No One Expects The IRB (Update x2)

When four scholars set out to show that “grievance studies” scholarship was gibberish, it was an astounding success.

The further one strays from reality, the denser the word salad. It’s not that they’re on to something profound, but they’re on to nothing and need bigger, more meaningless, jargon to make it feel as if it’s not complete gibberish.

How far did they stray from reality?

…a 3000 word excerpt of Adolf Hitler’s Mein Kampf, rewritten in the language of Intersectionality theory and published in the Gender Studies journal Affilia.

And that’s only outlandish if you accept the value of studying rape culture at dog parks (spoiler alert: dogs do not first obtain consent). The initial reaction of the punked was to rationalize why the publication of utter nonsense failed to prove the field of study was utter nonsense.

At the same time, though, any field can be hoaxed if you lie about the data you’ve gathered and hide your true intentions: The entire system depends on good faith and honesty. When people break it, by submitting bad-faith arguments or by manufacturing data, the system is not well-equipped to catch it.

Well sure, rape culture at dog parks was a fake data problem, not a foundational failing because the core concept is absurd. They’re dogs. But given the wealth of postdocs whose lives and careers depend on them not being jokes, the billions of dollars that universities spend, and take in, to promote a field of scholarship that barely exists, if at all, it wasn’t enough to defend its honor from the hoaxers. They needed revenge. They needed to burn the heretic at the stake.

Peter Boghossian, a professor of philosophy best known for his involvement in the “grievance studies” hoax papers, is now in trouble with Portland State University’s Institutional Review Board (IRB), which has accused him of violating its policies regarding the ethical treatment of human test subjects in the course of his experiment.

“Your efforts to conduct human subjects research at PSU without a submitted nor approved protocol is a clear violation of the policies of your employer,” wrote PSU Vice President Mike McLellan in an email to Boghossian, according to Areo.

For those unaware, colleges require that academic studies be approve by IRBs, who assure that the studies are ethical. On the far edge, they don’t want a Professor Mengele doing experiments. Closer to home, B.F. Skinner would get some unpleasant comments and limits. Shocking, right? But it’s left to a college’s IRB to sanction experiments so that no one is harmed.

But  was what Boghossian and the others did an experiment? Was it subject to IRB approval?*

This charge makes Boghossian sound like Dr. Frankenstein. But the “human subjects” in question are the peer reviewers and journal editors who accepted Boghossian’s hoax papers for publication. Their reputations may have suffered as a result of being duped—and they were indeed unwitting participants in the experiment—but their physical well-being was not compromised. Moreover, it may not have been obvious to Boghossian and his co-conspirators that research conducted outside his field, bearing no formal connection to Portland State University, was still subject to IRB approval.

This begs the question, however, as it assumes IRB approval was needed because the IRB says IRB approval was needed. Was this really an experiment? Were the journal editors and peer reviewers “human subjects”? In a world where gibberish passed for profundity without comment, where it’s good enough to get someone a really cool hood, why not?

Compliance with institutional review boards is a common facet of academic life; universities require assurances that professors are not conducting unethical experiments that harm their subjects. (Of course, since IRBs are change-averse bureaucracies, they often construct unnecessary hoops for researchers—see this amazing Slate Star Codex post detailing Scott Alexander’s futile attempts to gain IRB approval for a perfectly benign study.) But it’s possible that Boghossian didn’t realize he needed IRB approval for this project, since it had nothing to do with his job. (Portland State, I gather, considers all research conducted by its employees to be subject to IRB oversight, so this probably isn’t going to work as an excuse.)

The problem for Peter Boghossian is that he’s a prof at Portland State University, and so if the PSU IRB says it’s an experiment, says he violated protocol by neither seeking nor obtaining approval, then his job is on the line. Whether the IRB is right is too squishy a question to be subject to objective determination, but that the IRB says Boghossian’s “experiment” required its approval is all that matters, at least as far as Boghossian’s continued employment is concerned.

Then again, had Boghossian sought IRB approval of his “experiment,” would he have gotten it?

On Twitter, Lindsay opines that seeking the IRB’s approval was impossible—administrators would not have sanctioned the project, and this would have risked blowing their cover.

Musa al-Gharbi, a sociology fellow at Columbia University and director of communications for Heterodox Academy, tells me he thought it “highly plausible that had they followed standard protocol, the IRB board would have rejected their proposal for political/ideological reasons.”

Not only would the IRB have rejected the effort designed to undermine the integrity of an entire field of “scholarly” endeavor, but even seeking approval would have expose the effort. Or maybe the IRB would approve it, but do so with an announcement so that the journals targeted would have been alerted that they were about to be the target of a hoax. That would have worked well.

Robby Soave at Reason sees the significance of “Sokal Squared,” as this hoax has come to be known, as limited.

