Scottish law exempts academic work from the freedom-of-information laws, but the rest of the United Kingdom does not. Ireland also exempts, and although the United States is commonly thought to, it turns out that, as so often in that country, it is left to the courts to decide. So, just what should researchers make of freedom-of-information laws?

American climatologist Michael Mann, now at Pennsylvania State University in University Park, probably knows the score better than most. And in the latest twist in a long-running saga over who should be able to read Mann's e-mails, a Prince William County Circuit Court judge in Manassas, Virginia, last week tore up an agreement that would have given the e-mails, with conditions, to attorneys for the American Tradition Institute (ATI), a conservative think tank. Judge Gaylord Finch also granted Mann's request to join the University of Virginia (his former employer and holder of the e-mails) in a lawsuit to block their release.

Access to personal correspondence is a freedom too far.

As both sides argue about whether the messages should ultimately be made public, the two legal decisions come as welcome news to those (including this journal) who believe that access to personal correspondence is a freedom too far. But the case highlights, yet again, how woefully unprepared the academic community is to meet this kind of challenge. This must change.

Certainly, the University of Virginia caved in too easily when it signed the agreement that granted the ATI access to the e-mails last spring. Given the tone of public statements that have come out of the ATI, the university should never have agreed to hand over confidential material of any sort.

But the university and its attorneys deserve credit for rectifying the situation. And despite appearances, to fight such requests is not against the letter, or indeed the spirit, of perfectly proper regulations designed to improve the accountability of public bodies. In fact, Virginia's freedom-of-information law provides the university with a solid basis to deny access to this kind of blanket request for e-mail records: academic work is exempt. This is as it should be, and the university should fight to protect that exemption now and in the future.

Yes, the public has a right to know, and yes, greater scrutiny of public spending is a good thing. But research practice is typically protected for good reasons too. To protect academic freedom is a foundation for intellectual property and copyright laws, while in court, both Mann and the university warned of the chilling effect of such demands on communication between scientists. Certainly, many researchers are more wary of e-mail today, and given Mann's experiences, who can blame them?

His case is high profile, but scientists and academics watching it (as well as the related attempts by Virginia's attorney-general Ken Cuccinelli to force the release of the same e-mails) should be cautious about drawing broad conclusions from how it may pan out. Even within the United States, the eventual ruling won't serve as much of a precedent outside Virginia. Federal agencies in the United States are subject to the federal statute, but state universities and research institutions must all play by the laws enacted in their own states.

Across those states it seems that this kind of academic exemption is common, but not universal, and its application would vary according to precedents set locally. In other words, it will be up to individual universities to work out how to address these kinds of cases as they emerge in future.

Mann's decision to join the lawsuit was spurred by the initial decision of the university to grant ATI access to his e-mails, a move with which he disagreed. He suggests that universities may be limited in what they can do to fend off these attacks, or that their interests may not always align precisely with those of individual researchers.

Mann is also getting help from a new fund especially designed to aid climate scientists hit by legal challenges, and organizations including the American Geophysical Union, the American Association of University Professors and the Union of Concerned Scientists have weighed in as well. All of this is good and useful, but it is no substitute for a solid institutional defence. Individual universities and research institutions everywhere should review their own policies and make sure they know the applicable laws as well as do those who would use them for mischief, or worse.