"The Making of (Un)certainty in Knowledge Production: Rhetoric at Play in a Heritability Controversy" (by Angelica Camacho and Dominique Robert): Review 3

Vote: Publish pending major changes

This is an interesting paper that seeks to understand how fact claims are made and disputed within contentious scientific debates—using the debate on twin studies within Criminology as an example.

As one of the authors of the debate—a debate that took about a year of my professional life—I have intimate knowledge of what transpired and how the debate evolved. I’ll provide some of those details here.

We had heard from Dr. Simmons that he had a paper that was to be published on biosocial criminology. As we respected his work, we looked forward to reading his contribution, even if it was critical. However, when the Burt/Simmons paper was published in Criminology, we were shocked. First, never had we read a paper published in a reputable journal calling for the banning of research or a particular research method. People seem not to understand the broader implications of scholars arguing to ban and censor methods and people, but it is anathema to science. That said, if a statistical method can be demonstrated to produce consistently biased effects or effects that do not mirror reality, then it is incumbent on those making that fact claim to demonstrate mathematically those biases. Burt-Simmons did not demonstrate any inherent flaw in the mathematics behind twin studies. Had they done so, they would have invalidated decades of scholarly work by statisticians and behavioral geneticists and likely won a Nobel prize.

Second, I don’t know how to say this any other way, but they were scientifically wrong in many of their claims. The assumptions underpinning twin studies, for example, the assumption of equal environments, have been studied. Statistical genetics is a highly developed and specialized field—and one that has tested and retested its methodologies over time. Contrary to Burt-Simmons, there are no known inherent flaws in twin designs or in the mathematics that underpins them. They are robust, provide consistent evidence, and have evolved in complexity. Moreover, they are rooted in the mathematics of human evolution and genetic relatedness. Twin studies have been used to understand the genetic and social architecture of breast cancer, ADHD, homosexuality, and every other human trait and behavior imaginable (See Polderman et al., 2015). To say that estimating genetic and environmental variance is “nonsensical” is scientifically and mathematically wrong. Period.

Third, their paper was strident in tone and attacking in character. They went so far as to produce tables to “call out” those of us who worked in the area. While we don’t mind academic conflict, COPES guidelines dictated that we should have had an opportunity to respond. We did not get that opportunity. Instead, I wrote the editor demanding an opportunity. We were initially told that no such opportunity would be forthcoming, but after some cajoling, we could write a rejoinder. This is important for your paper because the debate would never have happened had me and Dr. Kevin Beaver not forced it.

I also want to point out that the journal’s editor was duplicitous in handling the Burt-Simmons paper and our exchanges. The editor worked with them to “call us out” and sent the paper to reviewers not qualified to review a paper on behavioral genetics. We know this by talking with Professor Simmons and through conversations with editorial board members at the time. So, this was not an effort to explore an important criminological topic engaged in good faith by the editor of criminology. It was, instead, an orchestrated effort to shut down scholarship in an area that had grown appreciably and that challenged core disciplinary narratives. It was, in my opinion, an illustration of editorial malfeasance.

The authors point out that Burt-Simmons selectively quoted and misquoted other scholars and left out of their discussion studies that would have provided evidence contrary to their views. Indeed, my colleagues and I spent an inordinate amount of time and effort researching their claims and citations because they showed a pattern of bias or motivated reasoning. Either way, they made claims contravened by the available evidence, and we pointed that out.

However, I think it is misstated that Burt-Simmons were trying to “promote the conversation about what needs to be done for biosocial criminology to be legitimized within the discipline of criminology.” No, they were not. While neither Burt nor Simmons are inalterably opposed to discussing the connection between biology and behavior, they are (were) keenly aware that the findings from statistical genetics likely invalidated their favored theories and, as such, could invalidate much of their life’s work. They were not trying to advance a conversation but were instead trying to eliminate a body of scientific work because the findings didn’t conform with their theoretical preferences. If there is a story here, it is how a discipline regulates and excludes evidence some dislike.

Moreover, we were not questioning Burt-Simmons “coherence” in their critique of the AddHealth data but their apparent hypocrisy. The AddHealth can’t be “ideal” for genotyping research when they use it, and yet inherently biased when we use it.

I disagree with the authors that “the authors do not explain the reasons why caring about the legitimacy of heritability estimates is important……...” We did. They contravene many established theories, and they call into question purely sociological perspectives. I also disagree that “fighting a socially and politically charged debate on highly technical grounds is a common way of producing ignorance…..” Math is technical. So, too, is genetic architecture. If a scholar makes a mathematically incorrect statement, they are wrong, regardless of how technical the math. This doesn’t produce ignorance. It produces correct, scientifically valid conclusions. The point was not to highlight the alleged controversy of the bio-paradigm, as the authors say, but to show that Burt-Simmons was simply wrong.

The authors draw attention to the hostility that emerged from the Burt-Simmons debacle and argue that perhaps the “pro-genetic faction” wanted to gain recognition or that we simply wished to be provocative. Let me suggest that there is no “pro-genetic faction.” We explicitly developed the term “biosocial” criminology to reflect equal parts biology and environment.

It is odd, too, to suggest we simply wished to be provocative in reaction to an editor and Burt-Simmons wishing to end our work and to besmirch our reputations. Why impute ill will on our part when they targeted us and called for censorship? This sounds remarkably backward. What is the proper response for scientists who have produced dozens of studies showing a genetic effect on crime and criminality (something widely denied in the literature) when other scholars seek to censor their empirical work and shame them in the field? What would be the appropriate response when scholars make unfounded allegations of racism, as did Burt-Simmons, and the editor of the top journal in the field allows it?

Again, let me suggest that calls to ban fundamental research into human behavior and to ban language emanating from that research is and should be considered “provocative.” Imagine the outcry if scholars attempted to shut down research on gender, race, or guns or tried to have banned methodologies used in those studies. To be clear, Burt-Simmons wanted our research canceled. They wanted us canceled, and they worked with the editor of criminology to make that happen. Their work was not a “conversation” but an accusation, and one in which they initiated.

I like what this paper tries to do, and I appreciate the authors' efforts. I would encourage the authors to think about the issues I’ve raised and, more broadly, about the purpose of scientific debate. By its very nature, scientific debate often involves highly technical issues. It sometimes involves sharply stated observations, and it sometimes involves the use of rhetoric to make and clarify points of difference. What it should not include is highly biased and selective reviews of the literature, mathematically incorrect statements, the casting of personal aspersions, demands for censorship, and editorial malfeasance. Yes, science is a human process. What this also means is that sometimes, the dark side of human motivation emerges.