This short essay follows a round table organized by the Reuters Institute for the Study of Journalism at Oxford in December 2016. It is a write-up of my contribution to the panel on Investigating big data: collaboration and best practice at the Journalism Festival of Perugia of April 2017.
The Panama Papers were a worldwide sensation in 2016. If we had Academy Awards for journalism, they’d be Titanic and La La Land rolled into one. Beyond the many records the projects broke, observers were struck by the collaborative nature of the effort. Hundreds of journalists worked together for months to produce groundbreaking reporting on tax evasion. The Panama Papers are not alone. The Russian Laundromat is a recent story coordinated by OCCRP, a non-profit. It brought together 60 journalists.1 I had the chance to coordinate a group of over 20 journalists in 2014 and 2015 in The Migrants’ Files, an investigation on the number of men and women who died in their attempt to reach Europe.
Such collaborations were successful. To produce successful stories, the thinking goes, journalists should collaborate more. This syllogism is attractive, but I doubt it’s true. I had the luck to coordinate many unsuccessful collaborations as well. The Belarus Networks failed because of a lack of involvement from a key partner. Turkish Puppets failed because the publication partners could not grasp why the story was important.2 Two or three other collaborations I was part of also failed to reach their initial goal.
Informed by a few successes and many failures, this text argues that we should consider collaborations in journalism for what they are: A tool among others that should be used only when appropriate.
Media companies joining forces is not something new. Back in the first part of the 19th century, newspapers in the United States pooled resources to maintain a pony express service (a kind of Uber, for horse-carried messages). From the 1850’s onwards, they institutionalized their collaboration by sharing a telegraph service. They created a joint venture to buy a steamer boat, itself a new and very expensive technology, that would fetch the mail directly from ocean-going ships arriving from Europe at their first American stop at Halifax, Canada. The contents of these fresh news was then telegraphed (an even newer technology) from Boston to New-York, where each newspaper would run it. That’s how the Associated Press was born.3
A theory goes that this early collaboration was just a cartel meant to put other telegraph lines between Boston and New-York out of business. But it’s not my point. My point is that this joint effort was driven by a necessity to share a piece of equipment, much like some media companies use the printing facilities of their competitors.4
I haven’t found examples of explicitely collaborative journalistic projects in the 19th and 20th century.5 The Muckrakers of the Roosevelt era in the United States, who pioneered investigative journalism as we know it, worked alone or in pairs.6 In 1973, Bob Woodward and Carl Bernstein of Watergate fame did not rush to share Deep Throat’s material with other newsrooms. Europe’s first data-driven investigator, E.D. Morel, did work in collaboration with others as he documented slavery in the Congo in the first years of the 20th century. He used pictures taken by missionaries to complement his data-driven analyses. However, this was never a joint effort - more of an alliance between activists.
Collaboration in journalism is a new thing, which began in the late 2000’s. Wikileaks pioneered the method when it brought together the New-York Times, the Guardian and Der Spiegel in 2010 over the Afghan War Diaries, a database documenting war crimes by NATO troops in Afghanistan.7 Since then, the method spread widely and is now considered normal whenever a major international story is released.
There are good reasons to go into a collaboration. In the case of the Afghan War Diaries or the Panama Papers, a coordinating organization offered exclusive material to newsrooms. The coordinating entity offered technical skills that newsrooms did not possess and pooled the reporting. Finally, it ensured that all partners published jointly, so that the joint publication itself became a news story.
This process is bold and brilliant. It is not, however, especially innovative. Rather, it is the adaptation to journalism of an ideology that favors networks over structures, which permeated the business world a couple of decades ago.
It’s a product of the dominant ideology
The ideal career of the 20th century was to remain loyal to a company for forty years, regularly climbing the hierarchical ladder. Since the 1980’s, this ideal changed dramatically. To be considered successful now means having a large network of contacts that one can activate for a specific project, hopping from one network configuration to the next according to one’s current needs.8
The apparition of the network as the best way to run an enterprise (be it a business or an investigation) fits perfectly with the idea that a system works best when individuals can apply their skills where they are most needed. Instead of building an organization that possesses all the possible skills it will ever need, it’s more efficient to look for a specific person possessing a specific skill for a specific need. The emphasis on individuals, as opposed to organizations, is one of the basic tenets of neoliberalism, the ideology that became hegemonic in the 1980’s and remained ever since.
The conservative nature of journalism ensured that this “new spirit” entered only slowly in the newsroom,9 but it’s slowly creeping in. Private donors lead the pack in demanding that the journalistic projects they fund be collaborative. JournalismFund.eu, for instance, offers grant only to cross-border teams. The same applies to Connecting Continents, another grant program run by the same organization. JournalismGrants, a program funded by the Bill & Melinda Gates Foundation, is less insistent on collaboration but mentions that “teams” should apply. The Volkswagen Stiftung recently asked that journalists team up with scientists.
Donor-funded outlets that operate outside of newsrooms also insist that they are networks or consortia, not publishers. The Organized Crime and Corruption Reporting Project and the International Consortium of Investigative Journalists are the most visible and successful, but many other exist at a smaller scale or at regional levels (Arab Reporters for Investigative Journalism for Arabic-speaking countries, the Balkan Investigative Reporting Network for the Balkan etc.) Public donors also consider collaboration a prerequisite for funding. The European Commission recently awarded grants for journalism where they requested that applying consortia be made of at least four members but wrote that the more members in the consortium, the better.10
My point is not to say that collaboration is bad - or good. But it did not arise in an ideological vacuum. It might be the case that collaboration emerged out of the needs of journalists who found a new way to organize. It might also be the case that the profession is just following a trend from the business world without giving it much thought.
It’s time to be smarter about collaboration
Collaboration is the action of achieving a precise goal with a group of persons without creating a new organization. Everything lies in the “precise goal”. If the purpose of the collaboration is uncertain or changing, the glue binding the team together won’t hold.
To state the obvious, the more interesting the documents that support the collaborative investigation, the more likely it is that the collaboration will be successful. If, on the contrary, the collaboration is not based on existing secret material but builds a story from the ground up, it will fail as soon as the initial goal gets out of view.
This can happen if the topic of the investigation isn’t clear. When we worked on the costs of Fortress Europe with the Migrants’ Files consortium, for instance, we set out to investigate who benefited from policies that prevented men and women from claiming asylum in Europe (walls, fences, electronic gadgets that sniff out people attempting to cross a border etc.) It turned out that “making money from Fortress Europe” was extremely complex to delineate, and harder still to investigate. If everyone’s mission is not perfectly defined, collaborative work becomes arduous.
More commonly, collaborations falter when the initial plan is changed. It happened that we based a collaboration on data that we were supposed to obtain through a freedom of information request. When it did not succeed, the whole project had to be revamped. Projects can also derail upon publication if a media partner misinterprets the documents that are made available by the coordinating partner, for instance. In rarer cases, partners can also fail to carry out their part of the agreed-upon work.
Collaborative projects have costs. Running a consortium requires good project management skills and enormous amounts of energy. The larger the project and the less clear the goal, the more so. Sometimes, these extra costs outweight the benefits of a collaboration and the project would be better done by a single person or a single organization.
Collaboration by itself does not make a journalistic project better. Sometimes, it can seriously hamper it by making it less able to cope with changes in the initial plan. It’s high time for funders to stop requiring that applicants work collaboratively, especially if they require that a project be innovative at the same time. Innovation, by definition, means making use of new and, ergo, untested processes. This implies that the risks of failure are high, thereby increasing the costs of collaboration tremendously. As long as they combine both criteria, they secure the failure of their grantees.
Instead, they should ensure that a successful project can be replicated or translated quickly. To do so, they should spend less money on grants that are bound to fail (the ones that require both collaboration and innovation) and more money on helping journalists find possible post-publication partners.
In a nutshell, this means spending more money on conferences and other get-togethers and - more importantly - on ensuring that a diverse set of people attend them. Diversity means to invite minorities, of course, as well as working journalists who cannot find the time to leave their newsroom - a recurring problem at European conferences.
2. Our story showed how European governments refuse to consider Muslims of Turkish descent as full-fledged citizens, allowing the Turkish government to meddle in their religious and cultural affairs. European-Turks who refuse to be treated as pawns in a geopolitical game are ostracised by both European and Turkish authorities. Most our media partners were fully unable to process a story that involved Muslims who were neither ‘terrorists’ nor ‘moderates’. I blame the lack of diversity in their newsrooms, though plain racism is probably also at play.
3. I read about this history of the Associated Press in The Nation’s Newsbrokers: The formative years, from pretelegraph to 1865, by Richard Allen Schwarzlose, page 80 ff.
5. There’s an exception. In 1976, 18 reporters got together in Phoenix, Arizona, to investigate the murder of a journalist. However, the choice of words in this New-York Times article (‘an unusual experiment in group journalism’, ‘nothing like this multinewspaper investigation of crime in one state has been attempted in American journalism’…) show that the project was a single outlier.
6. From what I gathered in The Muckrakers, edited by Arthur Weinberg and Lila Shaffer-Weinberg.
8. This is a very condensed summary of The New Spirit of Capitalism by Luc Boltanski and Eve Chiapello.
9. On the inherently conservative nature of newsrooms, Jane Jacobs makes a great point in Systems of Survival, page 208. Read it for yourself, it’s a fantastic book.