Lee Jussim: “Actually, stereotypes are often accurate”

Ajouté : 2026-05-09

[00:00:01]My guest today is Lee Jim who's a distinguished professor of psychology at Rutgers University and one of social psychies most persistent internal critics and he's also a fellow Substacker. So go and check out his excellent publication unsafe science on Substack after the podcast. I highly recommend it. For decades, Lee Jim has challenged a powerful story inside his own field. Social psychology often treated stereotypes as false by definition, implicit bias as a hidden force explaining discrimination, and ordinary human judgment as something suspect, something experts had to diagnose and correct. Jim's work points in the opposite direction. He argues that many demographic stereotypes are in fact statistically accurate, that this is one of the largest and most replicable findings in social psychology, and that the implicit association test does not measure what its strongest proponents claimed it measured. Hidden racism that reliably predicts discrimin discriminatory behavior. It doesn't show that at all.

[00:01:12]Ordinary people in his telling often read social reality far better than the discipline wanted to believe. In this conversation we talk about stereotype accuracy, implicit bias, DEI, academic conformity, and what happens when a science built to study bias becomes blind to its own. It's an American debate, but not only an American one.

[00:01:35]Much of what Jasim describes is just as relevant to Swedish universities and research. Ideological consensus passing as expertise, fragile findings hardening into policy, and the difficulty of questioning either. But now onto today's guest. You're listening to Ra Kger with me, Ilvarpi.

[00:02:18]>> [music] >> Welcome Lee Jim to >> Good to be here. Thanks for having me.

[00:02:29]>> And I just like I I I noticed you a few years back when really when I started reading Jonathan height and Philip Tetllock and you also you were like writing papers together with them and when you when you first sometimes I feel when I first notice someone I feel like they are new as well because I'm new to them but then >> yeah I'm old I'm old. Yeah I'm totally old >> and then it turns out you're really old.

[00:02:55][laughter] >> I'm old. I turned 70 this last year. I mean, I'm old.

[00:02:59]>> Yeah. Yeah. [laughter] Yeah. So, you've been doing this and then I I started reading up on like what you've been what you've been up to even before I I noticed what you've been writing and the interesting papers and books that you've been writing and >> and you've been sort of a s some people start to be like become like a sore point in academia late [laughter] in their careers so that their their colleagues get annoyed with them like later on.

[00:03:26]>> [laughter] >> But I think that maybe you've been you've been annoying for decades colleagues.

[00:03:32]>> Yeah.

[00:03:32]>> So like when did you uh like like biographically and intellectually when did you start to realize that something was wrong within your own field of social psychology? probably the the biggest turning point for me was started out interested in and I'm still interested in um social stereotypes that is people's beliefs about groups and especially how and whether those beliefs influence bias distort how they judge individuals from those groups as the earliest some of my earliest stuff in graduate school in the 80s was on that reading the literature on stereotypes was vividly clear at that time and I'm again at that time I I'm not you know I'm not a gadfly at that point I'm just I I I'm just doing I want to understand how stereotypes and when stereotypes bias judgments of very conventional social psychological things to be doing so but there's all this stuff on all these papers claiming that stereotypes are inaccurate and at that time I assumed they were they knew what they were talking about like I was just a lowly graduate why would I not assume senior faculty writing in prestigious outlets declaring stereotypes are inaccurate would be wrong. Why? You know, or wouldn't know what they were talking about. I they wouldn't even occur to me. All I wanted to do at that point was to get to the bottom of those claims. So, what's the original research finding that stereotypes are inaccurate and they lead to powerful and pervasive biases and judgments of individuals?

[00:05:06]Because those were the claims common in the 80s. And I I go into the literature and I find what I ended up calling a black hole and it with the black hole at the bottom of these declarations that stereotypes were inaccurate. So, and by black hole what I mean is some large portion half the claims that stereotypes were inaccurate didn't cite anything to support the claim. They just they just stated it like it's like saying the sky is blue.

[00:05:36]You don't need a reference to declare the sky blue. You don't need a reference to declare stereotypes are inaccurate.

[00:05:41]But they they don't seem like the same kinds of claims.

[00:05:45]>> We hold these truths >> to be self-evident. Yes. Right. Exactly.

[00:05:48]>> Like in the declaration, >> right? Right. Now, you could define stereotypes as inaccurate. Then of course they're inaccurate by definition, but that creates other problems which you asked me how I got this. So, so let me hold off on that.

[00:05:59]>> Yeah. Yeah. Yeah.

[00:06:00]>> So, so there's all these articles declaring them inaccurate without a citation.

[00:06:05]Okay, I don't know what to do with that.

[00:06:07]But then maybe half the articles have a citation. So then I would go to that citation. I'd go to that article, the article being cited. And that article would declare stereotypes to be inaccurate without an empirical citation. And at the bottom of all this was nothing. Just there was no empirical evidence there or almost no empirical evidence. So in so that first got in trouble, not not in too much trouble yet, but a little bit of trouble for saying, well, if st if psychologists want to declare stereotypes to be inaccurate, they should get evidence on this. And that was like, you know, that some people got pissed off. It's like, you know, this is not the kind of thing you need evidence for because guess I guess because it was self-evidence, self-evident. So then we we I and my several collaborators publish stuff making conceptual arguments why stereotypes wouldn't necessarily always be inaccurate and taking the field to task for making these claims without evidence that if they want to make the claims that's fine they should just get some evidence that supports it.

[00:07:12]>> Okay. So, so I and others there actually before me probably the first begin getting evidence on the accuracy of stereotypes was Clark McCauley and I think he he may be retired now you know he's like 10 years ahead of me 15 years ahead of me he's you know he's not no spring chicken at this point he's got to be close to 80 I mean he was publishing on this in 1978 you know in 1978 I didn't I I had dropped in and out of undergrad So in 78 I wasn't even in in college. So, so he had a small number of studies on both sex and race stereotypes that found people were very very accurate that if you ask people, you know, what they believe about he you were he would purposely choose criteria for which there were objective criteria objective like the census data you know how many how many uh black Americans have a high school degree you know how many black Americans are on public assistance you know these are things for which there's they're very clear data strong data on um and if you would then ask people you know what they think about you know what's the proportion of black Americans with high school diplomas and people were very very accurate they were very very accurate they weren't perfectly accurate but they were very very accurate you found the same thing with sex stereotypes again this was all before I came along or right maybe as I was beginning to become aware of this stuff and that eventually triggered outburst of research on the accuracy of stereotypes, not just by us.

[00:08:46]I continued to do some of it. So did he.

[00:08:48]But there was all sorts of stuff after that for, you know, starting maybe in the late 80s and early 90s into the 2000s. Um there were just scores of papers assessing the of accuracy of all sorts of stereotypes. You the basic demographic stereotypes, you know, age, sex, race. But then, you know, sometimes people found it useful to study these things kind of experimentally. So they wouldn't they use different kinds of groups. Um there were a series of studies on the accuracy of beliefs about sororities and fraternities. So that's good because they you know the sororities and fraternities are the advantage of that is they're not involved in like social and political issues in the same way that race and sex might be. But it didn't really matter.

[00:09:33]You know completely 100%. Most of the time, most of these studies found at least moderate, sometimes very high accuracy in people's beliefs about groups. It was really kind of staggering. So much so that just within the last year, a meta analysis came out of the accuracy of gender stereotypes that completely vindicated all that. It was by two women, both of whom have, you know, kind of both strong social science but also strong feminist credentials.

