I dislike writing responses to articles I feel need correction or clarification. As a journal editor, I do it frequently. In this case, I’m writing a response to a response about responses to a response. By way of introducing this piece, VOX published a critical article about a podcast discussion between Sam Harris and Charles Murray with a disparaging headline about “junk science.” The issue was whether any scientific data support the idea that group differences in average IQ scores might be partially explained by genes (a very minor part of the podcast). The VOX piece thought this possibility was junk. I submitted a response to VOX; they declined to publish it but Quillette did (June 11). VOX then published a second piece where the authors of the original piece responded to a number of online criticisms in considerable technical detail. That second piece had a headline of “still no good reason” to believe genes had anything to do with group differences. In my view, this article was much better than the first and the headline was less objectionable. Nonetheless, I felt it gave VOX readers a one-sided, non-mainstream view of the issue. I also felt VOX was putting a thumb on the scale by endorsing a particular viewpoint with two articles by the same group and no articles that presented other informed views. Reluctantly, I submitted a new response to VOX. Again, they politely declined to publish my response believing it is time to move on to other topics. They have a point. There’s not much more to be said on this hot button issue given current data. In what may be a futile effort to have a last word, I offer my second VOX-rejected piece below:
Let’s talk frankly about genetics and IQ differences
If you finished reading to the end of the new VOX piece on IQ, you may be confused about the role genetics plays in accounting for differences in intelligence. This was an informative article but it’s easy to lose the thread. The key issue discussed is whether the following statement is justified by the state of current scientific research: Average IQ score differences among race-defined groups are partially genetic in origin. Turkheimer, Harden, and Nisbett (respectfully abbreviated THN) assert the statement is not justified by available evidence. The impetus for this article, the second published by VOX on this topic by these authors, is a podcast discussion between Sam Harris and Charles Murray that included an affirmation of the statement. Both positions derive from a widely held formulation among researchers who specialize in intelligence research called the Default Hypothesis: Whatever factors influence intelligence differences among individuals will also influence average differences among groups.
THN do not like this hypothesis although each author acknowledges somewhat different reasons. Like good prosecuting attorneys, they lay out their evidence and their argument seems compelling. However, as observers of courtroom dramas know well, there are other equally reasonable interpretations of the same data and other evidence can be introduced when the defense has a turn. This give and take is common in science but discussions about different interpretations regarding intelligence research often turn on technical details and confuse non-scientists interested in the subject.
Based on my understanding of the research literature, many disagreements about the data result from a loose use of language. The words “intelligence”, “IQ”, and “g-factor” are not equivalent although they typically are used interchangeably. Each has its own relationship to data. Intelligence is a broad word that includes many mental/cognitive abilities, IQ is a summary way to assess or estimate a person’s rank on many of these abilities compared to others, and the g-factor is one component of intelligence that emphasizes reasoning ability common across a wide variety of situations. IQ scores are good estimates of the g-factor but an IQ score also taps other components. When we talk about differences in IQ scores, they may or may not be due to the g-factor alone.
Considerable research shows the g-factor has strong genetic influences and weaker environmental ones. However, there is a paucity of modern DNA evidence about group differences, best defined by DNA-assessed populations rather than race. But that is about to change dramatically as multinational consortia gather DNA and cognitive test data from large samples around the world. These databases coupled with new statistical methods potentially can answer fundamental questions about individual and group differences. That’s why a frank discussion about all this is timely.
The worst that detractors can say about the podcast is that Murray and Harris prematurely endorsed the Default Hypothesis as resolved. Similarly, in my view, the VOX piece prematurely rejects the Default Hypothesis as somewhere between unreasonable and not provable.
In my experience, presentations of research data to non-specialists easily fall into overly simplistic conclusions and charges of cherry-picking evidence. Here are three things to keep in mind as you form your own opinion of what the data mean: 1) intelligence is a function of the brain and no story about the brain is simple, 2) no one study is definitive, 3) it takes many years to do independent replications and sort things out until there is a compelling weight-of-evidence to support some interpretations over others. We all should be careful when advocates for a particular point-of-view claim the moral high ground. The science will sort itself out — it always does. In the meantime we need to encourage more discussions like the Harris/Murray podcast and be resolute in supporting the expression of controversial ideas and informed challenges to them like the THN piece.
Robert Plomin, an expert in behavioral genetics, wrote in 1999, “The most far-reaching implications for science, and perhaps for society, will come from identifying genes responsible for the heritability of g [i.e. the g-factor].” The Chinese government apparently is devoting considerable resources to this endeavor. Meanwhile in the US, the majority of the Congress apparently does not believe evolution is more than a theory. This is an important story in the 21st century.
In my view, the research data show that individual differences in intelligence, especially those related to the g-factor, are strongly influenced by genes. It’s a challenging story about multiple neurobiological cascades initiated by genetic and environmental interactions. How all this works will take years to disentangle; it is a complex but exciting and finite set of puzzles to solve. The data on group differences, however, has not yet established a reliable weight-of-evidence. New data are coming in the next few years with advanced methods of analysis and controversy is sure to follow. Respectful public discussion is essential and scientists have an obligation to provide non-specialists with understandable explanations of highly technical methods and results. Listen carefully to the Harris/Murray podcast and read what THN say. Both are worth your time. Neither is definitive yet but scientific progress is moving inexorably toward a clearer understanding about the origins and nature of intelligence differences. Let’s be open to what we find out.