Some weeks ago I stumbled upon a link to a collection of emails between Sam Harris and Ezra Klein. The two argued about an article that Klein’s site Vox had published in response to Harris’s podcast with controversial political scientist Charles Murray. While starting relatively well, it got worse. Whatever good will was there in the beginning had run out by the end, spent on failed attempts to get on the same page.
This sort of thing happens a lot, of course, but this was special: two intelligent, rational authors baffled about each other’s behavior in a disastrous misunderstanding regarding a complex, controversial issue. Thousands of words result, just counting that email exchange.
How disagreement works is my main theme on this blog, so of course this shit is catnip to me. I just had to look into it. I did and I’ve been obsessing about it ever since. Hence this 9,000 word article, cut down from twice as much material.
I came away from the email conversation with the impression that Harris got unhinged and retracted his invitation to have Klein come on his podcast for no good reason. And he had released this exchange himself, obviously thinking people would take his side and it’d reflect badly on Klein. That’s not what happened. Many thought Harris came off as the unreasonable party (and I can certainly see that). It’s quite remarkable how he could be so convinced that publishing the exchange would be primarily embarrasing for Klein. His anger also seemed out of character; I’ve listened to a few of his podcasts and he comes off as well-reasoned, balanced and rational. Nonetheless he appears to be the one responsible for the conversation deteriorating somewhere in the middle. Why this reaction from him? To find out one has to trace the conversation backwards.
He accuses Klein and the Vox article authors of slandering him and Murray. A “hit job” he says. I went back to read it. It didn’t exactly come off as a hit job to me, despite some inflammatory word choices. It seemed at least OK, if slanted in a normal-yet-regrettable journalistic fashion. So I realized this wasn’t enough. I’d have to listen to the two-hour-plus podcast too.
After that I thought that the validity of Murrays research is a big part of the disagreement. Ideally, in order to do this properly, I’d have to read Murray’s controversial book The Bell Curve, and after that all the research cited in it and all research on the topic as a whole, before and after.
I’m not going to that because it would be insane. Instead I’ll focus on the difference between Harris and Klein in how they see the issue and what caused the catastrophic breakdown of communication between them.
It’s worth pausing to note that the first draft of this article was written before Harris and Klein did a podcast together to discuss their spat. I was surprised at how quickly they arranged it and thought, hell no, now everything I’ve written might go down the drain because it won’t be relevant after the podcast clears things up. That’s not quite what happened. Some things did get less relevant and the target of some well-needed editing, but no one who’s heard their conversation can think they wrapped everything up. Some new information did come out which made me update some parts and write some new ones.
There have been a LOT of Discoursing on this whole affair and I haven’t been able to keep up with all the third party takes, response articles and the responses to their responses, but I read the emails a number of times, I read the Vox article by Eric Turkheimer, Paige Harden and Richard Nesbitt (“THN” from now on) a number of times as well, some responses to it, Klein’s articles etc. I’ve listened to the Harris-Murray podcast three times, Harris’s “Housekeeping” podcast once and the final Harris-Klein podcast twice.
Whew. Maybe I am insane. It’s just that if you’re interested in the philosophical, sociological and psychological underpinnings of disagreement this whole affair is the equivalent of an all-you-can-eat buffet that never closes.
It’s impossible to discuss all of it. Every single paragraph of the correspondence, the article it refers to, the final podcast and and the relevant parts of the Murray conversation deserve exegesis. But this is a blog post, not a book-length erisology dissertation. No matter how much I want to dissect hundreds of individual passages, I’ll by pure necessity focus on describing this conflict at a higher level of abstraction. I’ll interpret, compress and narrativize even though that’s the habit that got everyone into this mess in the first place. There’s no other way.
Occasionally I’ll speculate about what Harris and Klein are thinking. I want to stress that I don’t necessarily mean this literally. I’ll construct representations of them that embody opposing perspectives, and use them as tools to showcase various disagreements. This might sometimes overinterpret them or assign them opinions they don’t actually hold, and for that I apologize in advance and hope that the increased relevance for similar disagreements between others makes up for it.
Let’s dive in. We’ll start at the beginning: Sam Harris’s podcast with Charles Murray.
From Harris’s point of view
Harris starts his podcast (and his email prologue) with a list of what he says are scientifically uncontroversial facts. Paraphrased:
1) Human intelligence is a real thing and and a scientifically valid construct.
2) IQ tests measure it well.
3) It is strongly correlated with educational and economic success.
4) It is moderately to highly heritable among humans.
5) Test scores differ between races. Most notably and controversially, black Americans score between 10 and 15 points lower than white Americans on average.
What he says is correct. These five claims are broadly accepted among intelligence researchers.
Note that the list doesn’t include
6) The racial differences mentioned in (5) have a significant genetic component.
6-b) There are currently no effective means of permanently raising intelligence through environmental interventions.
It seems clear from the podcast that Murray also believes (6). If I understand him correctly, he doesn’t believe he can definitely prove it or that it’s completely certain. Rather, I get the impression it’s a judgment call he makes, eyeballing the evidence and finding that a significant genetic component to (5) is the most likely, and that the seeming irrelevance of shared environment (the environmental factor appears to be all nonshared according to behavioral genetics) makes effective interventions unlikely.
Harris separates 1-5 from (6), and this is important for understanding how Klein and the authors of the Vox piece interpret the podcast differently from how Harris does. In very condensed form, their disagreement is about the legitimacy of making that call. As we shall see, it depends on many factors.