[00:10:02]the met analysis shows that gender stereotypes are moderately to highly accurate you know so so but that you know that that that was my first so that was how I discovered that's how I began to discover that my field was riddled with these dysfunctions so by which I mean you can't be in the business of making empirical claims without empirical data like what the hell you know what kind of scientific field like we claim to be a scientific field and yet people are just making stuff up. So, and then I would say that more or less I mean I'm I'm talking in sort of casual language now but in academic papers I would say essentially the same thing. So of course that pisses people off and you know the people in my field didn't want to hear that. So they they doubly didn't want to hear it. They didn't want to hear that stereotypes had had this high accuracy component and they didn't want to hear me saying the field is in a is a mess. So they didn't like either part of that. So that's that's how it began.

[00:10:58]>> Like the stereotype accuracy, the whole the notion that stereotypes or biases towards groups can be accurate. It sounds like uh provocation almost.

[00:11:13]>> You know, it sounds like something like if if you're in a sort of a an argument in a pub or or a bar [laughter] and then you have like a loud guy saying like but you know you know it's right, you know it's correct, right? you know like but the stereotype is correct >> but like it's not it's not it's not so it's not so simple what you are suggesting though could you so could you go into a little bit more >> yeah it's not so simple so first place that you know when I say the evidence finds that stereotypes are accurate we're talking things like moderate to sometimes high correlations between people's beliefs and criteria so moderate correlation might be 04.5 which translating that to you know someone who is not familiar with correlations might mean people people's judgments are right maybe 70% of the time something like that sometimes the correlations are higher so the percentages would be higher but you know 70% is like pretty good compared to like inaccurate which is what I was arguing against but it's far from perfect even at 70% um which means to me even though from a scientific standpoint it is reasonable to say that in general with some exceptions stereotypes are at least moderately, sometimes pretty accurate.

[00:12:27]That doesn't mean any given person, including, you know, a guy drinking a beer in a bar, should assume that his beliefs about particular groups are necessarily accurate. That actually would I mean, they may be accurate. He may know what he's talking about, but that would require comparing his beliefs against some sort of evidence. Um and just just you you can't assume you can't assume that just because this is generally true that it's always true or it's always true for any particular individual. You can't you just can't assume that.

[00:12:58]>> Hey Fore pod new food chhatter needed and like it seems like social psychology got because I read uh sociology in the early north and uh a lot of it like was about these these things of course social psychology and you get taught at university and I I would say in like the public discourse at least in Sweden a lot like your bi check your biases.

[00:14:05]>> Oh yeah. And you have these like educational leaflets from from the government or uh or government agencies like actively trying to work against the bias that they presume that people have, >> right? And so it seems like both the science and the sort of the public discourse has been very invested in like you shouldn't be biased and you shouldn't and your biases are wrong like and you what your science what your research seem to suggest is that at least in part that might be morally correct in some way but >> Yeah.

[00:14:44]>> Yeah. But but why do you think how does your your research rhyme with the whole sort of progressive mission of like against antibbias work?

[00:14:55]>> Well, I you know rhyme is I'm not sure that it does although I'm not sure that it doesn't either. So as a moral statement I completely agree that agree with that.

[00:15:06]Yeah. You know you it's it's bad to be it's bad in all sorts of ways to have biased judgments of other people. So it's certainly bad in high stakes contexts like college admissions or employment or something like that and and we are most of us are imperfect and we are potentially vulnerable to all sorts of biases not just stereotype biases and so it does you know for a person who understands that I think it's useful to understand that we do have these to really understand it not as like a you know okay we all have biases but now let's move on that we do and that in important contexts there are things one can do to limit the effects of those of the potential effects. You may not even have these biases but you don't really know you know it's like one thing to conduct an experiment and you have whatever 500 participants and you can see that you know there's a difference in how these 250 versus this 100 but when you live your life you're doing one thing at a time. You can't live your life as a controlled experiment. So, it's very hard to know for whether your own judgments are biases, but but it's not really that hard to do things to limit your potential for bias, especially for things like college applications or work work applications, evaluating people for jobs or promotions. There's a very simple there's no guarantee there's no guarantee that any particular thing will eliminate bias, but there are things you can do to to to that that have a very high potential. Those are mainly for for things like college and work. It's focus on the merits and competence of the person. I mean, that's what you should be focusing on. Whether it doesn't matter whether a man or woman or what their race is. Uh what ma what should matter is how good they are at the job or how good they are at the precursors to the job. um you know if you're hiring an entry- level person or college and usually you either easily have those things at your fingertips certainly for college admissions and for most employment things you're going to have a person's resume um so you can see what they've done there's other ways you can also evaluate a person on their on their merits but if you focus on the things relevant to whatever the decision this personnel selection type thing you're doing then you know you're going Again, there's no guarantee, but you're going to do everything you can to limit that uh potential. And and you know, that's you know, if you want to if you want to admit the best students, you know, if you want to hire the best person, that's what you want to do. You want to focus on what they're capable of, not their demographic.

