this Convention.']

Return to top

E. Options for Authors and Rights-Holders

Note again the caveat above: I am not a lawyer (see B. The Purpose of this Paper). Nothing in this document should be taken as a substitute for qualified legal advice. I am simply describing the settlement agreement so far as I have come to understand it.

It is being widely stated by authors' organisations and other commentators that authors and rights-holders have the following choices:

a) to opt out of the settlement altogether

b) to opt in to the settlement and raise any objections or concerns they might have with the court

c) to opt in, and register their work with a new institution, to be called the Book Rights Registry

d) to wait until after the Final Fairness Hearing by the court, and if the settlement is ratified, register their work with the new Book Rights Registry

e) to do nothing.

Taking this last option first; if they continue to do nothing once the settlement has been ratified (assuming that it is), they will nonetheless find themselves opted-in by default and bound by its terms. The settlement agreement gives Google unprecedented rights to make use of the work of any rights holder who does not specifically opt out from the settlement. This agreement will last for the duration of their copyright(s). Rights-holders who remain inactive will not be able to sue Google Inc. for any use it may make of their work. They will be unable to make any claims for payment from the company unless and until they register their work with the Book Rights Registry.

This is counter to the principles of the US Copyright Act, and counter to the principles of international agreements on copyright and intellectual property, and counter to all custom and practice in the publishing industry, whereby each publication is the subject of individual contractual arrangements between author and publisher. Nonetheless, leading US lawyers tell us that it is the case.

Rights-holders who opt in to the settlement in order to object to it will have their views taken into account by the court. However, they will bound by whatever the court decides.

Under the terms of the settlement, the Book Rights Registry will be interposed as an intermediary between the registered rights-holders and Google Inc., to process claims, transmit any payments, and arrange for the arbitration of any disputes over ownership of rights. Opted-in rights-holders who wish to control the ways in which Google exploits their works must go through the Book Rights Registry.

Rights-holders who opt out of the settlement before 4 September 2009 retain the same rights over their books that they have always had.

However, it has been pointed out that they may not be able to exercise those rights in relation to Google Inc. if the company infringes them and they themselves do not possess the financial resources required to take legal action to protect them.

Google has stated that if authors and publishers who opt out of the settlement request the company not to digitise their work, or, if digitised, not to display any contents from it, it is its 'current policy to voluntarily honor such requests'. However, the company has been known to change its stated policies from time to time, and this statement explicitly leaves open the possibility that it may do so in this case. Moreover, one author-publisher who has, in the past, opted out all her books from the Google Library Project and Book Search reported in November 2008 that in spite of this, two of her books had been digitised and put into the Google Book Search corpus.

The Publishers Association of the UK has warned its members that if a publisher opts out, 'Google can use all of the publisher's Books, as Google sees fit … unless and until the publisher brings a copyright infringement suit resulting in a victory or in a settlement with Google that provides otherwise.' This seems, on the face of it, an excessively mistrustful view. In the past, Google has never claimed the right to use copyrighted works 'as it] sees fit', but only to the extent permitted under its (admittedly broad and disputed) interpretation of 'fair use'.

Note: some people have suggested that books that have been digitised by Google through its Partner Program with publishers are opted out from the settlement by default. I can find nothing at all in the settlement that confirms this, and some clauses that seem to indicate the opposite.

Rights-Holders Whose Works Have Not Yet Been Digitised

There seems to be an assumption abroad that the Google Book Settlement only affects works that have already been digitised by Google. This is not so. Only works digitised prior to the opt-out deadline will attract a payment from the Settlement Fund (see below, F. The Promised Benefits of Opting In: The Settlement Fund). But Google has made clear its intention to continue to digitise works published on or before 5 January 2009. Authors and publishers who do not wish to find themselves opted into the settlement should opt out, whether or not they believe that their books have been digitised. If they are opted-in, the settlement gives Google the right (but not the obligation) to digitise any of their works that were published in book form on or before 5 January.

Important Dates

5 January 2009: 'Notice Commencement Date': in the UK and Europe the rights in virtually all works published on or before this date in book form will be affected by the settlement, unless the authors or rights-holders opt out.

4 September 2009: 'Opt-Out Deadline': deadline for opting out of the Google Book Settlement, or opting in and objecting to it. (This was originally 5 May, but was extended.)

7 October 2009: Final Fairness Hearing by the court

5 January 2010: deadline for registering with the Book Rights Registry and claiming a share of the money Google Inc. is offering to rights-holders whose work was digitised without their permission on or before the opt-out deadline, 4 September 2009. (Google does not plan to stop digitising on 4 September, but rights-holders will only be able to claim compensation for works that were digitised on or before that date.)

5 April 2011: deadline for rights-holders who do not opt out of the settlement but who wish to have their books removed from Google's database. After that date, Google will only exclude their books from the database if they have not already been digitised.

There is no final deadline for registering books or inserts. Rights-holders who do not opt out can register their works with the Book Rights Registry at any time.

Information on Procedures

There is information on how to opt out, object, or register on the Google Book Settlement website.

Return to top

F. The Promised Benefits of Opting In

The Settlement Fund

1. Under the settlement Google Inc. will make a payment of $45 million into a fund to pay rights-holders whose work has been digitised prior to the opt-out deadline (now 4 September) and who register with the Book Rights Registry. It is guaranteeing minimum payments of $60 dollars for a book (about £35 at the current exchange rate); $15 for an 'insert' (just under £9): 'inserts' include 'forewords, afterwords, introductions, entire works included in anthologies, and entire poems, short stories, song lyrics or essays'; and $5 (just under £3) for a 'partial insert'(defined as 'an Insert other than an Entire Insert') . If enough rights-holders register that more money is required, Google has promised to provide the necessary additional funds.

