Mass murder and censorship
NOTE: I now write for Aporia Magazine. Please sign up there!
On 14th May, an 18 year-old white man named Payton Gendron gunned down ten people (all of whom were black) and injured three others in Buffalo, New York – the country’s latest racially motivated mass shooting. As is common in these kinds of atrocities, the shooter published a lengthy “manifesto” outlining his ideology and the reasons for his attack. (The manifesto also includes 96 pages describing in forensic detail the weapons and gear that he used.) In short, he was a white nationalist and anti-Semite who wanted to kill “replacers”.
After the section explaining his motives, the manifesto includes two sections titled “About blacks” and “About Jews”, respectively. These sections include figures from, and citations to, various scientific papers, as well as statistics plucked from sources like the FBI’s crime database. Among the scientific papers cited are several dealing with genes and individual differences in IQ/educational attainment, including a 2011 paper in Molecular Psychiatry, a 2015 paper in Molecular Psychiatry, and a 2018 paper in Nature Genetics.
The appearance of these papers in the shooter’s manifesto led to much consternation on Twitter. One commentator asked, “What’s the point of publishing such weak results if that manifesto cites you?” And in the words of another, “Maybe it’s worth considering that this paper, and others like it, that search in the weeds for “significant” PRS scores and ignore numerous important caveats, should not be published. They don’t have any scientific value, and they only serve as material for manipulation.” (He was referring to the 2018 paper in Nature Genetics, which found that a polygenic score can explain 12% of the variation in educational attainment.)
While neither commentator stated that research on genes and cognitive ability should be banned, both were hinting at this possibility – or at least hinting that scientists should voluntarily refrain from doing such research. A third commentator went as far as saying that violence “linked to misappropriation of research” is an “actualized harm that should factor into every decision made by IRBs, funders, and journal editors.” (Remember it’s individual differences we’re talking about here; not the much thornier topic of race and IQ.)
You’ll notice there are two elements to the argument: the research in question has little or no scientific value; and it has been misappropriated by a killer. It’s therefore “worth considering”, so the argument goes, that it “should not be published”. A lot of things are “worth considering”, so this isn’t a particularly strong claim. But what about the stronger claim that it simply “should not be published”?
I think we can dismiss the first of the two elements quite easily. Clearly, the research does have scientific value. The results are not just noise, and while you might not be impressed with a polygenic score explaining 12% of the variation, this is comparable to what traditional sociological predictors can achieve. The polygenic score has better incremental validity than household income, for example. And in any case, findings from twin, adoption and molecular genetic studies all triangulate on the conclusion that genes explain a non-trivial portion of the variance in educational attainment.
Of course, if the research really had no scientific value, that would be reason enough to say it “should not be published”. Why would we want scientists doing work that has no scientific value? Consequently, we wouldn’t even need to invoke the second element – that the research has been misappropriated by a killer.
Okay, so let’s deal with that second element – which is the more challenging of the two. Should we publish research that has the potential to be, or which has in fact been, misappropriated by a killer?
There are some cases where almost everyone agrees that censorship of science is justified: detailed instructions for how to make biological weapons, for example. Of course, this isn’t so much “science” as “technology that’s based on science”. A clearer example might be gain-of-function research on viruses, which has the potential to spark a pandemic that kills millions of people. In this case, though, it’s not that someone might draw twisted conclusions from the research and then do something terrible; it’s that the work itself might kill people.
I’m hard pressed to think of any research that should be censored purely because of its potential to be misappropriated (rather than because the work itself carries inherent risks). After all, there’s a large amount of research that has the potential to be misappropriated. Work on climate change could be used to justify ecoterrorism, for example. In fact, there’s a large amount of speech in general that has the potential to be misappropriated. The works of Karl Marx and Vladimir Lenin inspired a huge amount of killing during the 20th century. (One of the Khmer Rouge leaders told a court that he merely wanted “social justice” for his country.)