[00:17:38]>> Yeah. I mean it's sort of the whole discussion seems sort of it sort of lynches or it sort of rests on an assumption that the the biases are true because a lot of the progressives or what do you want to call call them they want to further groups that are disadvantaged or whatever term you want to use. So that implies that they like many of these biases are correct >> right? So, so because if you I mean if if let's say African-Americans, they're poorer, they're more prone to commit violent crime. If you have these kinds of uh you make these social claims, but then as a as a progressive, you would say, "Yeah, but that's why that's the reason we need these government programs. That's the reason why we need um affirmative action or we need uh diversity training and because that's the reason we we do it. But but you're not supposed to say the bias out loud but the whole argument rests on it being true basically.

[00:18:46]>> Well, so I I think you know doing my best to channel the progressive view which I'm probably not the best person to do that. [laughter] They were they were >> that's a that's the great great interviewer who forces the interviewee to like [laughter] channel the opposite view after like 20 minutes.

[00:19:04]>> Yeah.

[00:19:04]>> Right. I think they would many anyway would argue that the discrepancies in the academic records of of young black people compared to others Asians, whites, whatever that itself stems from racism and bias in the system. And so, yes, those gaps, those differences, you know, differences in high school degree, differences in GPA, differences in standardized test scores, they exist, but they exist because of racism. Um, and so so these this treatment is merely compensation for the biases that already exist in the system.

[00:19:44]>> So, it's not really they would argue it's not real. one two things I think they would dispute that there are genuine underlying differences in intellectual or academic capabilities or interests between groups or motivation other than those caused by racism and that um those diff because those differences are inculcated by a racist system. social action, programmatic action is required not to provide advantages to black applicants, but to compensate for the the discriminatory disadvantages that they've had to face.

[00:20:21]I I think that would be the argument.

[00:20:23]>> So, it's basically furthering meritocratic ideal, right? Rather than rather than uh >> That's right. It's impeding. It's saving me.

[00:20:33]>> It's like saving me. I think that would be the argument.

[00:20:36]>> Yeah. I think I think speaking as a a non-progressive I think I think you nailed it. Yeah. I I think this ties into another thing that you've been written about and others as well because one of the most influential ideas to come out of social psychology in recent times is the the notion of implicit bias and the claim that people even though they reject prejudice, they still car carry unconscious associations that shape their behavior. like that reinforces the system that we we just talked about. It's like I wrote about this like 10 years ago I think or or something like that in Sanska Dogbot which is a a daily newspaper in Sweden and and I went through like the claims I think I sort of probably referenced something you wrote. I don't remember but like okay these are the claims but they don't this very it's very tenuous and it's not clear if this is something that's really proven. The same day or the day after I published my essay, we had an uh mandatory course at the newspaper to teach us all the journalists and at least it was mandatory for the the bosses, but it was highly recommended for everyone else to like have a a core lecture about our implicit biases. And it was just such a so clear that there's like you have one level like in the public discourse you can like discuss it but you have another like within the HR departments and it's just like rolling out but I would like to ask you like is uh what's your view on only implicit implicit bias and the science is it like what what are the claims and how much like evidence is there for it? Well, so the early claims the workhorse method of assessing implicit bias is the IAT, the implicit association test. So without going into all the technical details that was p first published in 1998 and so you know it's been almost 30 years that this notion of implicit bias has been you know pushed in the literature and it really took off. there was a small and then growing cadre of people who thought this was a great method. So there was this outburst this explosion of research using the IAT to assess implicit bias in the 2000s into the 2010s. You know that kind of research has a unique insulation from critique from scientific critique and that is a version of we're assessing implicit biases to study racism. If you critique this, you're opposing our anti-racist efforts efforts. And so you're an anti-anti-racist, which is that is your, you know, why would who would do this except a racist, right? And so that is kind of >> why is this why is this so important to you?