Evidence that neither Harris not Murray considers the “dangerous” (6) definitely proven comes about 40 minutes into the podcast, where Harris devotes some time to saying explicitly that just because individual difference can largely be explained by genetics it doesn’t necessarily mean that 1) all individual difference in success or mental abilities or 2) all or even any of group differences are due to genetic differences. Murray agress and calls it “critically important”. Then they discuss the importance of environmental factors.
In Harris’s mind, the truth or falsity of (6) is not the issue. He does think that believing it is a reasonable extrapolation from 1-5. There are, as he says in the beginning of the Murray podcast, a series of “nested taboos” around intelligence research that concern all the claims from (1) to (6). Claims 1-5 are often wrongly treated as controversial or even wrong. But they aren’t, and from their non-controversialness, (6)’s legitimacy and status as reasonable hypothesis follows.
At the time Murray wrote The Bell Curve, these claims were not scientifically controversial, though taken together [my emphasis, JN], they proved devastating to his reputation among nonscientists. That remains the case today.
To Harris, Murray simply wrote a big, thick book about intelligence in which he put, as he himself puts it, a single paragraph saying that both environment and genetics probably matters for the black-white gap. And his treatment was horribly unfair. Harris reads an quote from a sociology professor (!) calling the book “Nazi propaganda” or a “scientific version of Mein Kampf”.
The Bell Curve (note that I haven’t read it, I’m going on reasonably reliable hearsay) is almost entirely about the uncontroversial 1-5. It’s in general quite careful in sticking to the established science, and race issues are only a very small part of it.
Similarly, (6) isn’t the topic for most of the podcast. Interestingly, this is far from obvious to a casual listener. I listened to the whole two-hour-plus shebang three times and I didn’t realize until the second listen how little time they actually spend on it. On the third listen it changed again and it became clear why I didn’t see it the first time. I now noted how (6)’s ghostly background presence colored the interpretation of everything said, whether it was Murrays otherwise mainstream opposition to affirmative action, the apparent irrelevance of home environment for intelligence as suggested by behavioral genetics research or the later discussion on class-based sorting and the election of Donald Trump.
Whether such subtext is “there” or not is matter of interpretation, and therefore of opinion. I don’t think Harris thinks of it as “there”, and therefore he doesn’t consider the topic of their whole conversation to be “the genetic underpinnings for racial differences in intelligence and its implications”.
Instead (and he says this many, many times) Harris considers the topic of conversation to be the tense political atmosphere around claims like (6) and by extension the conjunction of 1-5. He confesses he’s been complicit in shunning Murray for fear of the smell rubbing off on him. The recent events at Middlebury College, where Murray was deplatformed and attacked, changed his mind. In his introduction to the emails, he says:
So I decided to invite him on my podcast to discuss the episode, along with the mischaracterizations of his research that gave rise to it.
Needless to say, I knew that having a friendly conversation with Murray might draw some fire my way. But that was, in part, the point. Given the viciousness with which he continues to be scapegoated—and, indeed, my own careful avoidance of him up to that moment—I felt a moral imperative to provide him some cover.
When they get to the “radioactive” (Harris’s word) parts of the conversation around the one hour mark he frames it as wanting to “slay the elephant in the room” so they could move on to what he really wants to discuss — Murray’s experience at Middlebury and the cultural atmosphere that led to it.
Considering this, it makes sense for Harris not to be particularly critical of Murray. The authors of the Vox piece says he “fails to ask a single challenging question”, which isn’t true, but it is true that he doesn’t push particularly hard.
From the other side
It looks a lot different to Ezra Klein, and I suppose the authors of the Vox piece. The first point of disagreement is that according to THN, Murray and Harris do claim that the truth of (6) is scientifically uncontroversial.
The consensus, [Harris] says, is that IQ exists; that it is extraordinarily important to life outcomes of all sorts; that it is largely heritable; and that we don’t know of any interventions that can improve the part that is not heritable. The consensus also includes the observation that the IQs of black Americans are lower, on average, than that of whites, and — most contentiously — that this and other differences among racial groups is based at least in part in genetics.
I think they’re mistaken about the last part but I see no reason to think it’s not an honest mistake (it’s uncertain what exactly “scientifically uncontroversial” means). This difference has a massive impact on how the two sides view the entire controversy.
While Harris considers (6) not to be the central load-bearing pillar of his conversation with Murray, Klein does see it that way. In his mind, Murray is pushing a whole narrative, centered on racial genetic differences, meant to justify his preferred social policies. Klein says in his own later Vox article, and further develops in his later conversation with Harris, that virtually all of Murray’s work throughout his career in political science and in conservative think tanks is aimed at dismantling the Great Society and abolish many social services. He’s not just a scientist who got politics thrust upon him.
Consequently, far from being noncentral and not the focus, everything Murray says to Harris is in service of a particular conclusion, namely that black people are of lower intelligence than white people and there’s nothing we can do about it. The carefully designed caveats and the (in THN’s words) “anydone” exhortations to not discriminate are there to make that message more palatable.
The overwhelming thrust of your discussion features Murray arguing that racial IQ differences are real, persistent, significant, driven by genetic racial differences (he has a long discourse on how strong that signal must be to make it through the noise of racial mixing), and immune to virtually every intervention we’ve thought of. Yes, there are caveats sprinkled throughout, but there’s also a clear and consistent argument being made, or so it seemed to me. That was, as I understood it, the Forbidden Knowledge referred to in the title: you can’t just wish away the black-white IQ gap as a matter of environment and history and disadvantage.