Even contemporary left-wing speech has been used to justify lethal violence. In 2016, a black man named Micah Johnson killed five police officers and injured nine others in Dallas, Texas. According to Dallas Police Chief David Brown, “The suspect said he was upset about Black Lives Matter. He said he was upset about the recent police shootings. The suspect said he was upset at white people. The suspect stated he wanted to kill white people, especially white officers.” Does this mean journalists should refrain from discussing police shootings of black people?
In 2017, a white man named James Hodgkinson shot four people, including Republican Congressman Steve Scalise, during a practice session for the annual Congressional Baseball Game for Charity (all four survived). In the opinion of the Virginia Attorney General, Hodgkinson’s attack was “fueled by rage against Republican legislators”. The shooter was an avid Bernie Sanders supporter, who regularly posted vitriol against Republicans on Facebook, linking to stories on progressive news sites. Does this mean that Democrats should pause before criticising Republicans, and vice versa?
Returning to the Buffalo shooter’s manifesto, it’s worth noting exactly what it says in the section “About Jews”. In particular, there’s a graphic titled “The ‘Smart Jew’ Myth”, which disputes that “Jews have IQs of around 115”, describing this as a “lie propagated by Jews for the benefit of Jews”. This graphic even mischaracterises the work of Richard Lynn (a prominent hereditarian researcher), claiming that he “processed data from other tests”. In fact, Lynn wrote an entire book arguing that Ashkenazi Jews do have a high average IQ.
It’s obvious why the shooter would reject evidence that Ashkenazi Jews have a high average IQ: such evidence helps to explain why they’re overrepresented in elite professions and among Nobel-prize winners. The shooter’s clear double standards with respect to the black-white IQ gap and the Jewish-gentile IQ gap indicate (unsurprisingly) that he was not a dispassionate curator of scientific data. Rather, he welcomed the evidence when it suited his agenda, and dismissed the evidence when it didn’t. Suggesting that censoring the scientific literature might have done anything to prevent this particular attack therefore seems far-fetched.
All the cases I’ve discussed illustrate one thing: in a diverse society with people of different races and religions and political views, individuals with psychological derangements will develop grievances against other groups, and may act on those grievances up to the point of murdering innocent people. These acts needn’t involve race (as in the Congressional baseball shooting); although they often do. It goes without saying that all such acts are despicable.
Now, it’s possible the frequency of such acts is influenced by extraneous factors. This is essentially the argument of those suggesting research on genes and cognitive ability “should not be published”. They are saying that this research bolsters white nationalist ideology, and white nationalist ideology inspires people to go out and kill – which, as a general statement, may be true. In other words, if you actually did censor all the research on genes and cognitive ability, there might be fewer murders by white nationalists. I suspect the effect would be tiny, barely registering in the overall murder statistics. But I can’t rule out an effect.
However, as I’ve already argued, exactly the same reasoning applies to all sorts of other speech, including discussions about police shootings of black people, and criticism of Republicans – not to mention large swathes of religious discourse.
Which presents society with a dilemma: should we preemptively censor all speech that could plausibly be used to justify violence – whether through formal rules or voluntary self-censorship? Should we ban the works of Vladimir Lenin because they might inspire communist violence? Should we stop reporting birth rates lest they give ammunition to white nationalists? Should we gag neoconservatives in case they start another war in the Middle East?
The answer to all these questions is clearly in the negative. Rather than drawing the line at speech that could “plausibly be used to justify violence”, it makes far more sense to draw the line where liberal societies have traditionally drawn it – at explicit incitements to violence. Short of that, individuals should be able to speak the truth as they see it. Of course, one can argue that they have a moral duty not to be inflammatory, though this would rule out a lot of political speech, and wouldn’t justify censoring science – which is where this essay began.
Image: Memorial service following the 2016 Dallas police shooting, 2016
Thanks for reading. If you found this newsletter useful, please share it with your friends. And please consider subscribing if you haven’t done so already.