The sums that are promised are risible, far smaller than are normally payable for copyright licences. This is hardly fair payment; it is a pacifier for the desperate and the resigned.

By comparison, the lawyers who negotiated the deal on behalf of the Authors' Guild will receive, if the settlement goes through, a payment of $30 million. This fee will be paid by Google within ten business days of the court's approval of the settlement.

2. At least one writers' organisation has misleadingly told its members that 'the author of each book scanned is being offered at least $60 (it could be more)'. But this is plainly not the case. Only if the book is out of print and the rights have reverted to the author will the author receive the full payment.

a) If the book is still in print, the payment will be remitted to the publisher. It appears that this will apply regardless of whether the author has licensed the electronic rights and/or the US rights to the publisher. The publisher will pay to the author 'the appropriate splits or royalties as may be specified in the author-publisher contract for the Book or as the parties may otherwise agree'. Of course, no existing contract will make specific provision for a payment from the Google Book Settlement Fund. It may provide for electronic rights in some shape or form, but there may well be room for argument over how the one-off payment from Google relates to the precise terms of the contract. And in other cases it will simply come down to 'as the parties may … agree'. Whatever the author receives, it will not be the full amount.

b) In the case of out-of-print books where the rights have not reverted to the author, a share will again be paid to the publisher, this time on a fixed basis. For books with a publication date before 1987, the Registry will pay 65% of the revenues to the author and 35% to the publisher. In the case of books published during or after 1987, the split will be fifty-fifty.

Here again, no regard is paid to the question of who actually owns the electronic rights to the work, or the US rights either. In the case of nearly all works published before 1987, and many that have appeared since, the electronic rights will not have been licensed to the publisher. Similarly, in the case of many books that have never been published in the USA, the US rights may also not have been licensed to any publisher. In such cases the Google Book Settlement Agreement is effectively making a bid to supersede and rewrite existing contracts: assigning to publishers rights that they did not previously possess and revenues that they otherwise would not receive. This quite unnecessary intrusion into existing contracts is a very disturbing feature of the settlement agreement.

c) It may be noted that not all authors of 'inserts' will receive a payment from the settlement fund. In order to claim for an 'insert'a claimant will have to state that they 'did not give permission for Online use of the Insert as part of the work in which the Insert appears, or … if such permission was granted, it was no longer in effect on or after June 1, 2003'. Otherwise, it would appear that they cannot make a claim.

3. It has been pointed out that the initial sum of $45 million that Google has put on the table is wildly insufficient to pay all the rights-holders of all the in-copyright works it has digitised, even at the low figures specified in the settlement.

At a conference on the Google Book Settlement held on 13 March 2009 at Columbia Law School, Jule Sigall, formerly of the US Copyright Office and now senior copyright counsel with Microsoft, commented that at a rate of $60 a book, $45 million is sufficient to pay out on 750,000 claims. He noted that Google on their own report had scanned 7 million books, of which he estimated that 6 million are still in copyright. He suggested therefore that the proposed settlement is founded on the assumption that 88% of rights-holders whose work Google Inc. has digitised and now plans to exploit will never come forward to claim payment for the use of their work. It is no doubt possible to pick holes in his exact figures, but the main point is clear: Google is making a giant bet that most of the rights-holders affected by the settlement will never find out that their rights are being exploited and put in a claim for payment.

It may reasonably be assumed that, among those rights-holders who remain in ignorance of the settlement and the benefits that are promised from registering their works, a very large number, probably a majority, will be foreign authors and their heirs and assigns. One commentator has referred to the settlement in terms of a 'foreign land grab' of copyrights.

4. No money will be paid to rights-holders out of the Settlement Fund until at least a year after the settlement is approved by the court; and even with this provision written into the agreement, Google has thought it prudent to warn visitors to the Book Search Settlement website that payments will not come promptly.

Google's Plans for Making Money

1. Google Inc. has a number of plans for commercially exploiting the huge and still-growing corpus of digitised books it controls. Not surprisingly, given the company's core business model, it plans to run advertising alongside search results and online page views. It also plans to sell subscriptions to the whole corpus (or parts of it) to libraries. It plans to sell individual consumers the right to read books or parts of books online and print out pages. Other uses it envisages for the future include selling e-books, in the form of downloadable PDF files, selling Print on Demand copies, and offering custom compilations of pages and portions of books as course materials for the education and training markets.

It cannot be thought an irrelevance that on 2 June 2009 Google announced that it had plans to facilitate the sale of e-books by publishers in its Partner Programme.

2. In order to avoid being legally challenged under foreign copyright laws, Google plans (for the present) to confine these activities to the US market.

3. Under the settlement Google Inc. has promised to pay an initial $34.5 million dollars to establish a Book Rights Registry. This is to be a not-for-profit entity charged with representing the interests of rights-holders in connection with the Google Book Settlement. (It is envisaged that it may also represent rights-holders who are registered with them in arrangements made with companies other than Google, but at present this is only hypothetical.)