“The overwhelming thrust” here includes things that go on past claim (6), from science into social policy. In his article, Klein has a section about how similar ideas has been used to justify policies exacerbating racial inequality, and sees this as Murray’s purpose:
It’s not just that inequality is so deeply embedded in our genes and environment that we don’t know how to make progress — it’s that it’s dangerous to even try:
HARRIS: I guess one thing that must be occurring to listeners now — and this is my misgiving about having this conversation and going into this area at all — the question is why talk about any of this? Why seek data on racial difference at all? What is the purpose of doing this?
MURRAY: Because we now have social policy embedded in employment policy, in academic policy, which is based on the premise that everybody’s equal above the neck, all groups are equal above the neck, whether it’s men and women or whether it’s ethnicities. And when you have that embedded into law, you have a variety of bad things happen.
Note that even if Murray’s policy proposals are logically independent of his beliefs on intelligence (which Harris considers them to be), bringing policy into the conversation is going to, narratively speaking, change its center of gravity. Instead of focusing, as Harris wants, on the truth of 1-5 leading to the plausibility and respectability but not the empirical certainty of (6) and the “moral panic” around just entertaining the hypothesis out loud, inclusion of policies that appear to be based on (6) is going to be seen as arguing for the correctness of (6) by making it a load-bearing part rather than something pointed towards close to the edge of the frame.
Klein makes some fairly convincing points that this is intentional by Murray, who he claims is more focused on both race and politics than he lets on. While most of The Bell Curve is silent about race, the publicity campaign apparently wasn’t. I’m not qualified to discuss the publicity campaign of a book that came out when I was ten, so I’ll take Klein’s word for it.
I work with authors on book excerpts, and I can testify it is a known fact that the portion of a book you excerpt and put on the cover of an influential national magazine is going to get more attention than the rest of the book. Often, authors will not let you excerpt solely a particularly controversial portion because they’re concerned it will overwhelm the rest of their argument.
If this applies to Murray and The Bell Curve — and Klein is convinced that it does — then that clashes with Murray’s version of events where he didn’t much care for race at all but more or less had to put a little about it in because it became an issue. I don’t know which is true, but it certainly explains why Klein thinks of Murray the way he does.
A case of motte-and-bailey?
Is Murray saying that it’s likely there’s some nonzero genetic component to racial differences, or that black people are dumber and this isn’t going to change so we need to stop aiming for equality?
Consider what Klein said in one of his emails:
I was very prepared, reading this piece, for Murray and you and others to disagree with it. What’s confused me is the argument that the disagreement is invented, that this is all a misunderstanding. Something here is very off, and I am struggling to figure out what it is. My working theory is that there’s a strong version and a weak version of Murrayism, both are represented in the conversation, but though the strong version is emphasized in the presentation, there’s been a retreat to the weak version upon challenge. But perhaps that’s wrong.
Harris responds badly to this, which is a shame because it’s a great opportunity to get to the bottom of their disagreement. He’s angry and doesn’t think Klein could possibly have such a different impression in earnest. But as I’ve tried to say, I think he does.
It’s not an ususual situation. I don’t know if Klein is familiar with the term (my guess is yes), but what he described is a motte-and-bailey — an argument structure based on a medieval warfare metaphor where there is one conservative (as in cautious) and easily defendable version of an idea (the motte), and another more bold, far-reaching version that’s the one you really want to push (the bailey). You promote the bailey (often by implication) to get what you want, but retreat to the motte version when challenged. “I was only saying…”
Does this describe the message? Structurally, yes (although I’m not convinced it’s intentional). The motte is the largely uncontroversial claims 1-5, and them making (6) plausible. The bailey is that (6) is true and genetic differences plus the lack of reliable methods to raise IQ’s by manipulating the environment implies that the racial gap can’t be closed and this justifies Murray’s preferred policies.
While Klein clearly understands the structure, he doesn’t seem to get that Harris anger makes sense if you think of only the motte as in play. Klein says in an email:
I’m perplexed by the criticism, which I’ve seen some make and I think you’re implying, that there actually isn’t much daylight between the case you and Murray present and the one the authors present, and what disagreement exists is a matter of dishonest framing.
Harris does think that. He correctly judges there to be no major disagreement about the motte. In his words:
[Murray] doesn’t know how much of interracial IQ difference is genetic and how much is environmental, and he suspects that both are involved. His strongest claim is that given the data, it’s very hard to believe that it’s 100 percent environmental. This could be said about almost any human trait.
If you read closely, THN don’t reject this. They still say Murray is “peddling junk science”, by which they apparently don’t mean bad science but bad interpretation of science. To them (and to Klein) the message coming from the podcast, the “overwhelming thrust” of it, and the “consistent argument being made”, is the bailey. Claim (6) was not the end but the beginning of the important part. It was the seed of a narrative explaining racial inequality in America, and as such it was deficient, dangerous and worthy of some strident criticism.
Of course the disagreement is going to seem invented to Harris. It’s extremely small with regards to what he finds important (Murray’s policy prescriptions are largely beside the point and later Harris says he doesn’t endorse them), and he finds it intellectually dishonest to act as if at least the legitimacy of (6) as a likely hypothesis doesn’t follow from what they all agree upon. They’re making a huge deal out of nothing, and this is precisely in line with the problem (“moral panic”) he’s trying to draw attention to.
