The​ Italian newspaper Il Sole 24 Ore recently outed (or claimed to be outing) the writer of the Neapolitan novels concealed behind the pseudonym Elena Ferrante. Has the press – or anyone else – any moral right to do this? Is an author’s identity an aspect of her personal privacy, to be disclosed or withheld as she chooses? Or is it information which belongs as much in the public domain as the books she writes?

Anonymous and pseudonymous publication has a long history. It may now be the exception in literary and specialist journalism, but at the start of the 19th century it was pretty much the rule – to the extent that France in 1850 legislated to forbid the publication of unsigned articles on philosophical, political and religious subjects. A new book by Eric Barendt, Anonymous Speech: Literature, Law and Politics (Hart, £25), traces the contemporaneous voluntary abandonment of anonymity in England and the often pompous arguments that accompanied it. The fact was that journals’ recruitment of well-known writers – Thackeray, Dickens – was starting to put a premium on names. So when the Fortnightly Review started up in 1865, it announced that all its articles would be signed and free of editorial pressure. By contrast, from its foundation in 1913 the New Statesman anonymised its contributors, though the editor, having explained that this was necessary in order to establish a common style and tone, couldn’t resist announcing that Sidney Webb and Bernard Shaw would be writing for it. In 1925 the Spectator, after not quite a hundred years of unsigned articles, abandoned anonymity, and the New Statesman followed. Articles in the TLS remained anonymous until 1974, and obituaries in the Times and Telegraph are unsigned to this day. So are the entirety of the Economist and the bulk of Private Eye.

As an academic lawyer Barendt is interested in who benefits, who profits and who loses by anonymity. Important in the first category are whistleblowers, journalists’ sources and internet trolls; in the second, government and the media; in the third, trolls’ victims. Take trolls, whom the shield of anonymity enables to disseminate insult and calumny they would not have the nerve to transmit if it was known who they were. Psychologists have baptised this the online disinhibition effect. Not only where it involves threats to kill, as it frequently does, but also where it is obscene, menacing or racist, as it frequently is, it constitutes a criminal offence. It also does widespread, often lasting, sometimes lethal harm. Not every target of trolling has Mary Beard’s or Caroline Criado-Perez’s fortitude. But human rights law still does not tell us whether the right of free expression includes a right to withhold the writer’s or speaker’s identity. And behind this silence lurks a larger one which free speech theory has trouble addressing at all: is there a right to lie?

There are many kinds of lying which the law already catches: lying to defraud people of money or property; lying about someone’s character; lying on oath. In Germany and a number of other states Holocaust denial is a statutory lie and a crime. But, such offences aside, human rights law has found itself unable to draw a line between freedom to speak your mind and freedom to fabricate, falsify or mislead. The reason is not hard to see: a court called on to adjudicate on, for instance, some of the tall stories propagated in the course of the EU referendum campaign would have had to assume the role of a ministry of truth. At the same time, free speech advocacy needs to face up to the full magnitude of what it advocates; for nobody today can seriously believe in the existence of a utopian forum, a marketplace of ideas, where the true drives out the false. If anything, the converse is true. Timothy Garton Ash, acknowledging in his book Free Speech the ubiquity of lies in American political argument, responds: ‘Yet at least in most democratic countries there are two (or more) competing false narratives, and having two is more than twice as good as having only one.’ Are lies really self-cancelling, or is the candid answer that freedom to lie is part of the price we pay for freedom to speak?

Anonymity compounds the conundrum. To expose or refute falsehood can be hard, but it can be far harder when you don’t know who the falsehood is coming from. Internet providers, who know who their users are, claim to be as justified in protecting their identity as Ferrante’s publisher was in protecting hers. Does the law have anything useful to say about such claims?

The Court of Appeal has recently declined to accept that the right of free expression in Article 10 of the European Convention on Human Rights accords a source any right to the preservation of his or her anonymity. In consequence, and also because the right was being relied on to conceal criminal conduct, it afforded no protection to a prison officer who had sold a journalist information about some notorious prisoners. The journalist was acquitted on grounds of public interest; the prison officer went to jail for misconduct in public office. In 2009 a serving police officer whose blog, written under the alias Night Jack, had just won the Orwell Prize for citizen journalism, tried to stop the Times revealing his identity. His case was based both on his entitlement to privacy and on his right to freedom of expression. He lost. The High Court held that he had no reasonable expectation of privacy since blogging is essentially a public activity, and that in any event the public interest in knowing that it was a policeman who was excoriating politicians and the police service outweighed any privacy interest he might lay claim to.

So far so liberal, you might say. But there is an unwholesome contrast between the judicial vindication of journalists’ right to name names and the protection afforded by the law to the identity of journalists’ own sources. If, as Barendt points out, Night Jack, instead of running his own blog, had taken his criticisms and revelations to the Times and it had published them, any attempt by the police to force the paper to identify him would in all probability have failed. Since 1981 contempt of court legislation has restricted the courts’ power to compel the disclosure of journalists’ sources to cases where to do so is necessary for national security or for the ends of justice, law and order; and the Strasbourg court has adopted a parallel approach to the privacy and free expression rights in the European Convention. This works well enough when sources are honest and dependable; but when the source is plausible but malicious, or the story, though almost certainly invented, irresistibly sensational, justification of the protection begins to look dubious. Yet how can you judge whether information is contaminated by malice without knowing who the source is – assuming, what is not always the case, that the publisher itself knows?

