This week’s episode of Sam Harris’ Waking Up Podcast featured a fairly contentious conversation with Ezra Klein. While the conversation included a range of focuses, from my perspective there were two foundational cruces of disagreement. The first of which was over the ability to have a discussion of scientific data absent the broader context of the society in which the data is produced. The second of which was over a disagreement over whether Harris can possibly harbor tribalistic motivations in deciding who to converse with on his platform and who to reject out of hand. This post will focus on the former disagreement. I will say as little as I can on the validity of race and IQ science, since my focus is on the separability of science from societal context in a broad sense.
In the interest of disclosure, I am more sympathetic to Klein’s position on both of these points. This will surprise nobody familiar with some of my previous writing.
The episode revolved around social scientist Charles Murray, who is well known for an infamous chapter in his book, The Bell Curve. This particular chapter showed that black individuals on average scored lower on the IQ test than other races on average, and argued that heritable genetic traits are a factor that lower the intelligence of the black population as a whole. This particular subject is obviously a contentious issue, even on this blog.
Harris has previously invited Charles Murray on his podcast to discuss some of the backlash Murray has faced since then, up to and including being physically attacked. As part of this episode, Harris felt compelled to address the touchy subject of race and IQ. This is part of the issue that Klein took issue with, and where Harris and Klein spent much time going back-and-forth.
Harris’ interests in his episode with Murray were to simply raise the point that, by his assessment, Murray’s claims about race and IQ were fairly undeniable. At that point, they moved on to discuss broader speech issues and the ability to discuss controversial topics. Klein and several psychologists who research intelligence have since criticized this exchange and Harris’ characterization as “scientific consensus”. As Klein phrases it,
“To put this simply: You cannot discuss this topic without discussing its toxic past and the way that shapes our present.”
Much of the conversation between Klein and Harris seems to rest on the disagreement over whether or not it is ethical (or even possible) for Harris to simply discuss these IQ data without invoking broader political implications. I think it’s worthwhile to dig into the question of whether it’s possible to discuss science on its own merits, absent of a broader societal context.
As a scientist myself, it seems obvious and perhaps self-evident that science sans context is, at the very least, uninteresting.
Furthermore, it’s undeniable that presenting science within its proper context is part and parcel with mainstream scientific culture. As a researcher communicating research findings, it is essential to disseminate your results to your audience by motivating why you are interested in this type of work. Once you have presented and interpreted the relevant data in your work, it’s important to discuss as any implications the data might lead to. This is necessary whether you are giving a talk or writing a manuscript to be published in an academic journal.
Contextualizing the research is important enough to scientific process that it has been institutionalized within the mainstream journal format. Most long-form articles have an introduction section that motivates the reader’s interest in the science to be discussed. After the research methods and data has been presented and the scientific interpretations have been disseminated, the paper will usually conclude by discussing future implications of the research. This could be as wide-ranging as discussing the social or natural impacts of the findings, or localized to the research field itself by discussing what future related research should involve.*
Not even the scientific sources that Charles Murray pulls from in his infamous Bell Curve chapter present the data purely decontextualized. As an example, I pulled an excerpt from the discussion section in Vincent (1991), a paper that discusses IQ test performance as a function of age and race.
“The most obvious explanation for these findings is that Black/White IQ differences are in a very real sense a barometer of educational and economic opportunity. It appear that lack of early educational and economic opportunity has taken a toll on the black adult population that is still being reflected in the recent renorming of the WAIS-R and other adult IQ tests. This conclusion should not be surprising, for while it generally is conceded that IQ is malleable in the early years not even the most ardent environmentalist would be optimistic about a significant change in intellectual performance in adulthood and then only with extensive education. In other words, racial inequality has left its mark on Black adults, but the benefit of attempts to improve Blacks’ access to equal educational, economic, and environmental opportunity appears to be succeeding. The quarter-century push toward equal opportunity in the United States finally is beginning to show results.”
I would hope that readers can find here that while indeed a lot of these conclusions are things that are directly drawn from the data, there are very firm value and societally-based implications presented in one of the final paragraph of this paper. It’s very nice for Harris to just want to mention certain data absent of context, but even the supposedly firm unbiased scientific source doesn’t entirely decontextualize these data.