Of course he’s pissed.
Of course they think they’re justified in reading between the lines.
Science, politics and identity politics
One consequence of the motte-and-bailey difference is that Harris and Klein disagree on whether it’s a scientific or political question they’re discussing, and by extension whether scientific or political debating norms apply.
The difference is best illustrated by looking not at the muddled semi-rational practice of political debate, but at the purified rejection of scientific debate norms that is identity politics. They discuss identity politics a fair bit on their final podcast. Harris hates it and considers it prima facie illegitimate, while Klein insists that Harris is in fact engagin in it himself.
It’s clear that they don’t mean the same thing by the phrase, and Klein’s defintion is much more generous (there’s a motte-and-bailey structure here too). An unproductive conflict results, where Harris is too resistant to Klein’s quite obvious point that what you think might happen to you personally affects what you find important. He calls Harris’s concern over the treatment of Murray (his main motivation, according to himself) “identity politics” because according to him, Harris’s identity as a (white) public intellectual makes such problems feel greater than the problems (racism, sexism) such behavior is meant to address.
In this view, your identity affects your emotional responses and therefore what you consider good-bad, important-unimportant, interesting-uninteresting or valid-invalid. Hence what you argue for is inseparable from identity. This obviously applies to Harris, as it applies to everybody. He’s makes a mistake when he flatly denies Klein’s description instead of pointing out how it differs from what he means by identity politics.
In traditional scientific-philosophical debate, the content of what anyone says is evaluated according to agreed-upon, impersonal standards like logical coherence, empirical evidence and commonly accepted values. Motivations, personal history, possible consequences or hurt feelings are not considered to be on the table. Bringing them there, in the form of ad hominems, guilt by association, or non-sequiturs (in this case, extrapolation beyond someone’s literal words) are considered foul play.
The kind of identity politics Harris is referring to is a result of people rejecting the idea of impersonal standards and the practice of separating the argument from its social context into a sandbox-like logical-empirical domain. Instead of logical coherence, empirical evidence and appeals to commonly held values, we get credibility by victimhood, subjective experience and appeals to empathy and vulnerability. Its essence is the idea that claims should be accepted not because they’re logically and empirically sound or morally agreed upon, but because the people making them are owed it.
Harris isn’t doing identity politics in this sense. He doesn’t expect his identity to be an input to his arguments’ evaluation function, and from what aspects of his psyche they’re coming isn’t relevant. Not according to tradition Rational Style debate rules anyway, where evaluation functions only take the content of the argument. Identity politics means refusing to stay in this sandbox and the result is Activist Style, based on traditionally disallowed moves.
Ordinary politics and political journalism play dirty too, because when you really want to win you get out of the sandbox as soon as you think it benefits you. I think strategic use of Activist Style techniques is so normal in politics and political journalism (and frankly, everywhere except among philosophers, scientists and technologists who I, in a fit of typical-minding and wishful thinking, want to see as the norm) that members of those professions don’t think of it as playing dirty at all. At least not as playing dirtier than generally accepted and expected.
This is likely why Klein appears so surprised at Harris anger (to the extent that his surprise is honest). To him, political logic and its tactics are a fact of life and Harris being angry about him using it feels bizarre, like it would feel bizarre for a regular person just having a job to hear an anarchist yelling at them about “collaborating with the system”.
In support I’ll note that Klein several times says things like “you wanted debate, this is debate” or “this is well within the bounds of acceptable discourse” (about the Vox piece) or “this is how it goes” in response to Harris complaining about lack of charity. He doesn’t seem to object to Harris calling the Vox piece “propaganda”, nor does he see any problem with THN saying that Murray is peddling junk science and he himself implying that Murray is fixated on the black-white difference for murky reasons, when he apparently knows Murray and says that “he’s a lovely guy interpersonally”. To me, there’s a clear “this is war, not personal, the ends justify the means, and this is what happens” sort of attitude coming through here. In a former campaign worker it’s not that surprising.
Harris isn’t a political journalist and former campaign worker but a scientist, so to him lots of what’s considered normal in politics is disallowed and counts as dishonest. Applying political logic to an issue is by itself bad faith if you expect Rational Style.
One small step for science, one giant leap for politics
It’s okay to have different norms in different contexts, as long as everyone is on the same page. But sometimes we get to an issue two paradigms want to claim as their own: “My rules apply here!” *plants flag*. Except we don’t get that explicitly, which is a shame because then those claims could be discussed. Instead we get people assuming that their rules apply and go right ahead and apply them.
I discussed a similar situation in Science, the Constructionists and Reality, where the sociology of science and related research programs builds an account of science modeled on other idea-generating processes, making it incompatible with science’s account of itself. It didn’t go well then and it doesn’t go well this time.
Harris and Klein are champions in a turf war where science and politics (as thought styles and debate styles) want to lay claim to the question of racial differences in intelligence. The mismatch is simple: different reference frames assign different importance scores to different issues, and the political frame is here assigning an earth-crushingly large magnitude to an issue that’s not that big of a deal scientifically.
This transparadigmatic dissonance is well captured in an important passage from THN:s article:
Murray presents himself as coolly rational and scientific as he proceeds to his conclusion of genetically based racial differences: People differ in behavior, groups of people differ in behavior, people differ genetically, groups differ genetically. One way or another, genes are associated with behavior, so of course some group differences in behavior occur because of genes. No big deal. “This is what a dispassionate look at decades of research suggests,” Harris blithely says.
