I became disenchanted with the idea of e-books when Amazon reached into scores of Kindles and removed copies of (of all possible books) 1984 and Animal Farm. The notion that a major company had the power to deny access to any content they deemed problematic simply presented too many visions of reactive, totalitarian control.
I never considered that those very concerns might apply to the publishers of scientific research, who – in this age of online-only publications – have the power to remove properly vetted research articles at their whim. Now, it is not a nightmare vision but a grim reality.
The publishing company in questions is Frontiers, previously known for its pioneering approach to publication: instead of making money by charging potential readers to access the articles ($30 per article, for those not fortunate to have access through an institutional subscription), they make money by charging the researchers a flat fee to share their research. Although I was a little hesitant when I first blogged about a Frontiers article, I have since become more accustomed to the researchers-pay-to-remove-barriers-to-access idea (which I have encountered as an option at some traditional journals). I have also been part of the first round of a Frontiers peer-review process (the scientific version of quality control), evaluating the methods and conclusions of a developmental psychology submission. I was beginning to enjoy the open access to such a wide range of journals, and recent blog entries The Power of a Bedtime Story and Let There Be Meditating Light were based on Frontiers publications.
The article in question was written by Stephan Lewandowsky and is now known by its short name, Recursive Fury. It is right up my ally, investigating the psychology of belief, specifically the attitudes of those who rather vehemently object to any claims that climate change is caused by human activity. In this case, a number of bloggers and commenters had been objecting to the an earlier article by the same authors that investigated what other beliefs climate change deniers hold. Their online reactions – accusing Lewandowsky of running a scam, lying, and faking results in one way or another – were in themselves a treasure trove of what climate change deniers believe. So, Lewandowsky and his colleagues investigated them. Recursive Fury is the result of some carefully conducted Google searches to identify sites where the article was being contested, and analyzing what kind of logic and conspiracy theories were being aired.
This article successfully passed peer-review, with a careful and rather interesting qualitative analysis of the comments and how they match certain characteristics of “conspiracist ideation” – what we might more colloquially call paranoia about secret government thoughts or a conspiracy theory – and clear acknowledgement of the limitations of these particular researchers investigating the responses to just this one article. But the investigation strikes me as an entirely reasonable and very informative use of online commentary, which provides us with a treasure trove of insight into people’s thinking.
The bloggers and commenters themselves, of course, we not so pleased to see their words interpreted in this light. They complained rather vehemently to Frontiers. And Frontiers responded with an investigation that concluded that there were no “academic” or “ethical” problems with the study, but “the legal context is insufficiently clear” and therefore the article was being permanently retracted. In other words, the “public review” of a few disgruntled online opiners outweighed the peer review by fellow scientists.
The initial backlash against the retraction led Frontiers to provide a more detailed rationale for its decision, declaring “the article categorizes the behaviour of identifiable individuals within the context of psychopathological characteristics” and an explanation that, essentially, privacy concerns overrule all else in their publishing.
Recursive Fury is now being made public by the University of Western Australia (along with the story from the perspective of Lewandrowsky itself), so lets consider just how the authors did in regards to privacy.
All of the substantive claims in the article, about the specific ways the commenters displayed facets of conspiratorial ideation, are supported by specific quotes. Those quotes are then backed up by the URL of the website where that comment was originally made. In other words, the authors took care to be open about the specific sources analyzed and providing readers with the option to read the original text themselves, which allows us to see that quotes were not altered or taken out of context. What the researchers and peer reviewers saw as careful documentation, the commenters saw as a breach of privacy – and Frontiers took their side.
Yes, providing the URLs does mean that a reader can copy and paste (this one, perhaps) to see not just the complete text from which a quote was pulled, but also whatever identifying information the commenter cared to provide along with their words. But then, this is the age of Google. To find the source of a particular sentence, a URL is helpful but hardly necessary; you can copy and paste the sentence itself, put it in quotes, and find where it originally came from; I do it to check for plagiarism in student papers all too often. It might not have worked with every quote mentioned, but it led me to several of the sites discussed. In other words, “privacy” (that is, complete anonymity) of the commenters could only have been maintained if the researchers did not provide any original quotes at all. In which case, the paper could not have built a legitimate case for its claims, because no peer reviewer would have been satisfied by vague generalizations without actual quotes and at least the overarching website (the equivalent of contemplatingcognition.wordpress.com, if not the specific articles).
The problem with calling the links to the original comments privacy violations is that the comments were never private to begin with: they were posted to public websites, at least one of which takes pride in its public nature by identifying itself as “The world’s most viewed site on global warming and climate change“. I admit that we are not discussing commenters with the greatest logical skills and/or awareness of reality here, but it’s a bit disingenuous to declare that information freely given to every other random visitor of the website is supposed to be protected from the eyes of the scientists actually being discussed.
When Lewandowsky and his colleagues did have the chance to breach privacy, they very carefully did not. The original article was conducted with surveys provided on climate change blogs, with a note that 5 denial blogs were approached about the survey and declined to host it. This became a point of contention among the objectors:
Within short order, 25 “skeptical” bloggers had come publicly forward (http://www.webcitation.org/6APs1GdzO) to state that they had not been approached by the researchers. Of those 25 public declarations, 5 were by individuals who were invited to post links to the study by [the original research article] in 2010. Two of these bloggers had engaged in correspondence with the research assistant for further clarification. (p. 16)
Personally, I’m curious which of those 25 bloggers were misstating their position. But the article doesn’t reveal that; they didn’t run the risk of outright accusing any specific blogger of lying. In a way, they weakened their own position (that vagueness means they’re still vulnerable to accusations of making things up) in order to protect the identities of the people they are writing about.
So, any complaints about privacy have to be about the specific piece of “within the context of psychopathological characteristics”. It’s kosher to, say, accuse an explicitly named individual (researcher) of fraud, scams, and so on, but verboten to suggest that an online commenter might have a mental screw loose. I’m going to guess that this is Frontiers‘ take, not the commenters’, since it seems unlikely that they would self-identify as being crazy, accept that their comments showed any psychopathology, or grant Lewandowsky and his colleagues any legitimacy in passing judgment on their comments or mental state to begin with. Either way, though, the claims aren’t actually psychopathological. Believing in a conspiracy theory is not a psychological disorder, any more than a religious belief is. Analyzing either kind of belief certainly pushes people’s buttons, but identifying someone as a conspiracy theorist is not all that different, from a cognitive psychological standpoint, than identifying them as believing in a god. It’s a belief they have shown through words or actions. The difference is whether it’s a label they would like to have or not, but neither one is a mental disorder.
Meanwhile, let’s be clear: Frontiers changed its story, from originally stating that there were only some vague legal uncertainties to accusing the researchers of misconduct. “[D]oes not sufficiently protect the rights of the studied subjects” is a violation of confidentiality, one of the central ethical principles of scientific research, and the reason that I have to maintain separate locations to store my consent forms (which have names on them) and data files (which do not). Either the publishers are not aware of that fact, and therefore ignorant of their job, or they are making some rather….unfortunate insinuations.
So far, several editors have chosen to resign in protest over this bending to the “climate of intimidation” facing science today. The practical implications for more everyday researchers are less clear. This issue does not seem sufficient to reject all other research that is published by Frontiers; decreeing all articles from a publisher invalid would take a far more systematic flaw in how articles are accepted. I can hardly say that I would never submit to Frontiers, either, since I was never in a position to in the first place – I don’t have the kind of funding that supports the author-pay model. I am going to be more reserved about the future of Frontiers and its relevance to the future of scientific publishing. Sadly, even online publishing has some costs, and we are stuck with either big-name publishers who charge for the content but can withstand public kerfluffles and lawsuit threats, or open-source publishers who charge the researchers for the privilege of sharing their findings and don’t have enough of a war chest to defend their work. All of which will do nothing but feed the growing disbelief in science, because that science will either be hard to access, or easy to get rid of.
Frontiers in Psychology Editorial Office (2014). Retraction: Recursive fury: conspiracist ideation in the blogosphere in response to research on conspiracist ideation. Frontiers in psychology, 5 PMID: 24683402