While I’m far from convinced that submitting elaborate hoax papers was the best way to draw attention to the scholarly deficiencies of “grievance studies,” it would be troubling on academic freedom grounds if Boghossian lost his job simply because he did not ask his university for permission to conduct this little experiment.

Whether this hoax devastated a field grounded in empty jargon or merely shows deficiencies around the edges of what purports to be scholarship, may be worthy of dispute, but for Boghossian, the price could be his job.

Had Boghossian known that his “experiment” required IRB approval and went ahead anyway, he brought this problem upon himself. That the IRB wouldn’t have approved it, or would have blown his cover, is, unfortunately, part of the deal in academia these days. But the vagaries of whether this was subject to IRB approval at all is ironically linked to the hoax itself. In a world of gibberish in, gibberish out, how does one attack the gibberish without the Gibberish Inquisition knocking on the door?

*Edit: Inside Higher Education provides the IRB’s explanation for its assumption of “jurisdiction” over the “experiment”:

The IRB determined that the project, as discussed in Aero, was research since it was “a systematic investigation designed to develop or contribute to generalizable knowledge.” The determination letter continued, “The publicly available information about the project clearly indicates an iterative and systematic approach to performing the work, with an intention of generalizing the results.”

Does this make the case, or is it more word salad?

Update: Via this twit from Boghossian, see the attached letter from Harvard’s Steven Pinker to the PSU IRG.

Update 2: At FIRE, Robert Shibley provides a little history about the National Research Act, which gave rise to IRBs.

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.

And like all good things, mission creep took it from there.

42 thoughts on “No One Expects The IRB (Update x2)

  1. delurking

    “(spoiler alert: dogs do not first obtain consent)”

    Sure they do. It is obvious from their behavior that they consent.

        1. Casual Lurker

          “Woof means woof.”

          Tell that to some dogs, when my German Shepperd goes “Voof”!

          Fortunately, for him, most bitches dig the accent.

  2. delurking

    On a more serious note, the experiment looks to me like a pretty straightforward case of human subjects research. I don’t think there is a reasonable argument that it should not fall under an institution’s IRB review. When academic economists send out a bunch of fake resumes with ethnic names, in order to do research on bias in hiring, they get IRB approval. Mr. Soave’s reference to “physical well-being” is irrelevant; neither the Stanford prison experiment nor the Milgram experiment harmed anyone’s physical well being, but there is wide agreement that those experiments were harmful. IRBs came about to protect human subjects from experiments that weren’t obviously dangerous.

    I am surprised to read that Prof. Al-Gharbi is quoted as saying what he says. If that is true, then the field of academic philosophy probably has a problem.

    It is true that this experiment almost certainly couldn’t have been done with IRB approval. Perhaps IRB policies should be different, but as they stand today it is appropriate for PSU to investigate.

    1. Ron

      “the experiment looks to me like a pretty straightforward case of human subjects research. I don’t think there is a reasonable argument that it should not fall under an institution’s IRB review.”

      About as strong an argument as the IRB made. After all, isn’t it ultimately just Torquemada’s call?

    2. Patrick Maupin

      “When academic economists send out a bunch of fake resumes with ethnic names, in order to do research on bias in hiring, they get IRB approval.”

      An interesting example, since it’s practically indistinguishable — sending solicited words to external people who asked for them. If an IRB wouldn’t disapprove this, then they have no reason for disapproving the other, and if they have no reason to disapprove, they you’re normally in forgiveness/permission territory, where a lack of forgiveness where permission should have been granted is the hallmark of butthurt petty tyrants.

      “neither the Stanford prison experiment nor the Milgram experiment harmed anyone’s physical well being…”

      And, off in the weeds. In no rational sense is paying $50 to a starving student who then temporarily comes under your physical control in your space comparable to sending off some paperwork to random strangers who were soliciting exactly such paperwork.

      The resume experiment is one that can be done without any institutional affiliation; in fact, it more closely resembles an EEOC sting operation than it does the other kind of experiment, where institutional affiliation and the promise of easy cash can be used to lure unsuspecting students in for whatever abuse the experimenter can dream up.

      1. delurking

        Ron and Patrick, you both make good points. My only point is that it is unjustified to accuse the university of retaliating against the professor by means of an pretextual IRB investigation. I don’t actually know enough to have an opinion on whether this particular research crosses the line. It seems to me to be a closer call than the resume example, though. The total dollar value of services acquired under false pretenses is probably relevant to IRB review, and it probably exceeds $10K for this one.

        Patrick, your second point misreads mine, though. It is a response only to the idea that IRBs should be limited only to considering physical harms. They are not, nor should they be.

          1. delurking

            Well, we all draw our impressions from what we have read about the affair. It appears to me that Prof. Dawkins has read something that implies that the a University committee is considering punishing Prof. Boghossian for falsification of data. That is news to me, and if that is true, I agree with Dawkins that that is ridiculous.