It is a big deal. The conviction that groups of people differ along important behavioral dimensions because of racial differences in their genetic endowment is an idea with a horrific recent history. Murray and Harris pepper their remarks with anodyne commitments to treating people as individuals, even people who happen to come from genetically benighted groups. But the burden of proof is surely on them to explain how the modern program of race science differs from the ones that have justified policies that inflicted great harm. Is it simply that we now have better psychological tests, or more sophisticated genomics?
Note the frame shifting as we move from the first to the second paragraph. It’s right in front of your eyes.
It’s completely true that it’s not a big deal scientifically, for the reasons they mention — even if they mention it specifically to show how wrong it is to be cool and rational about this. At the same time it has massive political implications. So is making that call, like Murray does, a small thing or a big thing? Either something is a big deal or it isn’t, right? We humans aren’t good at dealing with this kind of dissonance. We think “bigdealness” should be an inherent property of the issue.
This isn’t the case. You need a whole system of interpretation to make judgments like that.
Harris, using the scientific paradigm, discusses the non-bigdealness of the dangerous claim a fair amount: given what we know about genetics and intelligence it wouldn’t be surprising at all, scientifically, if it was true because we know that many genetic traits like appearance and some medical conditions differ systematically by race and it would be unexpected if no mental traits (that we already know are heritable) did. Harris defends Murray by arguing that to assume a priori that no mental traits have such differences is instead a very strong claim that doesn’t represent the consensus in the field. This is true. Richard Haier, editor-in-chief of the scientific journal Intelligence mentioned in his defense of Murray and Harris’s podcast something called The Default Hypothesis:
I explained in a series of subsequent emails to the editors about the Default Hypothesis — whatever the factors are that influence individual differences in IQ, the same factors would influence average group differences. Since there is overwhelming evidence that genes influence the former, it would not be unreasonable to hypothesize that genes at least partially influence group differences.
To Harris, its location merely an inch off the mainland of uncontroversial science means you should be able to discuss, entertain and perhaps give this hypothesis some credence without being called either a pseudoscientist or a racist (even by implication). And since there’s no scientific reason for privileging a strict environmental hypothesis over the alternative, they should be treated equally. In the minds of the critics however, they should not be treated equally, and crossing that inch constitutes junk science and, in their words, “pseudoscientific racialist speculation”.
While Turkheimer apologized for “junk science” on Twitter, he also reiterated that “it’s just wrong to be neutral” (or “agnostic” that Murray claims to be) between genetic and environmental explanations.
This is obviously for moral and historical reasons. Note that Turkheimer and the others write “the burden of proof is surely on them” above, a phrasing suggesting they find this so obvious that it barely merits mentioning. They treat it as a strong argument in favor of their position instead of the very point under dispute, which is what it actually is.
Klein appears to agree, judging by how he seems almost shocked by Harris saying on the podcast that Murray’s ideas shouldn’t be considered part of a racist lineage. He thinks history justifies giving black people the scientific “benefit of the doubt”, as it were, as reported by Andrew Sullivan in yet another articleon the topic:
In an email exchange with me, in which I sought clarification, Klein stopped short of denying genetic influences altogether, but argued that, given rising levels of IQ, and given how brutal the history of racism against African-Americans has been, we should nonetheless assume “right now” that genes are irrelevant.
On pure substance, there’s not a lot of difference between this and Murray’s statement that he’s agnostic on the issue (but does seem to believe it). Only their opinions on what counts as a permissible interpretation differ. They differ because the interpretive frames they use assign different sizes to the jump from Uncontroversial Science to Dangerous Interpretation. Everyone agrees that small jumps are ok and big jumps are the hallmarks of cranks and pseudoscientists, but not on the size of the jump. That’s because it gets much bigger once you allow political and historical considerations to influence the size estimate.
High decouplers and low decouplers
The differing debating norms between scientific vs. political contexts are not just a cultural difference but a psychological and cognitive one. Beneath the culture clash there are even deeper disagreements about the nature of facts, ideas and claims and what it means to entertain and believe them.
Stanovich talks about “cognitive decoupling”, the ability to block out context and experiential knowledge and just follow formal rules, as a main component of both performance on intelligence tests and performance on the cognitive bias tests that correlate with intelligence. Cognitive decoupling is the opposite of holistic thinking. It’s the ability to separate, to view things in the abstract, to play devil’s advocate.
Speculatively, we might imagine that there is a “cognitive decoupling elite” of smart people who are good at probabilistic reasoning and score high on the cognitive reflection test and the IQ-correlated cognitive bias tests. These people would be more likely to be male, more likely to have at least undergrad-level math education, and more likely to have utilitarian views. Speculating a bit more, I’d expect this group to be likelier to think in rule-based, devil’s-advocate ways, influenced by economics and analytic philosophy. I’d expect them to be more likely to identify as rational.
This is a conflict between high-decoupling and low-decoupling thought.
It’s a member of a class of disagreements that depend on psychological differences so fundamental that we’re barely even aware they exist.
High-decouplers isolate ideas and ideas from each other and the surrounding context. This is a necessary practice in science which works by isolating variables, teasing out causality and formalizing and operationalizing claims into carefully delineated hypotheses. Cognitive decoupling is what scientists do.
To a high-decoupler, all you need to do to isolate an idea from its context or implications is to say so: “by X I don’t mean Y”. When that magical ritual has been performed you have the right to have your claims evaluated in isolation. This is Rational Style debate.