Google also promises to pass on to this Registry 63% of the revenues it receives from commercially exploiting the corpus of digitised books. The figure of 63% is arrived at by allocating a nominal 70% to the rights-holders, then slicing off 10% of this figure to cover Google's 'operating costs'.

It may be noted that under the standard arrangements of traditional book publishing the author is paid his or her agreed due and the publisher aims to make a profit out of the income that remains. The procedure laid down in the settlement agreement reverses this principle: Google Inc. will take its substantial (37%) slice off the top of the revenues earned by the books before any money flows by way of the Registry to the registered rights-holders.

The continuing costs of running the Registry are to be funded by taking a percentage of the revenues passed on by Google before what is left is divided among those rights-holders who have successfully registered a claim. An attachment to the settlement agreement estimates that the percentage withheld by the Registry for running costs will be between 10% and 20% of what it receives from Google. It should be noted that this is an estimate only, and does not bind the Registry's directors.

Rights-holders, then, should expect to receive at most about 53% of any income earned by the use of their works. If the Registry turns out to be expensive to operate, they may receive a great deal less.

4. Under the settlement Google Inc. proposes to deal with works differently depending on whether or not they are defined as 'commercially available' according to the terms of the agreement. The agreement defines a book as commercially available at a given point if the rights holder or his or her licencee were offering it for sale new in the United States 'through one or more then-customary channels of trade'. In that case Google will classify the book as 'in print' and will not make any 'display uses' of it, such as providing previews to searchers, including it in institutional subscriptions, or allowing consumer purchase of online access.

The definition of 'commercially available' has caused alarm among foreign publishers, since it seemed to imply that books in print but not published or directly distributed in the US would be made available by Google to searchers (in preview) and customers (for online access), unless and until the rights-holders registered the works at issue with the Book Rights Registry and changed the settings, or applied to have them completely removed from the book corpus. However, following consultation with the lawyers who negotiated the settlement on behalf of the AAP, the Publishers Association of the UK has reported that Google plans to classify any book as commercially available if it can be purchased new from within the US through a website.

If Google were to make a mistake in determining whether a book is available in the US, the agreement lays down one 'sole remedy': Google must correct the mistake within 30 days. A lot of damage might be done to the value of a copyright in 30 days. It is, however, open to the rights-holder to be proactive and direct Google or the Registry to exclude a book, or part of it, from online display or commercial exploitation.

5. In the case of books deemed to be in print, any revenue will be remitted to the publisher who will pay the author some kind of royalty, as described above under 'The Settlement Fund', 2 (a). (Presumably most if not all of this revenue is expected to come from fees for advertisements posted against permitted previews.)

6. Any revenue that flows to registered rights-holders through the Book Rights Registry from the commercial exploitation by Google of books that it does not deem commercially available (and that have not been excluded from such use by the rights-holder) is to be distributed along the same principles as the payments from the Settlement Fund: see above, The Settlement Fund, 2 (b).

7. For works that are included in the corpus of books offered for institutional subscription, an 'inclusion fee' is to be paid, in addition to fees based on the usage of individual books and 'inserts'. The inclusion fee is to be 'targeted at … $200 per Book' (about £119.50 at current rates). By 'targeted at' is meant that this is the figure aimed at, depending on sales of subscriptions. It may not be reached, or it may be exceeded. In any case, it will not be payable to rights-holders until ten years after the first payments for institutional subscriptions are made by Google to the Registry. Once an inclusion fee has been paid, the rights-holder will not be able to exclude the work from use in the institutional subscription corpus unless they repay the inclusion fee, or share of the fee, that they have previously received for the work. The revenue from inclusion fees will be shared among the rights-holders according to the same principles which are applied to other revenue sources.

8. The settlement agreement lays out the default pricing arrangements in considerable detail. It is open to rights-holders to specify their own prices, which the agreement states that only they can change. However, Google Inc. reserves the right to offer 'temporary discounts off the List Prices from time to time at its sole discretion'. The payment to the Registry will be at the list prices 'unless otherwise agreed by Google and the Registry': in other words, it is always open to Google and the Registry to agree to pass the cost of any discount on to the rights-holders. This is an important point which has not received much comment and which in my opinion should ring alarm bells.

9. Under the agreement no payment will be made to the rights-holders for any revenue earned by Google from advertisements served on web pages containing search results. The undisputed right to serve advertisements next to the results of searches on the Book Search corpus is one of the major potential benefits to Google from the settlement agreement. At present no advertisements are served by Google on Book Search results, apparently out of concern that this would weaken its case for the 'fair use' of the books.

I am inclined to suspect that the fees paid for such ads are likely to form a large part of Google's income from its use of the Book Search corpus. I even find myself wondering whether, so long as Google is pocketing the money from the ads served up next to search results, it may not be too worried either way by the success or failure of its venture into bookselling.

Inclusion in the Google Book Search Service

Apart from the one-off payments from the Settlement Fund and the possibility of some income from advertising and sales, the chief advantage of remaining in the settlement, according to its advocates, is inclusion in the corpus of books searched by the Google Book Search service. It is suggested that this is essential promotion, and certainly it offers some promotion. However, it has been questioned how far the chance to show up on searches conducted on millions of books will really help to promote most titles.

Moreover, it may not be necessary to opt in to the Google Book Settlement to remain in the Google Book Search corpus and show up on searches. The Partner Program (see above, C. History and Background: Google Book Search) will continue to operate, and will apparently provide an alternative way for rights-holders to make their books searchable by the Book Search service.