It is not widely appreciated that the largest beneficiary of journalistic source protection in Britain is central government. Whitehall’s regular off-the-record briefings, feeding friendly journalists unattributable departmental versions of events and policies, make it unsurprising that legislation conceived in Whitehall and delivered in Westminster has taken good care to shield the identity of journalists’ sources; though here, at least, freedom of information legislation offers diligent journalists some chance of checking on what they are told, and some journalistic codes – the Financial Times is a case in point – require two separate sources for stories that are to remain unattributed. But add to the gentlemanly Whitehall briefings the pseudonymous armies deployed by governments from China to Israel, Turkey to Russia, to post purportedly disinterested comments in their own favour, and the notion of the internet as a forum of information begins to wilt.

How fundamental in any event is the right to remain unidentified? The US Supreme Court in 1972 decided that the First Amendment prohibition on laws abridging the freedom of speech or of the press did not make the preservation of sources’ anonymity a constitutional right capable of defeating a grand jury subpoena. But they left it open to states to enact legislation protecting journalists’ sources, and at least forty states have since done this. In other words, anonymity in US law is not a fundamental right but one of the instruments that can be used to balance privacy and publicity.

Barendt is correspondingly sceptical of fundamentalist arguments for source anonymity. Knowing the originator of a press story, he suggests, will among other things make it easier for the public to gauge its reliability. Maybe; though when other publications then set out to discredit the originator, the public is more likely to find itself in a hall of mirrors than in the light of day. But these and other nuances are now all but eclipsed by the unprecedented fact that internet trolling has turned anonymity from a shield for the concerned and the opinionated into a bludgeon for liars and bullies.

There is no decent excuse for social media providers’ continuing refusal to expose trolls. There is neither moral justification for refusing nor technological difficulty in exposure, and some providers (for example Facebook and Google+) prohibit anonymity as a matter of principle. The greater number of providers who shield their users’ identity at least pay a penalty in their inability to sell advertisers full access to their customers; but the shield they provide for malice and worse is difficult to defend. Security services, for their part, have relatively little trouble in identifying the authors of transmissions, at least unencrypted ones. And there’s the rub: in many parts of the world, anonymity is all that stands between the purveyor of unofficial views or information and prison or torture. It is the reason the UN’s special rapporteur on freedom of expression, David Kaye, has defended the right to communicate anonymously on the internet, and that Germany has actually legislated to protect the use of anonymity and pseudonymity on social media.

Barendt advances two particular arguments against the protection of anonymity. One is the absence of any responsible intermediary, such as there is in the print media, able to filter out mendacious or defamatory messages and posts. The second is the fear experienced by victims of trolls and cyber-mobs, who cannot be sure how close they are physically to the source of the threats they receive. The first of these has an air of unreality in relation to agenda-driven media to which factuality and balance are all but strangers. The second is real and important and deserves serious police resources to deal with it. Under the Communications Act 2003, which criminalises the senders of grossly offensive or menacing messages, jail sentences were handed out in 2014 to two individuals who had directed violent online abuse at Caroline Criado-Perez and her supporter the Labour MP Stella Creasy over the not exactly revolutionary suggestion that it might be time to commemorate a woman on English banknotes. Barendt is right to suggest that social media platforms could do a lot more to block trolling, to ban known trolls and to call in the police where appropriate.

Libellous posts, as opposed to threatening or purely abusive ones, are more of a problem. One of the arguments against compelling internet service providers to identify the authors of defamatory posts is that nobody takes them seriously because they are anonymous. Barendt is rightly sceptical of this: the internet, he points out, can destroy a reputation worldwide in a few seconds, and there are not a few instances of journalists making uncritical use of such material. He endorses the view of the Australian judge Michael Kirby that libel law needs to be ‘technology-neutral’, in other words, to be the same whatever the medium; but that approach has not stopped courts on both sides of the Atlantic treating what the California appeal court has called ‘a looser, more relaxed communication style’ as modifying the meaning a damaging post might otherwise have borne. The 2013 Defamation Act supplies a trade-off here. A website operator sued for disseminating anonymous defamatory matter may be able to escape liability only if it assists the claimant to identify the author. The US courts, more chary of chilling free speech, will rarely order such unmasking where the defamatory material, however extreme, is political, but they will readily order it where the libel is personal.

Given that the 1948 Universal Declaration spells out the right to vote in elections by secret ballot as a fundamental human right, it is useful to be reminded that this paradigmatic form of anonymous speech is not that old and is still contentious. Its introduction by the 1872 Ballot Act, ending centuries of intimidation by local landowners and employers and bribery by candidates, was heavily contested. Conservative politicians contended that secret ballots were contrary to our traditions of honesty and openness. Perhaps ominously, the late Justice Scalia pointed out in 2010 that the US Supreme Court had never upheld a right to vote anonymously. Unlimited campaign funding, discriminatory voter identification, restricted polling hours: might anonymous voting be the next target for the Republican wrecking ball? Or does it no longer matter?

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences