One of my intellectual heroes is Alan Sokal. Sokal got a bit sick of the 'fashionable nonsense' coming out of Literature studies departments in the 90s and sought to expose them through a hoax. He wrote a paper which was called Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity" which argued, among other things, that gravity was just a social construct.
Sokal submitted the article to a top literary journal where it was published. Three weeks later he revealed the hoax.
In 2017, three authors conducted a similar hoax on what they believe is the fashionable nonsense of our time, namely 'critical theory' scholarship. The recent hoax has been dubbed the 'Sokal squared' (due to the number of fake papers they had accepted) or the 'grievance studies affair'.
In both cases the issue of the ethics of deceiving a journal editor has been raised. However, unlike the Sokal case, in this case one of the hoaxers has actually faced censor from his university. Portland State university has censured professor Peter Boghossian over the affair and has banned him from carrying out research involving human subjects.
But if journal editors are human subjects, then we are faced with a rather difficult problem. Between Sokal and the most recent hoax there have been a number of studies which looked into the issue of predatory publishing or publishers with low standards. Wikipedia actually has a page dedicated to "scholarly publishing stings". It includes such things as a 2012 maths paper which was randomly generated. This paper was intending to show the issues with predatory journals (I have written about these here).
Another paper from 2007, by Thomas Witkowski's attempted to show how poor quality articles on topics like NLP were being published in ostensibly high quality journals. His fake paper was not only published but the editors loved it so much they even helped the author to expand on it.
According to bioethics researchers then, are all these studies illegitimate? The original hoaxer, Sokal, disagrees and has argued that:
You may not have sympathy for grievance studies hoaxers or even for Sokal but decisions like these can have unseen consequences.
common sense suggests that something has gone seriously awry here, when rules initially written to protect subjects in biomedical research from physical harm -- and later extended to social-science research, where the harm could be psychological -- are applied blindly and literally to an "audit study" aimed at testing the intellectual standards of scholarly journals.
If it is wrong to hoax journal editors to expose pseudoscience or other nonsense, because it is unethically doing research on human subjects, is it also wrong, for example to send, as Ross Thornburn recently did, fake CVs to job companies in order to expose discriminatory hiring practices. In an excellent study, Ross shows that if you are white and trying to get a job in Asia, you have around a 64% higher chance of getting employed than if you are black.
Thornburn didn't, to my knowledge, have ethical approval to carry out this research. And he is not the first to have carried out research of this kind. Should we see all such research of this kind as misleading potentially vulnerable 'research subjects', or is there an argument that the good this kind of research can do outweighs the possible risks?
Comment from Robert Lowe (of TEFLology fame)
ReplyDeleteHi Russ,
I have reservations about the "grievance studies affair" for many more reasons than the ethics of it, but I don't think you are representing the ethical issues fairly.
Your post seems to raise a false dichotomy – can you do this research, or can you not? You use the example of Boghossian to show that people who do this kind of research may be ‘censured’, however as you note later in the post Boghossian was censured not for doing the study, but for doing it without approval from a research ethics board.
A research ethics board is there to assess the potential harm from research involving human subjects. In the case of this hoax, the human subjects were not just the editors, but also the reviewers. And was there harm? Well, potentially. Some of the reviewers were graduate students who took time out from their studies to review the hoax papers. That could have both psychological and financial consequences. Or maybe not. Who knows?
In any case, the job of a research ethics board is to check study design in advance for approval because they may catch a potential source of harm that the researchers missed. Failing to get ethics approval is punished because it’s a disincentive to engage in reckless and potentially harmful behavior. Whether you actually cause harm or not is beside the point – you COULD have, and that’s enough. You don’t avoid a speeding ticket by pointing out that you didn’t actually crash, and you don’t avoid a rebuke from your ethics board because you didn’t actually hurt anyone. You are being punished for acting recklessly and potentially endangering people.
Boghossian was found to have engaged in research on human subjects without approval, and has been banned from engaging in such research until he completes a refresher course on research ethics (which doesn't sound like an especially onerous task). That seems perfectly reasonable to me, even lenient.
None of this implies, however, that you can't carry out the kind of study he did. It just means you have to follow ethical practices as mandated by your research ethics board before doing so. Deception in research studies is fine as long as it is believed that the data that could be obtained would outweigh any harm to the subjects. Boghossian should have got research approval first (which probably would have been granted), and then perhaps the board would have recommended further measures such as a follow-up debriefing with the research subjects. You say in the last line of your post “is there an argument that the good this kind of research can do outweighs the possible risks?” – Exactly! Yes, there is, but you have to actually make that argument before you start.
Rob
Hi Rob,
DeleteThanks for the reply.
Obviously I don't think my post raises a false dichotomy. I ask whether any research can be legitimately done in which humans are deceived and point out examples of where this has been done to good effect (eg with CV studies).
You then suggest that because it was not just editors but also "grad students" who were hoaxed that this was somehow worse? I'm not sure about this point though as you seem to back track on it at the end of the para...so apologies if I have misunderstood.
I would say that grad students don't need any special protection and if they can't spot a hoax then they're probably wasting their money at grad school. I'm also curious how you know they were grad students? I don't remember seeing this detail.
You say that "Boghossian should have got research approval first (which probably would have been granted)". How do you know he would have been granted permission? I personally have no idea but seriously doubt that they would have got permission to do this study had they asked.
regarding his punishment, -I have little time for Boghossian myself, and like you see some issues with the study. That said, if he believes that he has done nothing wrong then then it would follow that he would not submit to they retraining course, no matter how lenient.
Thanks again for the comment.
Russ
Hi Russ,
ReplyDeleteInteresting post 😊
I think you miss Rob's main point: it's unethical to do research on human subjects without the approval of the ethics committee of your university. Whether the research brings as valid results or not, it's still unethical if it hasn't been approved or if you haven't sought approval.
You also latch onto somewhat irrelevant points, like the example of student reviewers. It's just an example. Maybe some were and maybe some weren't. I don't think it has much bearing on Ron's main point.
And to follow your line of rebuttal, how do YOU know he wouldn't have got permission had he asked for it?
Anyway, thanks for an interesting post. And keep up the good work