I picture Harris in my mind, saying something like “I was careful approaching this and said none of it justifies racism, that we must treat people like individuals and that general patterns say nothing about the abilities of any one person. In my mind that makes it as clear as can be that as far as I’m concerned none of what I’m saying implies anything racist. Therefore I’ve earned the right not to be grouped together with or in any way connected to nazis, neo-nazis, Jim Crow laws, white supremacy or anything like that. There is no logically necessary connection between beliefs about intelligence and racist policies, and it should therefore be possible to discuss one while the other remains out of scope.”
But “decoupling as default” can’t be assumed in Public Discourse like it is in science. Studies suggest that decoupling is not natural behavior (non-WEIRD populations often don’t think this way at all, because they have no use for it). We need to be trained to do it, and even then it’s hard; many otherwise intelligent people have traumatic memories of being taught mathematics in school.
While science and engineering disciplines (and analytic philosophy) are populated by people with a knack for decoupling who learn to take this norm for granted, other intellectual disciplines are not. Instead they’re largely composed of what’s opposite the scientist in the gallery of brainy archetypes: the literary or artistic intellectual.
This crowd doesn’t live in a world where decoupling is standard practice. On the contrary, coupling is what makes what they do work. Novelists, poets, artists and other storytellers like journalists, politicians and PR people rely on thick, rich and ambiguous meanings, associations, implications and allusions to evoke feelings, impressions and ideas in their audience. The words “artistic” and “literary” refers to using idea couplings well to subtly and indirectly push the audience’s meaning-buttons.
To a low-decoupler, high-decouplers’ ability to fence off any threatening implications looks like a lack of empathy for those threatened, while to a high-decoupler the low-decouplers insistence that this isn’t possible looks like naked bias and an inability to think straight. This is what Harris means when he says Klein is biased.
But Klein is a journalist, and as that he belongs with the literary intellectuals. To him, coupled ideas can’t just be discharged by uttering a magic phrase. The notion is ludicrous. Thinking in moral and political terms is not a bias, it’s how his job works and how his thought works. Implications and associations are an integral part of what it means to put forth an idea, and when you do so you automatically take on responsibility for its genealogy, its history and its implications. Ideas come with history, and some of them with debt. The debt has to be addressed and can’t be dismissed as not part of the topic — that’s an illegal move in Klein’s version of the rules.
Here is my view: Research shows measurable consequences on IQ and a host of other outcomes from the kind of violence and discrimination America inflicted for centuries against African Americans. In a vicious cycle, the consequences of that violence have pushed forward the underlying attitudes that allow discriminatory policies to flourish and justify the racially unequal world we’ve built.
To put this simply: You cannot discuss this topic without discussing its toxic past and the way that shapes our present.
I think Klein takes the historical background’s obvious relevance as a given to the extent that he sees Harris’s lack of focus on it as either an intentional attempt at pretending like nothing is wrong or an instance of gross negligence and ineptitude in failing to acknowledge it. Making that point isn’t slandering in Klein’s mind. Pointing out a connection to Nazism and slavery isn’t guilt by association because THN aren’t making that connection: it’s already there, obvious to everyone not in denial. That Harris gets upset by them mentioning it is just weird.
Klein focuses a lot on historical context in this whole conflict, to an extent that’s similarly “just weird” to a scientist (Harris responds to one of his discussions of America’s racist history with a simple comment that it’s irrelevant). But it’s not weird at all. It makes perfect sense, and Klein getting audibly upset at Harris’s brush-off is telling. The archetypal literary intellectual doesn’t see isolated facts, but narratives and gestalts.
To the archetypal scientist, the effects of a horrific history of slavery and discrimination can be summed up as “environmental factors” — a bloodless abstraction without the emotional punch offered by detailed accounts of the crimes. And that isn’t good enough. No matter how much Murray and Harris admit whenever asked that “yes, environment matters” and “yes, racism exists, which did and do affect the environmental factors” it’s not enough. Klein doesn’t want it acknowledged in some subordinate clause, he wants them to give it the spotlight. He wants equal time, because he knows that if you say that two things matter but only discuss one of them then it’s going to, in effect, grant support to narratives focusing exclusively on that even if other factors aren’t denied. In narrative logic, the point isn’t what facts you believe are true but what narratives you support and what images you evoke.
They get close to what appears to be the core of Kleins’s motivation towards the end of their podcast, when Harris uses an analogy about the racial makeup of world-class sprinters to argue that the possibility of genetic differences means we should not automatically assume racism as the sole reason for aggregate racial differences in outcome. It is to protect a narrative threatened by (in Klein’s earlier words)
the idea that America’s racial inequalities are driven by genetic differences between the races and not by anything we did, or have to undo.
Klein fears that Murray’s ideas will absolve white people for the historical crimes against black people so they no longer feel that racial inequality is their responsibility to correct. In other words it’s important that we do automatically view aggregate racial inequality as a product of racism. Otherwise justifications for anti-racist policies become, while not void by any means, weaker and an order of magnitude more subtle, complex and difficult (the same effect sex differences have on the justifications for feminist policies). That’s why he insists so forcefully that slavery and discrimination is what we’re supposed to be talking about, not genes. Eyes on the ball.
