Pinker Said What Now about Research Ethics?

Since I’ve been asked about this on Twitter….

Today, my colleague Carl Elliott posted a short note at his blog that surprised me. By way of background, I have long supported Carl’s quest to bring justice to bear in the Markingson research ethics scandal; I’ve written essays on it, participated in leading a call for the external review, etc. I’ve long shared Carl’s disgust at how hard it has been to get justice in this case.

Anyway, Carl today posted this at his blog:

Forget Tuskegee. Forget Willowbrook and Holmesburg Prison. Pay no attention to the research subjects who died at Kano, Auckland Women’s Hospital or the Fred Hutchinson Cancer Center. Never mind about Jesse Gelsinger, Ellen Roche, Nicole Wan, Tracy Johnson or Dan Markingson. According to Steven Pinker, “we already have ample safeguards for the safety and informed consent of patients and research subjects.”  So bioethicists should just shut up about abuses and let smart people like him get on with their work.

I was surprised by reading this at Carl’s blog, because Steve Pinker has been extremely supportive of all my work, including on prenatal dexamethasone for CAH, which is unquestionably an ethics travesty where the rights of patients and research subjects are concerned. Indeed, before I mount what will be seen as something of a defense of Pinker, I should disclose that Steve Pinker has for many years been a key supporter of my work; most visibly, he supported my Guggenheim application and blurbed Galileo’s Middle Finger, the book that came out of my Guggenheim, the book where I tell the horror story of prenatal dexamethasone for CAH. (You can read the short version at the NYT review of my book.)

I guess I could even call Steve a friend, considering that he’s drunk my husband’s scotch and complimented our kid on having “extraordinary parents” after Steve learned from our son that we let him sleep in a refrigerator box with his two pet rats for six weeks when he was little. (I think “you have extraordinary parents” was meant as a compliment; he didn’t turn us into child protective services anyway.)

So to the point: Contrary to Carl’s post, Steve doesn’t actually say to forget any of those research ethics travesties. Nor does he say forget about those who were harmed.

What he does say—what is clearly the main point of the essay—is that so-called ethicists often predict terrible things will come out of new technologies, and they are often wrong. Most of his op-ed focuses on ethicists who try to stop the development of new technologies not out of concern for research subjects but out of concern that they will lead to “brave new worlds.”

As an historian of medicine and science, I’d have to agree that people have often been wrong about the supposed doom that will befall us when we achieve new biomedical technologies.

Carl truncated the one sentence about research subjects. Here’s what it says in full, with the part that was cut off now underlined:

“Of course, individuals must be protected from identifiable harm, but we already have ample safeguards for the safety and informed consent of patients and research subjects.”

So Steve is not suggesting rolling back safeguards, so far as I can see, although he doesn’t advocate having more, either—which some of us would advocate. (For my part, I’d advocate neither more nor less per se, but much more effective. I’m not sure how to get that.)

And Steve’s obviously factually wrong about there being ample safeguards, given what a lot of us, including Carl and I, have found—although I guess, in the end, that’s a judgment call based on what you see at a population level as acceptable risk to informed consent and safety. A lot of us (myself included) might set the acceptable level very, very low. Steve seems comfortable where it is. He might change his mind if he did the kind of work Carl and I do.

Anyway, I suggest folks read Steve Pinker’s original op-ed for themselves. I rather wish he had not thrown in the issue of research subject protection with his main point—dire warnings about new technologies—as I think these issues are different and the point about research subject protection shouldn’t be in an essay that is mostly about the dire warning phenomenon. If Steve had written more about research subject protections, we could perhaps have a good argument on that point.


Steve Pinker sent the following to me on email and then at my request gave me permission to post:

Thanks, Alice, for the witty and welcome defense! We agree that human subject protections are not where they should be – there may be too little in some areas, but there surely are too many in others, as when they were used to persecute Mike Bailey and Beth Loftus, or, more parochially, when I have to formally request permission from a committee to change a single item in a questionnaire about irregular verbs. And many critics have pointed out that current IRB and privacy regulations are wreaking havoc with effective research and patient care (not least in the case of Jesse Gelsinger, who, it has been argued, was a victim of too much bioethical argumentation, not too little). So I’d stick with “ample,” though much of it is, as you’ve shown, misdirected.

Those interested in the Gelsinger case, including how he became “a victim of too much bioethics argumentation,” may wish to see Carl’s book, White Coat, Black Hat: Adventures on the Dark Side of Medicine.


My mate, Aron Sousa, read all this and said I could quote his response:

Nice essay. Pinker has been reading his own work about the better angles of our nature and thinks less about the individual outliers who are harmed. Dex is one of those system failures where everything that could go wrong has gone wrong: prenatal, off-label, intersex, fraudster, oil-rich donors, retrospective study design….

Those of you who know the story of dex from Galileo’s Middle Finger will know what he means by all that–except maybe the “oil-rich donors” bit, which I left out of the story lest people think I had to be making it all up.

But Aron’s main point is that Steve is thinking more about overall progress and “less about the individual outliers who are harmed.” It’s impossible for someone like Carl or me to do that, because after you spend years of your life tracking research harm in a particular case, you become pretty obsessed with “outlier” harm, and increasingly aware of how it may not be so rare.