[00:23:38]>> That that Yes, that's right. Why? Well, it's more than just why is it so important. You're you're saying, you know, you're criticizing it. So you're doing anti- anti-racism. So, you know, so two negatives is a positive. I mean it's this very simplistic thing but that was what happened and so it was very very difficult for people to publish critiques in the 2000s and 2010s >> but some did get published >> and you know they'd get published and then they'd be you know supposedly refuted you know there'd be more reputations than there would be critiques but the critiques didn't go away because the problems they identified were real.

[00:24:20]>> And starting around 2015 or so, more and more critical skeptical stuff about implicit bias in the IAT in particular started to get published. So that's like 20 almost 20 years from when the stuff first came on that this sprouting of you know really thoughtful scientific methodological statistical critique.

[00:24:43]Okay, that's the general that's the lay of the land. Your question was what are the claims? So the original claims you know circa 1998 2000 2005 you know this is the era before the there was much in the way of critique it's the era where the critiques were dismissed as as you know as just racist and nobody needs to pay attention to it. So that's when the the implicit bias began to get got a lot of scientific traction and then even public traction. uh it was Hillary Clinton referred to it in our 2016 campaign, you know, that we need to I don't remember the exact quote, but that uh it was some condemnation of implicit bias.

[00:25:21]>> Okay. The claims were things like 80 to 70 to 90% of Americans are unconscious racists.

[00:25:29]Um uh that the IT uh uh uh captures something different than explicit measures of prejudice. So explicit measures just a questionnaire measure.

[00:25:39]So well like one of the most common questionnaire measures for prejudice is for to ask people to rate various any group on what's called a ceiling thermometer might go from 0 to 10 or 0 to 100. You know it's a temperature. So the higher the temperature the warmer you feel to the group. That turns out to be a very very good assessment of explicit prejudice that does a very good job of predicting discrimination actually. So saying the critique of implicit bias is different than saying, "Oh, well, you know, prejudice and discrimination are dead." You know, implicit bias can be a a kind of methodological and scientific borderline and maybe more than borderline disaster area. That can be true and there still could be prejudice and discrimination.

[00:26:22]It's just the method is is bad.

[00:26:24]>> So anyway, 80 90 70 90% of Americans are unconscious racists. the uh the implicit measures IIT there's others besides the IIT capture something very different than explicit measures and really strongly predict things like discrimination and that real world gaps are a function of implicit bias to some degree. So those were the you know those were the those were the common claims and I think that in part to the extent that institutions were sincerely motivated which I'm slowing down here because I don't know how much they were sincerely motivated to do something about this. There's a paper just came out which I've only read the abstract that much of the push for implicit bias trainings was by elite white liberals and it seemed to be serving like a social signaling mechanism much more than a let's actually do something about racism mechanism. So that's why I'm hesitating. But to the extent that it was sincere, which is, you know, for some people it probably was to some degree sincere, if you believed the original rhetoric around implicit bias, if you believe that, you know, 80% of Americans are unconscious racists and that this is, you know, really a powerful predictor of their discrimination, you might come to believe that, well, one way to address this is to bring this unconscious bigotry to consciousness so that at least the well-intended people could do something about it. I ever think that's the you know that the to the extent that it was earnest and not just like this virtue signaling on the part of elites that would be the logic the but that logic has several problems one this I'm laughing because you can't make this up it's now over 10 years ago 2014 paper comes out finding so this is the first in 2014 they probably did the research in 2012 or 2013 these were the first people to test whether what's assessed by the implicit association test is unconscious.