For comparison purposes, it may be noted that the terms and conditions applied to UK users of the Book Partner Program are available on the web. Compared to the terms of the Book Settlement Agreement they are, at least, blessedly short. They bear reading with care, since though the agreement can be terminated, Google retains the right to continue to host and index the content, and use it to provide search results. However, that probably is not something to which most users of the program would object.

It may be noted that the settlement agreement contains nothing to prevent Google from manipulating the results of a search through the Book Search service in any way that it chooses. This point was confirmed at the Columbia conference by Richard Sarnoff, Chairman of the AAP, and one of the chief negotiators of the settlement.

Return to top

G. Burdens Imposed by the Settlement on Rights-Holders

Those authors and rights-holders who are aware of the settlement and make the choice to register, either in order to receive a share of any money earned by their works or to be able to control (to the extent allowed by the settlement) the uses to which their works are put, will have to provide full details of all their publications in book form up to and including 5 January 2009. This will include all their works and excerpts from their works that have appeared in collections with more than one author. (The settlement agreement terms such works 'inserts'.) They will also have to register all editions.

The work involved in registering their rights will be a burden on all authors and rights-holders. For prolific authors, especially those who have had a long career, and for publishers, except the very smallest, it seems likely to prove an enormous burden, imposed on them in the interests of Google, a commercial company, so that it will not have to spend its own money clearing copyrights in the traditional, and legal, manner.

At the conference at Columbia in March, Lois S. Wasoff, a copyright and publishing lawyer who has previously held positions with Houghton Mifflin and Simon and Schuster, made the following comments on the administrative burden that the terms of the settlement would place upon publishers:

The publishers that don't feel there will be very substantial resource demands for the most part are the publishers who haven't yet started looking at the actual implementation. The publishers who have been on the website, started thinking about claiming their books, started thinking about how they are going to deal with the different categories, have - in my experience in talking to a range of people - expressed a lot of concern about the amount of resources it is going to require to comply with this. … The demands will be enormous. … There is also a concern about ongoing maintenance of the data. This is not going to be static data; populating it once is one issue, maintaining it, changing it, is going to be a problem. … Maintaining a database of rights holder information, of information about copyrighted works, is a lot of work.

Rights-holders who opt out of the settlement will also have to provide details of their works. However, they will not have to keep their records up-to-date with the Registry, or track and manage over time the specific authorisations relating to Google's use of each of their individual works.

Return to top

H. Some Further Points of Concern for Rights-Holders

The Likelihood of Piracy

The settlement agreement goes into some detail about the restrictions that will apply if a consumer should purchase a work through Google. The purchaser will be allowed to view the work online, but not to download an electronic copy, or to print more than 20 pages with a single command, or copy and paste more than four pages with a single command. The settlement agreement states that any pages printed out will have a watermark containing 'encrypted session identifying information' that could be used to identify the user who printed them. Recently the Engineering Director of Google Books has stressed in a public talk that the books will not be downloaded; they will be held on a server, 'in the cloud'.

The point of all this is evidently to reassure the rights-holders that their works will be safe from piracy. But of course, it provides virtually no protection.

Anything that is displayed on the screen of an ordinary computer can be copied and saved and/or printed by the user, using the PrintScr key on every standard keyboard, which saves an image of the screen contents to a buffer, from where it can be pasted into a graphics program. For greater convenience, there are screen capture programs available online, some of them free to download. Screen capture is a perfectly legal technology with legitimate uses.

But ingenuity even of this kind will probably not be needed. In an email sent on 1 July 2009 to the book editor and translator Claude Almansi, Dr. Joanne Zack, of Boni and Zack, counsel for the Author Sub-Class in settlement negotiations, has stated that 'the purpose of these provisions that is, the restrictions on printing or on copying and pasting] is to create "speed bumps," so that users will find it simpler and more desirable to simply buy the book rather than copy/paste or print it out in its entirety'.

In other words, all a user would have to do would be to keep on copying and pasting, or printing. Kent Fitch, a programmer at the National Library of Australia, pointed out earlier this year that 'printing' can as easily be to a file as a printer: it is 'up to the controller of the system on which printing is done'.

Once the page images are available, it is an easy matter with modern optical character recognition software to extract the text; indeed, these days there is at least one free online service that will do this, as well as free software that can be downloaded. Any watermarks would be lost in the conversion to text.

Summing up the whole matter, Fitch states:

Given the reality of inevitable piracy of digitised books, the interests of rights holders and Google are seriously misaligned. Google has little incentive to be very worried about piracy, and in any case, they're smart enough to know there's nothing they can do about it. All they need is to sell 40 odd copies (or get equivalent per-book institutional subscription revenue to their book database) and they're in the black. If they sell 100, they've got a 200% return on investment, whereas the rights holders haven't even covered the costs of the layout artist.

Digitised books from the Google repository will be pirated and there's nothing that can be done about it. DRM wouldn't help a bit, copies will be untraceable, watermarks will be removed.

One may speculate that so long as Google is making money from advertisements served on search pages and previews, it will probably be fairly contented. The circulation of pirated copies is unlikely to reduce such uses much, if at all. But it could (and I would hazard, it probably will) amount to a massive destruction in the value of the copyrights and copyright licenses that are held by authors and publishers.

