How should we handle controversial science?
Dual-use technologies in the human and social sciences
Geneticists and evolutionary scientists are doing a lot of soul-searching on social media right now. An 18-year old massacred 10 people, injuring another 3 at a grocery store in Buffalo, New York on Saturday. Eleven victims were Black.
It was a horrific, racist, pre-meditated attack. In a live streamed video, the killer steps out of his Ford and wantonly opens fire with his cheaply purchased assault rifle, modified for a larger magazine and emblazoned with a slur intended for his victims. In his inhumanity, he achieved what he wanted. Families in mourning, communities in fear, and a nation arguing over the “Great Replacement” theory, genetics, and group differences.
There was no ambiguity about why the perpetrator thought he needed to take those lives. He left a Google doc screed, where he detailed his plans and motivations. He discusses the “Great Replacement” theory—it’s the idea that there is a deliberate attempt to make Whites a minority by increasing non-White populations through immigration or by raising the status of non-White groups. He also cites several prominent articles on group differences and genetics as evidence. From those articles, he copy-pasted graphs, that “opened his eyes” to the “truth”.
The Internet is a powerful tool that allows us to peek into conversations we would normally never have access to. In many corners of Reddit, Quora, 4chan, and websites that don’t normally appear on the first page of Google, you’ll find entire communities for whom the contents of the screed are common knowledge, for whom the killer is a hero, the murders to be celebrated.
Focusing on this particular shooter is pointless. Even judged by his own twisted logic, his unprovoked, targeted murder of Black Americans makes no sense. Black Americans are a community as old as America and upon whose backs the country was built. They have genetic and cultural roots as deep as any non-native American can claim. The bigger concern is the community that provided his cultural input.
Mass shootings are an all too frequent tragedy in the United States. That some are motivated by the misuse of science is almost inevitable in such a large, diverse society. There are 330 million people in the United States. Our larger collective brain makes it highly likely that useful ideas, necessary capital, and networks of opportunity intersect for motivated innovators. Our larger collective brain also makes it more likely that a shallow understanding of genetic and evolutionary science, access to weapons, and the motivation to do something, intersect for would-be murderers. But as scientists, we do have a responsibility to reduce the probability of that intersection. To not provide fodder that fuels division and harm. But how we do that matters.
As a scientist working in cultural evolution, these concerns are constantly on my mind. Cultural evolution has many potentially controversial discoveries. That we belong to multiple overlapping and embedded cultural-groups of ideas vying for dominance through the forces of cultural-group selection or the equilibrium selection problem isn’t far from the dynamics of demographic and cultural change or the Great Replacement theory minus the conspiracy. Culture-gene evolved local adaptations are a potential source of group differences. In my own work, when you measure cultural distance it reveals that some countries and some religions are more culturally distant than others. And that’s not even getting into the potential implications of Eurasian ad mixture with ancient populations of Neanderthals and Denosivans.
My training in engineering makes it natural to think about these concerns through the lens of dual-use technologies. Nuclear technology gets you powerplants and bombs. Rockets can launch satellites and warheads. As researchers working in the human and social sciences, we are long overdue for a more thorough discussion and debate over how best to handle the equivalent of dual-use discoveries.
Our challenge is analogous to engineering, but thornier because the mere existence of a discovery of genetic distance or even our words alone can inadvertently support hatred, cruelty, and xenophobia. The nuance of the science is easily lost to the lesser angels of human nature. In one solution, unwelcome hypotheses are under explored and unwelcome results are suppressed for fear of possible societal harms. A fuller discussion never takes place, the topics themselves become taboo. This is an intuitively tempting, but dangerous path.
Censoring science is self-defeating. It feeds into the narrative that scientists are suppressing the “truth” in an attempt to prevent harm. That in turn leaves open a space for less scrupulous scientists or even non-scientists to fill with questionable work, driven by an agenda. These become the sole sources that appear in random Google research reported in dark corners of the Internet and sometimes by the media.