Harris of course has a narrative of his own to protect. He fears that politicization of whole intellectual fields (including sometimes science), pushed by identity politics activists using their norms, will lead to the displacement of Rational Style as the norm. Klein would say that Harris fears this because it would hurt him personally, but I don’t think that’s the whole story. Harris cares about things besides himself, just like Klein. He cares about protecting Rational Style and the high-decoupled thinking it runs on because it is tremendously valuable for finding out the truth and for ensuring civility in the public sphere. Unlike identity politics activists, he doesn’t think such rules lose their legitimacy because power differentials exist.
How to tie this together? I think by bringing us back to my entry point: Harris and Klein’s email conversation.
I’ve reread it after having finished this article and my final impression is that both of them fall short of ideal disagreement conduct in different ways.
Harris lets his anger get the better of him, at least near the middle of the conversation. He accuses Klein of cynically engineering the controversy for clicks and insists that Klein knows what he means and is just feigning ignorance, which I’m doubtful of. I also think he made a mistake and overstated his case in his Housekeeping podcast where he insisted the Vox article was intentional character assassination. I don’t think that was its primary purpose, and being so strident about it hurts his case by offering an easy way to dismiss him.
Another problem is his handling of important clues. When Klein says “something is very off” he answers with:
The thing that is “very off” is the highly moralistic/tribal posture some people take on every topic under the sun, which makes rational conversation on important issues nearly impossible. If we do a podcast, that should be the central topic of conversation.
That was a thread that could’ve helped unravel the conflict. Instead of pulling it he doubles down on confrontation. He does this social equivalent of trying to fix a tangled cord by yanking it really hard, getting himself stuck in discussing the scientific evidence instead of the differing interpretations of his podcast what led to them.
I’m being a bit tough on Harris here because as a fellow high-decoupler he has my sympathies by default and I’m trying to be fair, but Klein doesn’t come off well either. He seems to have a better grasp of the situation and remains cooler and appears to be the better communicator — at least in the beginning. But as the exchange continues, the accusation that Klein dodges and stonewalls seems more and more valid. Since he does seem to have some idea of how Harris sees the situation it’s all the more frustrating how he refuses to give an inch and grant any validity to the view that Harris (and Murray) has been dragged through the mud. “They’re still successful, so it doesn’t matter” seems to be his attitude. His reasons for refusing to publish the Haier piece also remain weak and transparently partisan.
Their later conversation on the podcast wasn’t much better. They spent two hours talking past each other. Harris says in his intro that he did his best, but it didn’t work.
The roots of their disagreement are different ideas about what’s acceptable conduct in debates, and whether we should be treating politically relevant scientific ideas in a high-decoupled or low-decoupled way. I’m disappointed in them for not honing in on this during the podcast. Both make slight movements in the right direction, but nothing comes of it.
They might just be so enmeshed in their respective spheres that it’s become difficult for them to bring their unstated assumptions up explicitly, so instead they engage in indirect proxy-argumentation. This is a common pattern: you think your interlocutor implicitly believes something you disagree with, so you try to disprove that with counterexamples (like Harris did with his rhetorical “do you think I’m inferior to John von Neumann?” that didn’t land). It doesn’t work, because as long as it’s implicit people dodge it and focus on the specific point instead of the idea it’s meant to adress. It has to be tackled head on. Make it the subject.
This requires that you focus specifically on dissecting the disagreement, and not on pushing your own points. Someone suggested to me that they’re bad at this because they have big egos. Maybe that’s true. Maybe it’s also true that they know they have an audience, which damages the quality of their conversation because good disagreement requires stepping out of rhetorical, point-scoring mode, which is difficult when there’s something at stake.
Truth be told, most people just aren’t big enough nerds to disagree about complicated things effectively. You need to be much more literal, much bigger sticklers with semantics (because semantics are important) and much more eager to spend time pinning down exactly what the other person thinks — even if this is boring, pedantic and doesn’t lend itself to monologuing the way both of them spent too much time doing during their talk.
They use more advanced argumentation than the situation permits. When people are nearly on the same page complex arguments work because they are understood and received correctly, but when people don’t understand the subtleties of their differences that well and good faith is in short supply, you need to advance incrementally and with quick, tight feedback loops to make sure you don’t go off the rails. Harris and Klein went flying into the countryside shrubbery way too many times.
In the end these problems plagues all complex communication. Perhaps full comprehension is not possible when a message becomes too big for us to fit the whole of it in our short term memory. A two hour podcast, an article of a few thousand words or and often even something as short as a few paragraphs of speech cross that line. Selective, lossy compression through interpretation and narrativization becomes necessary. With that comes inevitable bias.
In your response to me, it’s clear you thought I couldn’t possibly have heard the original discussion to think that this piece was fair, which means I’m either a terrible listener, or the discussion landed differently on some listeners than you think it did, or both.
Klein’s right, except that there’s no “or”. Things certainly did land differently with many people than Harris expected, and Klein is a terrible listener. Both are true because we’re all terrible listeners.
Including me. This account, as long as it is, is far from complete. I’ve ignored some things, cherry picked my examples, connected my dots and filled in the blanks with my own projections, all in the service of creating a narrative. I’ve tried my best to be fair.
• • •
This controversy is so large and sprawling and touches on so many separate disagreements that it’s impossible to do it justice even if you restrict yourself to just applying high-level abstractions. There were many relevant issues I had to leave out, including:
The difference between criticism and discrediting, whether the second amounts to silencing, and the implications for free speech.
The trolley problem-like distinction between putting forth words that end up hurting someone and hurting someone with words.
The justifications for Activist Style vs. Rational Style argumentation in relation to identity politics.
