But, within the US, the company has scanned both in-copyright and out-of-copyright works. In its defence, Google points out that it displays only snippets of books that are in copyright – arguing that such displays are "fair use". But critics allege that by making electronic copies of these books without first seeking the permission of copyright holders, Google has committed piracy.
"The key principle of copyright law has always been that works can be copied only once authors have expressly given their permission," says Piers Blofeld, of the Sheil Land literary agency in London. "Google has reversed this – it has simply copied all these works without bothering to ask."
In 2005, the Authors Guild of America, together with a group of US publishers and publishers, launched a class action suit against Google that, after more than two years of wrangling, ended with an announcement last October that Google and the claimants had reached an out-of-court settlement. The full details are staggeringly complicated – the text alone runs to 385 pages – and trying to summarise it is no easy task. "Part of the problem is that it is basically incomprehensible," says Blofeld, one of the settlement's most vocal British critics.
Broadly, the deal provides a mechanism for Google to reimburse authors and publishers whose rights it has breached (including giving them a share of any future revenue it generates from their works). In exchange for this, the rights holders agree not to sue Google in future.
The settlement stipulates that a body known as the Books Rights Registry will represent the interests of US copyright holders. Authors and publishers with a copyright interest in a book scanned by Google who make themselves known to the registry will be entitled to receive a payment – in the region of $60 per book – as compensation.
Additionally, the settlement hands Google the power – but only with the agreement of individual rights holders – to exploit its database of out-of-print books. It can include them in subscription deals sold to libraries or sell them individually under a consumer licence. It is these commercial provisions that are proving the settlement's most controversial aspect.
Critics point out that, by giving Google the right to commercially exploit its database, the settlement paves the way for a subtle shift in the company's role from provider of information to seller. "Google's business model has always been to provide information for free, and sell advertising on the basis of the traffic this generates," points out James Grimmelmann, associate professor at New York Law School. Now, he says, because of the settlement's provisions, Google could become a significant force in bookselling.
Interest in this aspect of the settlement has focused on "orphan" works, where there is no known copyright holder – these make up an estimated 5% to 10% of the books Google has scanned. Under the settlement, when no rights holders come forward and register their interest in a work, commercial control automatically reverts to Google. Google will be able to display up to 20% of orphan works for free, include them in its subscription deals to libraries and sell them to individual buyers under the consumer licence.
"The deal has in effect handed Google a swath of intellectual copyright. It is a mammoth potential bookselling market," says Blofeld. He adds it is no surprise that Amazon, which currently controls 90% of the digital books market, is becoming worried.
But Dan Clancy of Google dismisses the idea that, by gaining control over out-of-print and orphan works, Google is securing for itself a significant future revenue stream. He points out that out-of-print books represent only a tiny fraction of the books market – between 1% and 2%. "This idea that we are gaining access to a vast market here – I really don't think that is true."
James Gleick, an American science writer and member of the Authors Guild, broadly agrees. He says that, although Google's initial scanning of in-copyright books made him uncomfortable, the settlement itself is a fair deal for authors.
"The thing that needs to be emphasised is that this so-called market over which Google is being given dominance – the market in out-of-print books – doesn't currently exist. That's why they're out of print. In real life, I can't see what the damage is – it's only good."
It is by no means certain that the settlement will be enacted – it is the subject of a fairness hearing in the US courts. But if it is enacted, Google will in effect be off the hook as far as copyright violations in the US are concerned. Many people are seriously concerned by this – and the company is likely to face challenges in other courts around the world.
Over the coming months, we will hear a lot more about the Google settlement and its ramifications. Although it's a subject that may seem obscure and specialised, it concerns one of the biggest issues affecting publishing and, indeed, other creative industries – the control of digital rights.
No one knows the precise use Google will make of the intellectual property it has gained by scanning the world's library books, and the truth, as Gleick points out, is that the company probably doesn't even know itself. But what is certain is that, in some way or another, Google's entrance into digital bookselling will have a significant impact on the book world in years to come.