When professional organizations put out statements on what the definitive truth of the world is knowing that the science can change, it leaves a feeling of “the lady doth protest too much”. When researchers make broad statements on groups such as “immigrants” knowing that it’s too broad a category to generalize, it undermines our perceived commitment to faithfully discovering and reporting the truth to the best of our ability.
This was something we dealt with in a recent paper on the “Paradox of Diversity”. We didn’t shy away from difficult findings, but we did try to give readers a fuller picture of what the research says and how we can think about it. For example, to ask if “immigrants” contribute more or less to the economy, commit more or less crime, or are happy or unhappy in their new homes is as crude as to ask if “native-born citizens” contribute more or less to the economy, commit more or less crimes, or are happy or unhappy. When we talk about our ingroup, it’s more obvious that we need to disaggregate broad sweeping claims, be they positive or negative. We need to disaggregate by factors such as the nature of the economy at the time of immigration, the characteristics of immigrant and local populations such as culture, age, economic conditions, and level of education.
I explored these broader debates about potentially harmful science in a recent BBC radio show where I interviewed David Reich, Steven Pinker, Brandeis Marshall, Caroline Criado Perez, Pedro Domingos, and Emily M Bender on AI, science, freedom of speech and so called “cancel culture”. A theme on both “sides” was that to solve social problems, we must first understand them. And because the world is a complicated place that doesn’t always conform to what we think or hope the answers are, we can only do that in a diverse environment of different backgrounds, considering all hypotheses, and ideas. Evolution needs diversity and transmission. This kind of exploration needs people with different experiences, discussing in a safe space of unfettered free speech.
The fact is that none of us is unbiased. Indeed, none of us can even see the full extent of our bias. Science doesn’t work because we’re enlightened humans who see past our incentives and our life experience. It doesn’t work because we readily change our minds in the face of new evidence. No, science works because we commit to a method of discovery, agreement on what counts as evidence, and most importantly, are incentivized to show others when they’re wrong. It’s a collective act that slowly converges on the truth. Our findings can only be trusted if we could have found the opposite to whatever current political sentiments suggest is the “right answer”. Take behavioral genetics for example. There are many critiques to be made.
We made a high-profile critique in a recent paper on “cultural evolutionary behavioral genetics”. My coauthors and I argue that there are fundamental problems in theory, measurement, and interpretations of behavioral genetics that can be resolved through the intergration of cultural evolution. As part of this argument, we explain why the bulk of work on human evolution suggests that between-group differences are almost certainly best explained by culture and environment, not genes. And that groups themselves are a challenge to define. You can read the paper for the details.
Obviously, I think we’re right and perhaps you might too. But it’s essential that scientists who think otherwise can freely argue for their position. Doing so not only helps us arrive at the truth—a truth essential for knowing how best to make our world a better place—but it also increases trust in scientists and the perception that all views are on the table and ones that are trumpted are not trumpeted because they are palatable, but because they are true. We hope that some of that discussion comes out in the many commentaries and our forthcoming responses. None of this abrogates our responsibility to dual-use discoveries and careful consideration of the broader impact of what we discover and how it can be contextualized for mass consumption.
From my perspective the solution to misinformation is more information. The answer to the infamous “fire in a theatre” analogy is that when someone shouts “fire” in the theatre, we need other voices to shout “no there isn’t”. And we need to track reputations to turn that person into the boy who cried fire when there was none.
But maybe I’m wrong.
When 13 strangers are shot while picking up groceries on a Saturday afternoon. When the person who shot them says he did it because he didn’t like the color of their skin. And when that person cites evidence that our scientific community published—evidence that is then widely shared in online communities dedicated to uncovering “truths” they claim the media and scientists are downplaying or hiding to political ends—well, there’s a problem. As we mourn more innocent lives lost to ideologies, it is appropriate and imperative that we all need to do some individual soul-searching and have a collective discussion about how best to handle dual-use discoveries.