While scientific papers aren’t divorced from being reflections of the society they have been produced in, it’s probably fair to say that academic literature probably does not discuss touchy subjects in the way that Ezra Klein states they should, where I quoted him above. The above cited paper doesn’t discuss post-hoc genetic rationalizations of black performance or pseudoscience of phrenology in its contextualization of race science. It doesn’t discuss the long history of justifications that were labeled as science to enforce white supremacy in America.
Likewise, papers on germ line editing don’t begin their papers on the history of eugenics and discuss how class-based access to control over genetic heritability will inevitably lead to social stratification. Papers on artificial intelligence research aren’t expected to have discussions of personhood and the ethics of producing a conscious being. Many, many scientific problems are inherently tied to difficult philosophical and ethical problems. Does this seem to counter Klein’s point?
Perhaps Klein does think that academic papers would be more responsible by emphasizing how this science can be misused and abused, though I cannot speak for him. However, I would argue that this type of specific contextualization may not be necessary. When communicating science, writing for the intended audience is a primary concern, and journalistic literature is designed for scientists who are more concerned with answers to scientific questions in an abstract, academic sense.
This is not the case for a broad, general audience like Sam Harris’, or amateur science enthusiasts. One of the well-known problems in science popularization is that press releases tend to oversell results, and vastly over-emphasize research implications found in an article’s discussion section. It’s distressingly too common that science journalists and communicators will discuss a mere speculation in the final paragraphs of a recent article without even touching on the data found within the paper.
It’s apparent that in order to sell science to an audience, sexy groundbreaking landmark findings sell ad revenue and magazine subscriptions, while the pure data absent of context simply is uninteresting. Does Harris truly think that the majority of his audience is simply interested in a single difference of an IQ measurement, and an insubstantial amount of his audience doesn’t care about potential implications?
To return to the conversation with Murray and Harris, I would say that it’s borderline impossible to actually discuss these data absent of context. Even if the conversation on race and IQ were literally simply bringing up two Normal IQ distributions’ means and standard deviations and nothing else, Harris would still be bringing attention and a spotlight to Murray. As Klein points out, Murray’s career is largely focused on explicitly tying racial differences in IQ to things like affirmative action and social support networks. Even if Harris were to explicitly refuse to discuss social implications, the act of bringing an audience to a man who ties policy to the data is, in itself, a political act.
Additionally, the mere act of pretending that inviting a polarizing political figure onto a large platform is an apolitical act is political. This is much like when anti-gun control proponents ask not to politicize a mass shooting. By trying to suppress certain types of political discussions, they are in fact trying to shape the political landscape in a certain manner.
As a final criticism of Harris, it seems that if he wanted a dispassionate discussion of the scientific data on race and IQ completely absent of a social context, framing the podcast episode as forbidden knowledge against political correctness does not seem to be a good way to go about doing this. While he will claim that most of the discussion is on how Murray has been treated and they only raised IQ and race so they could get rid of the elephant in the room, this still recognizes the political component of the data.
Ultimately, while we like to think of scientific discovery in the abstract as a purely rational, dispassionate, noble pursuit of nothing but the facts, I find that to be a modernist fantasy that borders on scientism. It’s fairly clear to me that while science is a fantastic tool for discovery and eliminating personal biases to explore the natural world, it’s impossible to entirely divorce from the broader social context.
Perhaps it is easier for me to see this as a researcher who is familiar with the flaws and biases that still continue to crop up in the scientific literature. It’s all too apparent to me that while scientific methods are profoundly powerful as an epistemic basis for discovery, it is still a tool employed by flawed humans with their own motivations (and funding, especially in Murray’s case). Scientists have narratives just like everyone else, and it’s the job of the scientific community to try and curtail these narratives before they are disseminated to the public.
*Notably, even article abstracts will often include a brief sentence or two of contextualization. The abstracts are supposed to be a brief bare-bones summary of the article, reduced entirely to the essentials. It would seem that even when an author wants to reduce the paper as much as possible to the barest of information needed, explaining why the research was performed continues to be necessary.
Stay in touch! Like A Tippling Philosopher on Facebook: