The Defamation Act 2013 replaces Reynolds privilege with a new defence of ‘publication on matter of public interest’. On one view, the statutory defence aims to provide a broader protection for expression than was previously found under the common law.
Unlike Reynolds, the 2013 Act provides no list of standards to help decide whether the publisher has acted with sufficient responsibility to warrant protection. Instead, the defendant now has to show that ‘the statement complained of was, or formed part of, a statement on a matter of public interest’ and that the defendant ‘reasonably believed that publishing the statement complained of was in the public interest’.
Just how much broader it proves to be than Reynolds privilege will, of course, depend on the way the provisions of the statute are interpreted. In particular, will the courts look to the type of professional standards found in the Reynolds test to decide whether a publisher has a reasonable belief that a statement is in the public interest? The existing Article 10 jurisprudence may encourage the courts to do so.
The European Court of Human Rights has stressed on many occasions that the freedom to engage in discussion of political matters is matched with duties and responsibilities – which are often defined by the ethics of traditional journalism. For example, in Axel Springer v Germany ([2012] ECHR 227) the Court reiterated:
‘the safeguard afforded by Article 10 to journalists in relation to reporting on issues of general interest is subject to the proviso that they are acting in good faith and on an accurate factual basis and provide “reliable and precise” information in accordance with the ethics of journalism.’
Following this approach, the courts may well consider the steps taken by the defendant to verify the statement, as was the case under Reynolds, when deciding if the publisher’s belief was reasonable.
In a paper to be published in Public Law later this year, I argue that much of the European free speech jurisprudence was formulated with the traditional mass media in mind. This is unsurprising given that newspapers and broadcasters were the dominant means for disseminating political information when many of the key cases were decided. But there are dangers in simply transposing some of those principles to all speakers in the digital context. The public interest defence in defamation illustrates this point. If the courts use professional journalistic standards as a benchmark that all publishers are expected to follow in order to rely on the defence, then there is a danger that the amateur or smaller scale speakers online will get limited protection.
The first reason is obvious – it may not be reasonable to expect the amateur to meet the standards of a professional investigative journalist. For example, when verifying a factual claim, we may not expect an amateur blogger with a small circulation to contact a politician for comment prior to publication. Furthermore, most amateurs will not have access to the type of legal advice that journalists (under the earlier law) had relied on to make the article Reynolds-proof.
However, the courts need not use the standards of the professional journalist as the benchmark against which to judge speakers. The need to interpret the defence flexibly and take account of context was emphasized several times when the Defamation Act was debated in Parliament. Different standards from the professional could be required when an amateur/small scale speaker wishes to rely on the public interest defence. This may seem obvious, but the trickier part is determining what ethical standards the amateur blogger and publisher can be expected to meet.
The second difficulty with applying the standards of the professional media is that it overlooks the new methods for newsgathering and producing knowledge that now occur in the digital media. We tend to think of articles published in a printed newspaper as a finished product. The journalist has undertaken research behind closed doors and then only after refinement (fact checking, etc) the product is presented to the audience. That mode of thinking is reflected in the old Reynolds test – the journalist was expected to go through various processes prior to publication.
With the digital media, the production of knowledge can occur through a more transparent and collaborative process. For example, one person may openly post raw data, allowing others to take on the task of verifying various points and commenting on each other’s findings. In his book, The Wealth of Networks (2006) [pdf], Yochai Benkler provides a good example, where a network of people found flaws in electronic voting machines. The initial publications are not a finished product that makes a claim to accuracy, but the opening of a conversation. The knowledge produced is not contained in a single publication, but is the product of the ongoing publication. To apply Reynolds-type standards here may discourage people starting such a conversation.
What I describe here is an ideal. I don’t buy into the cyber-utopian view and I accept that such transparent exchanges can in some cases result in misinformation. We should also be concerned that such transparent conversations could have pretty serious consequences for people’s reputations and privacy. But the point is that some publications that serve the public interest do not present themselves as finished products and will not have undergone the types of processes found in traditional journalism. The approach taken by the court to the new public interest defence may determine whether people feel free or inhibited from participating in such conversations on controversial topics.
The choice is not simply between two extremes of either (a) expecting all speakers to fulfill the standards of the professional investigative journalist or (b) allowing a free-for-all for bloggers to ride roughshod over reputations in the hope that some useful knowledge is produced along the way. The Defamation Act 2013 envisages a mid-way and flexible approach to be taken. It remains to be seen how this will be achieved and whether a new balance will be struck between free speech and competing rights. This raises difficult questions, but the key point is that the courts cannot simply apply the standards formulated in the previous century without taking on board the new ways that certain types of digital communications can serve the public interest.
Jacob Rowbottom is a Fellow of University College, Oxford.
Jacob, the basic thrust of the above sounds clearly right. What springs to mind is a concern about legal certainty: newspapers apparently already complained that Reynolds was not predictable enough in operation. If s 4 requires the courts to develop several different levels of the test, tailored for different kinds of communications and different types of ‘speakers’, there is an obvious risk that the law will become extremely complex. (This doesn’t mean that you’re not right, just that this approach will give the courts a real humdinger of a task). I wonder how many different ‘grades’ courts might need to develop, each demanding a different standard of care from the ‘speaker’. I think you’d need two ways of mapping it, too. One axis concerns the status and experience of the speaker, and might go: (1) professional journalist (2) professional blogger (‘Guido Fawkes, Tim Mongomerie) (3) regular, but amateur blogger (like us?) (4) no experience.(Obviously you could divvy those up differently and have far more, fine-grained categories). But there is also the second axis which is concerned with the nature, not of the speaker, but of the output: on that axis you could have: (a) online newspaper article (b) ‘paid-for’ or professional standards blogpost (e.g. one that is moderated, like here or the UK Con Law blog (c) non-moderated blogpost; (d) comment on a blog post; (e) posting on social media (facebook entry/tweet). (Again, I’m sure you could multiply examples). Point being, a professional journalist (1) might do any of (a)-(e) and presumably we expect less care from a journalist in their tweets than in their proper articles. And so on.
Thanks for your comments Gavin (I only just saw them, weeks later). I think you are right about the dangers of uncertainty and there cannot be separate tests to accommodate every contextual variable. Maybe there could be a rough distinction relating to a category of speakers (just as the law of negligence demands higher standards of care in relation to some professionals). But I accept that this is not without problems. My main point is that if the defence carries on along the Reynolds lines, then it will in practice give limited protection to various non-professional speakers. That is something we need to be aware of. If it is thought that a single standard of conduct is appropriate for all speakers (avoiding any distinctions), then maybe it should be set at a level lower than those normally expected of professionals – though this might be thought to offer too little protection to competing interests. Ultimately, difficult choices need to be made about how to strike the right balance, and (given changes in digital communications) the balance may be different from that struck in the previous century.