How, like I’ve said before, it’s essential to productive disagreement to acknowledge the legitimacy of the other party’s views before criticizing them.
The difference between judging conclusions by their arguments and judging arguments by their conclusions.
The question of whether Harris is being hypocritical when not wanting Murray’s ideas to be seen as part of a larger pattern but doesn’t mind making Vox responsible for the moral panic it’s part of.
How the mechanics of partial narratives explain some of the dysfunction of genetics vs. environment debates.
The difference between science and the interpretation of science and its intersection with the difference between science and pseudoscience.
The symmetry and asymmetry between Harris blaming Klein for indirectly suggesting things about him and Murray, and Klein blaming them for indirectly suggesting things about black people.
That while preference for high or low decoupling is a cognitive style, most people are capable of either when it’s in their interest.
What high decoupling looks like in the political realm.
What responsibilities sender and receiver have when they disagree on the proper interpretation of a message.
How individually innocuous ideas can add up to dangerous gestalts, and what that means for responsibility attribution.
I understand it better in retrospect. It’s been clear that Harris is extremely aware of the dangers of getting yourself connected to something connected to something connected to racism, and he likely thought it was wisest to talk about this as little as possible.
The closest thing that exists is a book I never tire of recommending. Defenders of the Truth: The Sociobiology Debate by Ullica Segerstråle is a fantastic and very thorough discussion of the fights around EO Wilson’s book Sociobiology in the 1970. It’s very much still relevant today and I can’t recommend it enough to anyone who enjoyed this article.
As I’ve mentioned in a previous case study, the word “race” weirds me out. In my mother tongue it’s an archaic term (outside some unsavory circles), except when referring to breeds of domesticated animals. Words like “ethnicity” or “people group” are used instead. I’ll still use “race” here since it’s used in English and can’t really be avoided.
In support of this Harris notes that claims 1-5 have often been attacked, sometimes for the the purpose of undermining (6). Criticisms asserting that intelligence isn’t even real, that IQ tests don’t measure anything or that they’re are culturally biased and that environment is all that matters have been and still are common (and Harris and Murray spend some time talking about them). The popularity of these arguments suggests that (6) is considered at least somewhat implied by 1-5 by others as well.
I learned and extremely important lesson from this: Anything contentious, anything that’s an even slightly complex or subtle argument with ambiguous implications has to be read several times. Everything here that I’ve read more than once has felt distinctly different the second time. When you don’t have to take in the surface level information you can observe yourself reading or listening, and this makes for a different experience.
I don’t think Harris agrees this is the reason for the title, his “nested taboos” comments suggests he considers the whole range of 1-6 to be “forbidden” to some extent. I also don’t think Harris agrees about what the “thrust” of the argument (or that there’s a single overarching argument going on at all) is either — because this is highly subjective; the “long discourse” mentioned is under one minute long, less than a third of the length or their later discussion of how immoral and irrational it is to treat people as members of demographic groups rather than as individuals — a point Murray refers to as shockingly difficult to get to stick.
What I want to bring up here that no one appears to have mentioned about this, is that there’s a difference between 1), policies that assume everyone is equal above the neck, 2) policies that assume demographic groups are unequal above the neck in some defined way (which would be genuinely discriminatory) and 3) policies that assume neither of those things. I think many who advocate 3 over 1 are read as advocating 2 over 3.
This part in particular was interesting to write. Because of how convincing both Klein and Harris are on this point (Murray’s focus and motivation) I’m genuinely uncertain about what to think of it, what to think of him and which account should come first — in other words, which should be the signal and which should be the corrective. It was an unpleasant feeling at first and I know people, me included, will do anything to quell such uncertainty. But then I told myself I didn’t need to make a judgment here, that it’s fine, even virtuous, to simply remain agnostic (and it might be more accurate than anything else). The unpleasantness disappeared right away. I recommend doing this.
I agree with Harris that there’s something dishonest and evasive about Klein’s conduct. It just think there’s genuine conviction underneath it that I prefer to focus on.
Transparadigmatic dissonance is a common reason for particularly tricky disagreements, particularly on topics we’re not willing to nor good at discussing openly. Like sex. Transparadigmatic dissonance about the nature and purpose of sex and relationships gives rise to incompatible ideas of the moral significance of promiscuity, abortion, contraception, pornography and infidelity etc.
Harris accused Klein and the others of being biased here and he has a point. Klein calls the Murray podcast “curiously ahistorical” for not mentioning slavery, segregation or past racism in science, but I very much doubt he’s call a similar discussion focusing exclusively on the historical background “curiously abiological”. It’s a common pattern. I remember taking introductory psychology and hearing some people loudly complain that the chapter on the biological basis of behavior ignored culture. When we got to the chapter on cultural factors, they did not complain that it ignored biology. The justification for this asymmetry is of course political implications. Whether this counts as a bias or a valid concern is the difference between scientific and political thinking.
THN are of course scientists and thus not typical literary intellectuals. I still think the distinction holds, as they don’t rely as much on low-decoupling as Klein does in their argumentation. Even if they would have, it wouldn’t fully disqualify the model since even among scientists there are lots of people who are quite capable of using low-decoupling when they find it justified — it’s our natural mode of thinking, after all. Note that Stephen Jay Gould, author of the (in)famous (and historically focused) anti-IQ book The Mismeasure of Man is known for his unusually ‘literary’ style for a scientist, which contributes to his popularity among nonscientists (and nonpopularity among scientists).