[00:28:27]>> It's just by assumption. It's sort of like the stereotype and accuracy stuff.

[00:28:30]It's just they have this test and they assume it's unconscious. So they tested they asked they told people what the IAT was and asked people to predict their scores. Their predictions were massively accurate. Those correlations like seven and higher. So people know if people know what their IT score is going to be.

[00:28:48]The idea that it's unconscious is kind of falsified. It's really pretty strongly falsified. So, they're not unconscious.

[00:28:55]That's number one. And then all this stuff comes out showing that this the scores themsel IT scores themselves are biased. So, um again, without going into all the details, scores on the IT range from negative to positive. So, a negative score might be pro black, a positive score might be pro white. And the presumption was that zero was egalitarian. Right? It, you know, it's sort of like, it's like a discrimination. It's a judgment task, but it's sort of like discrimination.

[00:29:23]It's okay. Um, but it's a it is a reaction time task. People are categorizing stimuli into categories.

[00:29:31]And whether zero actually corresponded to egalitarian behavior was never assessed. H >> it could have been assessed because there were studies that gave the IAT and also assessed discriminatory behavior.

[00:29:48]So they could have identified through simple correlation analysis really regression but it's a simple a fairly simple analysis what is the point on the IT that corresponded to the egalitarian behavior. They could have done that never occurred to anybody to do it. So around the time that the criticisms start picking up around 2015 or so, one of them does this with like three or four papers and finds that rather than zero, the point of egalitarianism flips around from study to study and ranges in the pos it's in the positive range like 3.6 something like that. I mean most uh the average IAT effect I don't know is probably around 0 2 or 3. Again, what that means doesn't really some sort of it's interpreted as pro-white bias, but what they're finding is scores.3 and above are are corresponding to egalitarian behavior. So what that means so so the claims that 70 or 80 or 90% of Americans are unconscious racists was based on scores above zero. Scores above zero are anti-black bias. But if zero is not the cut off for egalitarianism, all the 70 80% claim is nonsense. It's the people above.3 or 04 or point 6 which is going to be a lot lower than the people above zero. So then the 80% of Americans are unconscious racist. That was nonsense also. And there's a paper that just came out like a month or two ago.

[00:31:13]And that was probably the single best paper on the IAT and discrimination that I've seen so far. It was it has a big team and it was an adversarial collaboration. So should I explain what that is in adversarial collaboration?

[00:31:28]>> People who are usually not the best of friends or agree on >> yeah well >> the basic assumptions political assumptions but you you could explain but I wanted I wanted to be graded apparently.

[00:31:43]>> Yeah.

[00:31:44][laughter] >> I mean that's basically right. It doesn't have to be political. It could be theoretical. You know there is an intersection of political and theoretical. You know, this anti-racist stuff is both empirical and theoretical.

[00:31:55]You know what? It's a theory. How unconscious are IT scores and how much does that predict bias and discrimination? It's a mini. It's not theory like theory of evolution or relativity, but it's like a mini theory of how racism works. And yes, to collaborate, there are two main ingredients. the to have an adversarial collaboration, there need to be people who disagree, you know, who have probably um published opposing viewpoints on this issue. Right now, we're talking about the power and importance of implicit bias. So, they would disagree about that, but they'd agree that this is subject to empirical test, and they'd agree to work together to provide the best tests of that.

[00:32:38]>> Okay? So, that's what this group did.

