In 2004, the book House of Bush, House of Saud reached the bestseller list in The New York Times. It was on sale in bookshops in France, Germany and Japan. But it was not published in Britain. It thus joined the ranks of books — Nobel Dreams by Gary Taubes was another — that, one can only infer, were in effect censored owing to the reluctance of publishers to risk a libel suit under English law.

The reason? At the heart of the issue sits the burden of proof of unjustified defamation. In the United States, aggrieved 'claimants' who believe that a book, magazine or newspaper has significantly damaged their reputations without justification generally have to prove to the courts that the allegations are false. In Britain, the author and publisher (assuming both are sued) have to prove that the allegations are true. Also critical to the outcome is an interpretation by the court of the meaning in the mind of a reasonable person of what was originally published.

There is no reason to give a special privilege to science within libel law.

From time to time, Nature's journalism has stimulated people to get their lawyers to send us threatening letters. Because we have been able to show that our reporters tackle issues of genuine public interest, do their homework and give people the opportunity to reply to allegations within our stories, such threats have generally been rebuffed.

Simon Singh, a leading independent science writer, was less fortunate. In an article in The Guardian, he tackled an issue of genuine public interest: whether or not there is any evidence that chiropractic treatment is effective, as some chiropractors have suggested, for various childhood conditions, including asthma. The British Chiropractic Association (BCA) claimed that he had defamed its reputation and threatened to sue him for libel. According to Singh, the BCA did not take up an offer by The Guardian of space in which to respond.

Again according to Singh, he was expressing his own opinions and was not accusing the BCA of being dishonest. But last week a judge controversially decided that Singh's language amounted to factual statements (rather than opinion) and that the meaning of Singh's article was that the BCA was deliberately using unfounded claims. This meaning makes defending a libel case extremely difficult, as he would have to prove that that was indeed the case, even though it was not the meaning that he intended. So he has decided to appeal against the ruling on meaning.

The campaign that was launched in support of Singh makes the valid point that debates about scientific claims should be conducted in the open rather than pursued in the courts. In a statement, the BCA claims to support this ethos and that it was only opposing Singh's characterization of its integrity. However, many observers, including Nature, believe that its resort to the courts is tantamount to suppression rather than toleration of debate.

There is no reason to give a special privilege to science within libel law. What is required is that English law be made not only consistent with that of other countries, but also with Article 10 of the European Convention on Human Rights, which protects the right of freedom of expression. The letter of English law may not breach the latter, but the threat of the costs involved, which can be huge, and the burden of proof are enough to scare many writers and publishers into a restraint of freedom of expression.

Anyone who believes that this is an issue only for those living in Britain could not be more mistaken. It affects any online content that can be read in the United Kingdom. Wisely, there is legislation in the United States that, in some instances, protects its citizens from English court judgements.

The English law needs to be changed. There urgently needs to be a review of the burden of proof and a restriction of the costs that can reasonably be awarded to whoever wins the case. Only then can Britain claim to be a country that supports free speech.