Portland State University professor Peter Boghossian, who was one of the three participants in the project commonly known as “Sokal Squared” or the “Grievance Studies Affair” last year in which several academic journals were fooled into publishing fabricated papers, is now facing discipline for his involvement. On Saturday, his collaborators, James Lindsay and Helen Pluckrose, revealed PSU’s charges against Boghossian in the online magazine Areo. (Pluckrose is Areo’s editor-in-chief.)
According to Lindsay and Pluckrose, the PSU administration told Boghossian in late December that his “efforts to conduct human subjects research at PSU without a submitted nor approved protocol is a clear violation of the policies of [his] employer.” While further discipline might be forthcoming, he has meanwhile been “restricted from performing any human-subjects research, at least until [Boghossian] has completed an appropriate training protocol ‘to be identified and proctored by the Assistant Vice President for Research Administration.’”
The controversy over PSU’s decision revolves around a requirement of academic research that is little known outside of the academy: the need to have research on human subjects approved by the Institutional Review Board at an academic institution before the research can begin.
IRBs are required by federal law. In 1974, Congress passed and President Nixon signed into law the National Research Act, in the wake of the exposure two years earlier of the infamous Tuskegee Syphilis Study, in which hundreds of black men who researchers knew tested positive for the disease went without proper treatment for up to 40 years, without their consent. IRBs were intended to prevent this sort of abuse from happening again by ensuring that research that could affect human subjects was approved by a panel of fellow scholars who were tasked with addressing any physical or psychological harm to human subjects by requiring modifications to, and sometimes denying, research plans.
Over time, the use of IRBs has become increasingly commonplace, and seemingly required, even for social science research or experiments that have a far less direct effect on the humans who might be involved. As Columbia law professor Philip Hamburger, a prominent critic of the current role of IRBs, has pointed out, even oral history projects and opinion poll research, which simply consist of asking people for their own stories or opinions, can be subject to change or simply forbidden by IRBs. (Oral history, at least, was relieved of this burden by federal regulatory changes that took effect just last year.)
Particularly when removed from the medical context, it becomes all too easy for some fundamental IRB rules — such as the requirement that studies be done only with the informed consent of all human participants — to fail to work well. As Lindsay and Pluckrose point out, the Grievance Studies Affair is one of these situations, as “it is impossible to conduct a valid quality assurance investigation, which this audit was, after informing those being audited that they’re under examination.” Assuming it’s correct to characterize the journal editors as subjects of an experiment who needed to be protected from its potential physical or psychological harm, the IRB process would at the very least have required that the authors inform all of the potential “subjects” that faked research papers were coming their way. Truly “informed” consent might have required rather more specificity than that. It doesn’t take a scientist (or a whole group of them on an IRB) to understand that such a restriction would make this particular research effort pointless, but PSU nevertheless determined that the research violated its rules and was worthy of discipline.
Some critics have also suggested that Boghossian and his fellow authors were engaged in academic fraud. For example, Brandeis professor Joel Christensen told Inside Higher Ed that the trio “did misrepresent themselves, they did falsify their evidence and they did commit a serious infraction of research misconduct.” Boghossian and the other authors of the faked papers have argued since the premature exposure of their project that they intended to reveal that they had faked the research at its conclusion, which makes sense given that there would otherwise be no reason to believe their results. While an element of deception was necessarily involved, an academic fraud analysis doesn’t make much sense here: The authors could not benefit in the traditional way from faking research results, as they were using pseudonyms and were not even scholars in the actual fields in which they submitted. And with the papers’ inevitable retraction, the potential damage to actual science from those who might credit the falsified results would be eliminated. These factors may be why even Christensen said, “We doubt that this rises to the level of an offense warranting termination. And the bar for professional sanctions should be very high in the case of an academic with academic freedom.”
Professor Boghossian would not be the first person, nor will he be the last, to be adversely affected by the application of rules that make no sense in a specific situation. Nevertheless, this highlights the obvious problems for social science research presented by some IRB rules, especially when that research is geared toward determining the honesty, true attitudes, and/or competence of its “human subjects.” (The numerous studies on employers’ differing reactions to resumes that have differing racial or ethnic identifiers, but that are otherwise identical, are obvious examples.) When it comes to this type of research — in which Boghossian, Pluckrose, and Lindsay were certainly engaged — it’s hard to avoid the conclusion that if the rules forbid it, it’s the rules, not the researchers, that have gone wrong.