analysisBy Nick Ishmael Perkins
World Press Freedom Day was observed on 3 May as usual this year. It marks one of the handful of moral absolutes in our civilisation: the desire for journalists to hold the powerful to account and document current events. But it also presents an opportunity for media professionals to reflect on their practice.
Two events in the wake of this year's observance suggest that there is room for science journalists to think more about how censorship affects their work.
First, I chaired a session at the 13th International Public Communication of Science and Technology Conference (PCST2014), held in Brazil earlier this month (5-8 May), which examined how power can be wielded to silence science journalism. Then the recent calls for the closure - or at least a reboot - of the UK Science Media Centre because of its perceived ties to lobbyists have sparked something of a debate.
These two events raised various issues. Perhaps most strikingly, they highlighted the paucity of rigorous studies on the impact of censorship on science journalism. At the PCST meeting, a number of delegates said they rarely thought of science journalism as a field affected by censorship. This is worrying. SciDev.Net works with more than 400 correspondents around the world and many complain that the biggest threat to their livelihoods is getting access to scientists.
Our experience indicates that this is at least in part because of censorship: when our organisation has tried to survey science journalists about how they obtain information, we found various government ministries - particularly, but not uniquely, in north Africa - wanted to vet our questions.
Developed world concerns
This is not only an issue for the developing world. Journalist and academic Kathryn O'Hara of Carleton University, Canada, has observed that the strictures the Canadian government has placed on the research community mean there are now four times as many people employed to restrict access to information as there are to facilitate it. In the United States, an editor at Scientific American claimed last month he was censored during a program on Fox News. 
Wendy Yared, director of the Association of European Cancer Leagues, addressing the World Conference of Science Journalists in Finland last year, said that the control lobbyists have on policy and the media in Europe was a major strategic concern to her.
We shouldn't assume these issues are exclusive to the global North. All the examples above illustrate the pressures science journalists face in developing economies too.
But with limited research, the scale of the problem is unclear. In my experience, when editors from this publication and Nature have been asked to quantify the number of times they had faced censorship, they were hard pressed to give figures. Also the Science Media Centre's chief executive, Fiona Fox, seemed genuinely intrigued by the criticism it received and the data that is meant to support it.
What these anecdotes reveal is that one clear challenge facing those who want to do something about the censorship of science journalism is that we have too few, up-to-date studies on this issue. So why is more effort and resourcing not put into documenting this?
First, it might be because the vast majority of such censorship is quite low-level. When an organisation such as Article 19, which lobbies for freedom of expression, speaks about censorship, it starts by talking about how to manage threats to your life and how to get in touch with networks that will remove you from harm.
This is out of sync with the reality of most science journalists, whose experience of censorship is typically less about death threats and more about sources sitting behind walls of bureaucracy.
There are also the frequent requests by sources for story approval before publication when they agree to talk. Many will argue that this is barely censorship - just caution borne out of experience of inaccurate reporting of complex technical interviews by journalists. This is an understandable point, but to avoid stepping onto a slippery slope, only editors should approve stories.
Self-censorship is also common practice among science journalists. If you anticipate that your attempts to cover a story might result in alienation, or reprimand, from the expert sources you depend on or the media outlet that pays you, then you may have to make a judgement call about that problem relative to society's need to know.
Faced with such decisions, it is unsurprising that journalists will often choose to maintain their professional position and their livelihoods, or to avoid working with certain editors whose stance they disagree with.
A particularly insidious form of self-censorship is the reliance on user-friendly texts supplied by lobbyist or public relations specialists. In these cases, journalists opt for convenience over the protracted effort of investigated objectivity.
In any event, addressing self-censorship does not invoke the same glamorous call to arms that we associate with defeating dictators or corrupt captains of industry.
Tough to define
There was some discussion at PCST about whether embargoed press releases constitute censorship. These materials, typical of academic journals, are where journalists are banned from covering a story until the journal is published.
While I agree that this is the kind of behaviour that makes academic publishing unfashionable and boosts the case for open-access models, I am unconvinced that this is truly censorship. Embargoes are normal in all forms of journalism and, indeed, having some information beforehand can help a journalist to prepare a sound, well-researched piece.
What is clear though is that the apparent low intensity of censorship in science journalism means that developing a working definition of such censorship would be hard. One person's censorship is another one's way of working.
To be fair, the science journalism community, perhaps more than most fields of journalism, would consider itself vigilant about its independence.
They need to be able to navigate the perfect storm of technical disciplinary expertise, academic ambition and investor interests to get to the heart of a story on a daily basis.
The problem though is that journalism is itself like the scientific method: it is not acceptable to practise it well only some of the time.
The challenge here is to generate support and guidelines that address the kind of censorship that is most typical for science journalists.
This is especially important for journalists working in emerging economies where the stakes are so high, as is the opportunity for their coverage to enable local people to leapfrog in terms of technology and scientific application.
This brings us to the problem of who will take on this challenge; a difficult question if we accept that there is currently not much agitation on the subject within science journalism.
The first step may be for science communication researchers to attempt to quantify the social and economic costs of this 'low-level' censorship. If they do this, science journalists will have substantive evidence to reflect on when World Press Freedom Day next arrives.
Nick Ishmael Perkins is the director of SciDev.Net @Nick_Ishmael
 Dominique Mosbergen Science editor: Fox News told me not to talk about climate change (Huffington Post, 30 April 2014)