Porous Territorial Boundaries

Another security issue was raised in a point made from the floor at the Columbia conference. The participant noted that regardless of the fact that the complete Book Search corpus was only supposed to be accessible from within the US, people outside the US could use a proxy server located within the US to access the service. No one responded to his point. However, he is quite right. Proxies are offered as a free service by some websites, and they make territories meaningless. It would probably be hard, and might be impossible, to fool the Google Book Service into letting one open an account with it from an address outside the US, but it is likely to be an easy job for those so inclined to access the extensive additional preview facilities to be offered under the settlement, unless Google takes a determined approach to detecting and blocking proxies.

The Book Rights Registry: Lack of Foreign Representation

There is no provision for foreign publishers and/or authors to be represented on the Board of the Book Rights Registry. This was raised as a point of concern from the publishers' side by Herman Spruijt, President of the International Publishers Association, who spoke at the Columbia conference.

The Board of the Book Rights Registry is to consist of an equal number of publishers and authors. The settlement agreement does not specify how the directors are to be appointed. However, it does state that the Registry will be set up by the plaintiffs, that is, the Authors Guild and the AAP. At the conference at Columbia Law School, Michael J. Boni, lead counsel for the Authors Guild in negotiating the Google Book Settlement, told a questioner from the floor that the author directors will be directly appointed by counsel for the Authors Guild and the Guild itself, and the publisher directors will be directly appointed by counsel for the AAP and the AAP. The Authors Guild and the AAP have no plans to appoint directors who are not members of their respective bodies. This was also the substance of the answer to a question on this point put by the UK Publishers Association to Debevoise and Plimpton, the lawyers representing the AAP.

The lack of representation of non-US rights-holders is troubling for many reasons. US and foreign authors and publishers cannot be said to have identical interests in the management of the Registry and the proposed operation of the book service.

One point that suggests itself concerns compliance. Since the digitised book corpus will only be accessible for commercial purposes within the US, foreign rights-holders who register will find it difficult to know whether their work is being used or excluded from use in accordance with the terms that they have stipulated. Moreover, they arguably have a much greater interest in Google's taking steps to maintain territorial security; and there is no doubt that if Google is really serious about confining access to users within the US, determined measures will have to be taken (see above, Porous Territorial Boundaries).

The Expense of Funding the Book Rights Registry

The settlement specifies one of the main tasks of the Registry as the establishment and maintenance of a database of book rights information. This would include ownership of rights, and details of whether, and in which ways, the rights-holders wished their works in the corpus to be commercially exploited by Google.

This is a very large undertaking. At the conference in March held at Columbia Law School, Tracey Armstrong, President and CEO of the Copyright Clearance Center, the US collecting society that administers collective photocopying licences, commented on the amount and complexity of the data that would need to be assembled and organised. She pointed out that this would necessarily include information about 'multiple rights holders per work, works within works, chapter and sub-chapter level rights, types of use or licence type,' the latter depending on the choices made by the rights-holders]. The database would also have to track 'ownership transfers, in bulk and individually, including inheritance'. All this, as she pointed out, 'is a lot of ones and zeros', a lot of data.

And all this data would have to be continually kept up-to-date. It cannot be doubted that this would be an expensive business.

One may note, again, that while it has been estimated that the Registry would require between 10% and 20% of the revenues that are projected to flow from Google, this figure is not binding on the Registry's directors (see above, F. The Promised Benefits of Opting In: Google's Plans for Making Money, 3). It might well turn out that the Registry engrossed a considerably higher percentage of the available income. Pamela Samuelson, lecturing on the Google Book Settlement at the University of North Carolina in April, memorably commented, 'Most collecting societies] … spend a lot of money on themselves'. At the Columbia conference the author Eugene Linden observed that he wanted to be sure that the proposed new system would benefit authors 'and not just the intermediaries'. Similar concerns about the extent to which the fees charged by the Book Rights Registry are likely to cut into the income that flows to authors and publishers have recently been expressed in an editorial on the Authorlink website.

Compulsory Arbitration

If a rights-holder were to pursue a dispute against Google Inc. and/or the Book Rights Registry over a mistake or a disagreement, their only recourse under the settlement agreement would be to submit themselves to binding arbitration. The arbitrators are to be drawn from a pool of arbitrators to be selected by Google and the Book Rights Registry. The arbitrator's fees and costs are to be shared equally by the parties to the arbitration, each of whom must pay their own legal expenses.

Lynn Chu, a New York attorney and literary agent, has predicted that if the Google Book Settlement is accepted by the court it will open 'a Pandora's box of disputes' over rights.

Indeed, it does not take much foresight to recognise this as an inevitable outcome of placing a value, however modest, on rights that in some cases have been dormant for years, not to mention the attempt to pass a huge number of publishing contracts, the diverse products of specific negotiations, through the one-size-fits-all machinery of the provisions in the settlement agreement.

Under the settlement, all disputes between authors and publishers about ownership of rights must, again, be submitted to binding arbitration by an arbitrator from the pool chosen by Google and the Registry. In such a case, the side bringing the dispute to arbitration must pay the Registry a filing fee of $300 (about £179 at current rates; the exact fee is 'subject to adjustment by the Registry'). Half of this will be returned to them if they win. The only exception to the arbitration rule is in the case of disputes between publishers, who are allowed to use the court system to sort out their differences.