            1. SHG Post author

              It appears the gravamen of the IRB’s beef is that the dog park rape culture post faked the data, as he didn’t actually interview 10,000 dogs. What else would an IRB do with such a liar?

            2. Patrick Maupin

              The IRB’s response here is practically indistinguishable from that of corporate America when faced with white-hat hacking: “You have put the system at significant risk by showing its inadequacies with dealing with bad actors. You knew this would cause problems for people in authority, but did it anyway; this makes you a bad actor yourself.”

              Pulling back the curtain, or pointing and laughing at the naked belly protruding over the tiny dick — these things have always been unforgivable.

  3. Patrick Maupin

    The Spanish Inquisition is so last millennium. All hail the new and improved Iberian Review Board!

  4. RedditLaw

    A thought experiment:

    If the journals in question had rejected the submitted gibberish, and Prof. Boghossian had publicized that the journals were actually subjecting proposed articles to rigorous review, thus complimenting the fields that he had intended to criticize, does anyone here think that the IRB would now be after his job?

    Anyone?

    Bueller?

      1. RedditLaw

        It’s because I am ensconced in The Paranoid Style of American Politics. Come and join us. It’ll be fun.

      2. Onlymom

        Just think the media is getting another chance tonight to show the public tbey can and will expose bs when they see it. Using the oval office speech tonight if darth cheeto. Have it on a 5 minute delay to do a quick fact check and display the truth under the lie in text in real time

    1. Patrick Maupin

      PSU-IRB MOUSE!
      (What the fuck?)
      PSU-IRB MOUSE!
      (What the fuck?)

      Forever let us squelch discussion
      NOW! NOW! NOW! NOW!

      Go along to get along
      and burn an effigy!
      P-S-U I-R-B M… O…. U. S. E!

        1. Patrick Maupin

          Wait, I’m starting to think like McLeod?

          I need some comfort food to graze on while I mull over the implications of this.

          Are Peeps in season?

      1. B. McLeod

        Jacking up Boghossian,
        So everyone can see,
        PSU (U can’t think HERE),
        IRB (B. . .more careful next time),
        M….O…U…S…E

  5. Julia

    If the board treats the dog park hoax as research, what’s the difference for them between research and a hoax?

    1. SHG Post author

      Good point. Of course, had someone proposed the dog park “research,” they would have approve it provided they offered palliative care to the victims.

  6. szr

    Counsel for the Defendant moves this Court to dismiss the IRB’s complaint.

    The IRB complaint alleges Defendant’s actions constituted “research” because they were a “systematic investigation designed to develop or contribute to generalizable knowledge.” However, this statement is unsupported by evidence and is merely conclusory.

    Portland States University’s Human Subjects Research Review Committee Policy defines “research” as a “systematic investigation designed to develop or contribute to generalizable knowledge.” HSRRCP at 1.3. The Policy defines a systematic investigation as “one that applies a defined set of questions or steps across a number of individuals or points in time in order to answer a research question.” id.

    The IRB failed to describe any “defined set of questions or steps” Defendant is alleged to have taken. See id. Because IRB failed to allege any “defined set of questions or steps,” the Board’s complaint fails to allege Defendant engaged in a “systematic investigation.” This failure is critical to the IRB’s complaint.

    The IRB has jurisdiction over “research” as defined by the Policy. The IRB failed to allege that Defendant engaged in a “systematic investigation” within the meaning of the Policy, thus the complaint fails to allege that the Defendant’s conduct constituted “research” for which the Board has jurisdiction.

    For the reasons described above, this Court should dismiss the IRB’s complaint.

  7. Casual Lurker

    There’s a lot I could say about IRBs, but won’t. How they’re being used to shield an institution from embarrassment or scrutiny has already been covered.

    For IRBs overseeing medical research, I expect them to be a pain in the ass. The “informed consent” and HIPAA issues alone put the brakes on many proposals. But a non-med IRB asserting jurisdiction over something akin to a fraud investigation, conducted outside of Prof. Boghossian’s official duties, seems similar to the kind of mission creep I’ve mentioned before. To wit, the case of Mats Järlström, the “Oregon Man Fined for Doing Math Without a License”,* who was represented by the Institute for Justice.

    Assuming Prof. Boghossian’s employment contract didn’t explicitly prohibit his actions, and he winds up being severely sanctioned, it seems to this non-lawyer like a lawsuit in waiting. (Looks like SZR has already drafted the complaint).

    If not SZR, then maybe the IJ can help the Prof. out. (I fully expect a response along the lines of “Of course they will… They’ll ask him which way he came in!”)

    *Reason
    Hit & Run blog
    June 2, 2017

    1. SHG Post author

      I hate to be the one to tell you this, but SZR’s complaint was dismissed, with costs and attorneys fees, on the motion to dismiss. It was a sad day for all involved.

    2. Patrick Maupin

      “Doing math without a license?”

      The bullshit headlines even the nominally good guys dream up for page clicks these days…

Comments are closed.