Last week, a US judge in Manhattan made a landmark decision. As to what it means, opinions vary. Some see it as arresting the cultural progress that began with the Enlightenment; others are celebrating Judge Denny Chin's ruling as the blocking of a predatory move by a giant corporation to control access to the world's cultural heritage. The truth, as always, lies somewhere in between.

First, the backstory. In 2004, Google decided that it would digitise all the books currently held in a series of major libraries with the aim of making their contents globally available to anyone with a web browser. The project was the brainchild of Larry Page, one of Google's two founders, who is soon to become the company's CEO. It was a typical Google project – audacious, difficult and fiendishly expensive – but one that fitted naturally with the company's stated mission to "organise the world's information".

In the past few decades, everything that has been published in print also exists in digital form, but most of the world's books existed only in paper form and digitising them was a task that most people regarded as insuperable. That, however, is exactly the kind of challenge that turns on Google engineers. So they set to work, and to date have scanned upwards of 15 million volumes, the texts of which reside on their server farms, ready to be called up by a simple web search.

For some reason, (possibly because it feared rejection) Google omitted to consult publishers and copyright holders about this bold endeavour. The results were predictable. In 2005, the Authors Guild and the Association of American Publishers sued for "massive copyright infringement". Google claimed that the project represented "fair use", arguing that it was providing full text only of works that were in the public domain and showing just a searchable online summary for books still under copyright protection. But the publishers were not placated and the action continued.

Then, in an astonishing move, the company and the plaintiffs reached an agreement: in return for a payment of $125m, together with arrangements for collecting and distributing royalties, Google would be free to continue with its project. In retrospect, to call this astonishing is an under-statement. If ratified, the agreement would effectively hand over mankind's cultural heritage to Google, because it would indemnify the company from liability for infringing the copyright in so-called "orphan" works – works still in copyright but whose rights holders are unknown or untraceable.

This is a big deal for two reasons: first, a large percentage of extant works fall into this category; and second, copyright infringement carries "strict liability", which means that there are no limits to the damages that might be awarded for infringement, even if it was done inadvertently. Strict liability is the reason why nobody has hitherto contemplated large-scale digitisation: the potential financial hazards are too unpredictable.

The negotiated agreement between Google and the authors and publishers was submitted to the court for Judge Chin's approval. It was, predictably, opposed by Google's competitors (Microsoft, Amazon, etc), but also by some copyright experts and, most significantly, the US Department of Justice. Until Tuesday, nobody knew what the judge would decide.

Now we do. In a 48-page decision the judge concluded that the $125m deal would give the internet giant the ability to "exploit" books without the permission of copyright owners, echoing the Justice Department's concerns about the deal. "While the digitisation of books and the creation of a universal digital library would benefit many," Judge Chin said, the agreement would "simply go too far" because it would "give Google a significant advantage over competitors, rewarding it for engaging in wholesale copying of copyrighted works without permission"" It would also "arguably give Google control over the search market" for books – which is what the Justice Department had argued in its submission.

So Judge Chin rejected the agreement. He also indicated that he would consider a revised agreement that would better protect copyright owners, say, by requiring that they opt into the digitisation process. But an agreement modified along those lines would effectively cripple the Google project, because it would mean that most orphan works would be digitally invisible. So it looks as though Larry Page's dream has been derailed.

There is, as it happens, an obvious solution. The US Congress, followed by the EU, could change the rules about orphan works, and in particular about the strict liability that copyright infringement carries. Nobody is arguing that infringement should not be punishable, only that the damage resulting from it should be impartially assessed and the penalties should be proportionate to the damage. It's just common sense. The only problem is that common sense about copyright is the one thing that Congress doesn't do.