This inequality of provision sharply points up the fact that authors are being corralled into a system that is likely to be disadvantageous for them. Lynn Chu has this to say about the proposed system:

Access to your rights as an author at common law in federal court is very important, because U.S. courts have historically been very author-friendly, often overturning publisher contracts of adhesion to uphold author rights. … Expect arbitration to be publisher-centric, and focused on letter-of-the-contract analities, not author-centric. Arbitrators are never as open as federal courts to arguments of justice in equity to exploited authors.

Writing under the pseudonym 'Jane Litte' on the Dear Author blog, Jennifer Gerrish-Lampe, an attorney in Des Moines, Iowa, has also commented on potential pitfalls for authors in the arbitration system. She points out that under the terms of the settlement, arbitrators are expressly not bound by precedent: hence, there is a strong likelihood that the system will breed 'confusion and inconsistent decisions'.

Recently the investigative journalist Pam Martens published a powerful exposť of the corruption and partiality rife in the private arbitration system as it operates in the US.

An Option to Press Rights-Holders to Change Their Settings

Google Inc. is eager to make what the settlement agreement calls 'display uses' of as many works as possible, and many of the default arrangements laid down in the agreement tend towards this end.

As reported above, the company undertakes not to make display uses of books designated as commercially available (see F. The Promised Benefits of Opting In: Google's Plans for Making Money, 4). But once a year has gone by from the date when the settlement is ratified, if Google then decides that a book is not after all commercially available according to its criteria, the agreement provides that the company may request the Registry to change the classification of the book to 'display'. From this point the Registry has 120 days (just over three months) to contact the rights-holder and find out whether they want the book to remain classified as a 'no display' book. 'If the Rightsholder of the Book provides evidence that the Book is Commercially Available or otherwise directs Google that he, she or it wants the Book to remain a No Display Book', then that is how it will be left. But if the rights-holder does not reply within the time specified, Google may change the book to a 'display' book, and make commercial use of it, without their explicit consent.

The clause does not specify any particular methods by which the Registry must seek to communicate with the rights-holder. If a message were to go astray, and a rights-holder suffered harm to their copyright as a result, their only prospect of redress would be by way of arbitration under the arrangements described above (H. Some Further Points of Concern for Rights-Holders: Compulsory Arbitration). Since they will be deemed to have agreed to this ramshackle procedure, it is not clear that they would have much of a case.

It may also be noted that there seems to be no bar to Google's making the same request at a later point. Indeed, there seems to be nothing in the agreement to prevent Google from importuning the rights-holder repeatedly, should it choose to take this course.

Lack of Control over Advertising Uses

Google will not pay out a share of the income generated from advertisements placed on search results pages, but only for those on preview pages and other online page views. The placing of advertisements is controlled under the settlement agreement: for instance, Google may not place ads on pages, or use pop-ups. However, there is no control over the content of the ads, nor is there any scope for an individual rights-holder to state objections to certain kinds of ad. The only option is to exclude a work from advertising uses altogether (and thus from any revenue from ads).

In the recent past, Google and its subsidiary Double-Click have on occasion served ads that are highly offensive to the gay community. For example, it was reported last November that Google Adsense served anti-gay marriage ads to a large number of gay websites. In May this year Google's subsidiary Doubleclick (apparently) served an anti-gay marriage ad to at least one page on LiveJournal.

On 1 July 2009 the feminist website Feministing reported that Google was disallowing ads for abortion services in a number of countries, including Germany, France, Spain and Italy. In none of those countries is abortion illegal. At the same time, 'to create] a level playing field' (in the words of its representative), it had changed its rules to permit 'religious associations' to put up ads campaigning against abortion.

Many writers would find it repugnant for their work to be displayed alongside anti-gay advertisement campaigns. Many would also be repelled by anti-abortion campaigns. Some, of course, would be unhappy about ads for abortion clinics. Altogether, the rights-holders' lack of control over the content of the ads that Google proposes to display alongside their work is a serious matter for concern.

Poor Quality Scans

Tucked away among the 'Miscellaneous Provisions' at the end of the settlement agreement is an item labelled 'Scan Quality'. Google 'makes no guarantees … regarding the Digitization quality … of any Book or Insert'. And this is wise of Google, since among people who make use of the public domain library books made available for download from Google Books, the poor quality of much of the scanning is notorious. Blurred pages, missing pages, cropped pages, folded pages and snapshots of scanners' hands obscuring the text are all familiar problems. (This is not inherent in the scanning process: the quality of the scanned texts made available by the Internet Archive is much more reliable.) Google is said to have begun to address these problems; nonetheless, they are still serving up a large number of defective files. This is irritating even when they are being made available free of charge; it is not going to please paying customers. It seems all too probable that Google Books will spoil the market for quality scanned copies, while offering both readers and rights-holders a very unreliable service.

No Audits

There are provisions in the settlement agreement for the Registry to arrange for an independent auditor to audit Google Inc. and for Google to arrange an audit of the Registry. Rights-holders have no right to audit either Google or the Registry. Moreover, it appears that Google's audits of the Registry are to be confined entirely to the payment of the unclaimed funds (paid in respect of works that have not been registered by the rights-holder) and of any funds paid for public domain works. There is no provision anywhere for auditing the payments made to registered rights-holders.

A Secret Agreement Between the Parties

Article 16 of the settlement agreement mentions the right of either party to terminate the agreement 'if the withdrawal conditions set forth in the Supplemental Agreement Regarding Right to Terminate between Plaintiffs and Google have been met'. However, it then states that 'the Supplemental Agreement Regarding Right to Terminate is confidential between Plaintiffs and Google'.

The whole agreement, then, is subject to conditions that are being kept secret from the overwhelming majority of the settlement class. Authors and publishers are being urged by the promoters of the settlement to opt in to its provisions, but they are not being told all that they need to know in order to assess their best interests. This is yet another highly disquieting feature of the whole business.

Return to top

I. The Unrepresentative Nature of the Plaintiffs

Professor Pamela Samuelson of the University of California, an authority on intellectual property law, has noted that at the start of the legal process, Google Inc. disputed the claim of the Authors' Guild to be representative of the class of all authors everywhere. She points out that most of the authors it represents do not write the kinds of scholarly works found in the university research libraries which have been supplying the books for Google to digitise. She believes that if the proposed settlement, which, of course, confers major benefits on Google, had not been agreed, Google would 'almost certainly' have strongly resisted the move to have the Authors Guild certificated by the court as representative of the class of authors. She also believes that Google would have won its point on this.

In order for the settlement to be ratified by the court, the judge must be satisfied that the Authors Guild is an appropriate representative of the class of people affected by it. So the question is one of considerable importance.

On the matter of its representativeness, it may be noted that the Authors Guild is relatively small: it has roughly the same number of members as the UK Society of Authors, in a country with five times the population.

Charles E. Petit, a US lawyer specialising in intellectual property law, has queried whether the individual plaintiffs, in whose names the original law suits were brought, are adequate representatives of the two sub-classes of the settlement class: the publisher sub-class and the author sub-class. He concludes that they are not. In the case of the author sub-class, he gives a longish list of important categories of author who are not represented among the group of five named plaintiffs.

Using information freely available on the web, I have investigated the origins and publishing histories of each of the plaintiffs. None of them originate from outside the US. (Two were born in New York City, one in Yonkers, one in Chicago, and one in Texas.) All have published mainly or exclusively in the USA. There is no representative here of authors from outside the USA, though the settlement class comprehends a very large number of non-US authors.

Moreover, under the rules of the Authors Guild, membership is specifically restricted to authors of works that have been published 'by an established American publisher'. Many of the foreign authors whose works have been digitised by Google have never been published in the US, and so are not eligible to join the Guild. This is another strong argument that the members of the Authors Guild cannot be taken as representative of the whole author sub-class.

Return to top

J. The Google Book Settlement, 'Orphan Works' and Intellectual Property Rights

'Orphan Works'

There has been much use of the term 'orphan works' in the publicity given to the Google Book Settlement in the press and on the web. Initially, this misled me, and has certainly misled other authors also, into assuming that the settlement agreement could not and was not intended to affect the rights of authors who were still alive and could easily be contacted through the usual channels.

In the past, the term 'orphan works' has been used to describe out-of-print works whose rights-holders cannot be traced. As any anthologist knows, tracing rights-holders can be a time-consuming and demanding process; that said, there are probably far fewer books whose rights-holders are genuinely untraceable than is sometimes stated or implied.

It is plainly Google's hope and expectation that if the settlement goes through it will be relieved of its obligations under copyright law to make a diligent search for all the individual rights-holders whose work it proposes to exploit commercially. It aims to bypass the process of negotiating licensing deals with them, directly or through their agents, at a fair market price.

If this is allowed to happen, it will be a very serious diminution of the rights and benefits currently enjoyed by copyright holders. It would, of course, be of considerable economic advantage to Google. For one thing, it would take a great deal of work to do a diligent search for all the rights-holders, as Google, it cannot be doubted, is keenly aware.

Rights-holders who remain in ignorance of the settlement will have their works exploited and will have no control over the uses to which it is put, nor will they receive any income from it. In many, perhaps most cases, those works will not be 'orphans', in the sense of having no identifiable owners. It will simply be the case that not enough effort has been put into the task of tracing the rights-holders. It is likely that the majority of untraced rights-holders will live outside the USA.

Unclaimed Funds

One provision of the settlement that has attracted much critical comment is the plan for distributing what the agreement terms 'unclaimed funds', revenues flowing from the use of works belonging to rights-holders who have not registered with the Book Rights Registry (the so-called 'orphan works'). These funds would be held for five years in case the rights-holders turned up. What then remained unclaimed would be applied in the first instance to paying 'the operational expenses of the Registry'. If there were any money left, it would be distributed differently depending on its source.

(a) If it was earned from advertising, or sales of access to works by consumers, it is to be used to top up the payments to the registered rights-holders until the sum arrives at 70% of what their works have earned from Google. The thinking behind this was explained by Alexander Macgillivray, Senior Product and Intellectual Property Counsel at Google Inc., in a lecture at the Berkman Center, Harvard on 21 July 2009. Under the settlement, 70% is the notional percentage of a work's earnings allocated to the rights-holders, but 10% of this will be sliced off by Google to pay for 'operating costs', thus arriving at the figure of 63%. The intention is that where there is money available from the 'unclaimed funds' it should be applied to returning this 10% slice to the registered rights-holders. If there is any money left after this, and only if, it is to be allocated to charities with educational objects, such as the promotion of literacy.

(b) After the expenses of operating the Registry have been met, if there are unclaimed funds earned from the sales of library subscriptions, these are to be distributed in their entirety among the registered rights-holders.

The bland language of 'unclaimed funds' glosses over the fact that this is money that cannot, in equity, be applied to the purposes of the Book Rights Registry, since it belongs to rights-holders who are not registered; nor is there any justification for dividing all or part of what is left among the registered rights-holders, since they have no copyright interest in the works that earned it.

Strong critics of this aspect of the settlement agreement include Professor Pamela Samuelson, and James Grimmelmann, an Associate Professor at New York Law School.

It seems a well-founded assumption that only a minority of rights-holders would be likely to register their works with the Book Rights Registry (see above, F. The Promised Benefits of Opting In: The Settlement Fund, 3). The unclaimed funds would therefore make up the bulk of the revenues earned by the Google book corpus. It is noteworthy that the money is to be applied first to paying the expenses of running the Registry, and that the language used ('any remaining funds') clearly opens up the possibility that the whole of this money may be lavished on that task.

One quite persuasive inference from this is that Google Inc. and its partners in the Authors Guild and the AAP suspect that the revenues flowing from the proposed uses of the book corpus might turn out to be paltry; while the establishment and maintenance of the book rights database would inevitably swallow up a good deal of money (see above, H. Some Further Points of Concern for Rights-Holders: The Expense of Funding the Book Rights Registry). Without the unclaimed funds to draw on the Book Rights Registry and its database may not be an economic prospect.

The diversion of the income from the so-called 'orphan works' (in reality mostly works by non-US authors) would thus be a crucial feature of the whole scheme: and it is equally crucial that not too many of these 'orphans' are claimed by their rightful owners. Under the settlement, one of the functions of the Book Rights Registry is to 'attempt to locate Rightsholders with respect to Books and Inserts'. However, this obligation is not set out in any detail, and it is an indisputable point that every rights-holder they locate will reduce their available funding.

It may be the case that the Google Book Settlement, if it comes into force, will function in a way not dissimilar to a Ponzi scheme: if every rights-holder whose work has been appropriated and added to the book corpus were to register a claim for the revenues earned by their work, it is possible that the system may collapse.

International Copyright Law

The Berne Convention does the work, very conveniently for Google Inc., of bringing non-US authors and rights-holders within the scope of the settlement. Yet at the same time the settlement agreement aims to dismantle some of the fundamental protections for intellectual property enshrined in international copyright law.

Under the terms of the settlement agreement, authors would have to register their works with the Book Rights Agency in order to retain and protect their rights in them. This contravenes a key principle of international copyright law, spelled out in Article 5.2 of the Berne Convention: 'The enjoyment and the exercise of these rights shall not be subject to any formality'.

Article 9.1 of the Berne Convention states: 'Authors of literary and artistic works protected by this Convention shall have the exclusive right of authorizing the reproduction of these works, in any manner or form'.

This clause is subject to certain limitations and exceptions, set out in the one that follows it: 'It shall be a matter for legislation in the countries of the Union to permit the reproduction of such works in certain special cases, provided that such reproduction does not conflict with a normal exploitation of the work and does not unreasonably prejudice the legitimate interests of the author' [9.2]. The limitations and exceptions specified in the Berne Convention are restated in the TRIPS Agreement of 1994 [Article 13] and the WIPO Copyright Treaty of 1996 [Article 10].

The conditions specified here, only in 'special cases', non-conflict with 'a normal exploitation', and protection for the author's 'legitimate interest', are often referred to as the 'three-part' or 'three-step' test. But before I consider the 'three-part test', there is another point that calls out to be made: though the Google Book Settlement seeks to have a quasi-legislative effect, it is not a piece of legislation.

Having regard to the 'three-part test', it is not hard to argue that the Google Book Settlement, if put into operation, would interfere with authors' 'normal exploitation' of their works and damage their 'legitimate interests'. Indeed, this is very much what the plaintiffs asserted in their original complaints against Google Inc. Causes of 'irreparable injury' specified in the complaint from the authors include 'depreciation in the value and ability to license and sell their Works'; 'lost profits and/or opportunities' and 'damage to their goodwill and reputation'. The publishers are less specific, but no less adamant that Google's unlicensed use of their publications is causing 'harm' to their copyrights.

This was over scanning, indexing and displaying snippets. How much more harm will be caused to rights-holders whose written content, without their permission or knowledge, is to be exploited for advertising purposes, made available for purchase, and placed at risk of being pirated!

Further: the Google Book Settlement certainly does not conform to the established definition of a 'special case'. In a case brought under the TRIPS agreement, a WTO Dispute Resolution Panel found in 2000 that an exception made as a 'special case' must be narrow in its scope, 'in quantitative as well as a qualitative sense'. The Google Book Settlement, of course, is drawn so as to encompass an enormously large number of rights-holders and works.

The US is a Contracting Party under the WIPO Copyright Treaty. Article 14.2 of the Treaty states: 'Contracting Parties shall ensure that enforcement procedures are available under their law so as to permit effective action against any act of infringement of rights covered by this Treaty, including expeditious remedies to prevent infringements and remedies which constitute a deterrent to further infringements.'

Private individuals cannot take action to enforce adherence to international treaties and trade agreements. Only nation states and organisations of nations are in a position to do that.

As I said at the beginning of this paper, it is my trust that the governing bodies of the EU and the government of the UK, together with other European governments, will take action to uphold the rights that are guaranteed to European authors and copyright-holders under international law, and defend their intellectual property, now under threat in the US from an audacious attempt at private legislation by way of a court-approved settlement.

Gillian Spraggs


10 August, 2009

Lightly revised, 28 August, 2009

Return to top