[00:32:39]That's who this group was, and that's what this group did. They did a slew of studies um which I recently summarized as part of a talk. So, so they administered the IIT four different measures of of behavioral discrimination as well as explicit measures of prejudice, you know, like this. I don't remember exactly what they were, but they were just questions like the feeling thermometer that I just >> how much do you like this group?

[00:33:03]>> Yeah. Yeah.

[00:33:04]>> How warmly do you feel towards these people? So, you could correlate with actual behavior.

[00:33:09]>> That's right. Exactly. So, so here I'm I'm actually reading off my slide. So, across the four me measures there was no anti-black discrimination. Actually, on three of the measures they had problack discrimination and one there was egalitarian. So, that's the start. This was also a registered report. So, what that means is that they had to lay out what they planned to do before they actually did it. and the journal that published it agreed to publish it based on their plans as to what they were going to do. And what's great about this is it ties researchers hands. They can no longer make stuff up after the fact.

[00:33:46]Like this was our standard. This is how we were going to test it. If the results were this way, it would mean this. If the results were that way, it mean this.

[00:33:53]Okay. So, continue. So one of the things they did was they used both the implicit measures which included things beyond the IAT as well but they they had several implicit measures and the explicit measures and they would use both to predict behavioral discrimination right that could go either way right I mean if it's a low enough measure it could predict pro prob black discrimination and if it's high enough it it would correspond to uh anti-black discrimination so they by using all of those together they could look at how much each predicts over and above the other. So how much do implicit measures predict over explicit ones? How much do explicit ones predict over implicit ones? And what they found is no matter how they did it, the explicit measures predict powerfully predicted discrimination. They like 56 these are high effect sizes. Whereas the implicit measures, the effect of the implicit measures when controlling for the explicit measures was on the threshold of trivial. So they had an effect size in advance that they said, "Okay, this is below this is trivial. Above this is non-trivial." It was right at that threshold. It was right. So I can't say it was below the threshold of trivial, but it was right at the threshold of trivial. Okay, that's the second thing.

[00:35:10]the explicit measures and maybe I said this predicted discrimination far more than did the implicit measures and that the zero on the IAT corresponded to anti-white bias not not to egalitarian treatment that there were such a good study there's more beyond that but those would be the headlines so you put all that together so you're you're being subject to an implicit bias training and it's at this point it's probably mostly nonsense because it's probably not based on the most recent data it's probably based on this from the 90s and early 2000s where you know because it takes a long time for that to percolate into the real world. And so they're probably covering all this stuff as if you know implicit bias is this powerful pervasive you know deeply predictive of discrimination unconscious phenomena for which there's no currently no evidence.

[00:36:00]>> I mean it also ties into like a view of people or like people in in our current times. I mean, if you have a a view, I don't know, in the 1900s, at least sometime in the 1900s, that you could become a better person, that people are mostly rational and can make rational decisions. Now you have these sort of bias trainings, diversity training and also I've been writing a lot about like racism and diversity in Sweden and uh the training and the the public discourse or the discourse from the political or and intellectual level and there are mandatory courses that you have to go through and you have to get like uh certified if you want to be a practitioner in the in a hospital for example and if you ask critical questions as as you said uh before before like if you ask the critical questions then you're actually an anti- anti and that cancels both out. So you're like you're anti- anti >> Yes. So now now now you're a pro-racist.

[00:37:02]You're racism >> and you can be misogynist anti-trans like it's it's even though even though you are like protrans, you could be like protrans in this way like everybody has their right to decide for themselves but you can still be critical of certain claims >> absolutely >> in the field. And it's the same goes for feminism or the gender sciences. like you could be uh I myself I I am I'm I'm for I'm probably on the extreme left on a global scale but in Sweden because you question sort of premises or doctrines you become like an anti-feminist or you're like misogynist because but I was because you've been been a researcher for for long and you're you're in in these very contentious and fraught questions and you're researching them like do you agree that the the view it ties into like a view of people that they are you can't really trust yourself. You have to trust us to make the like we have to train you and you have to listen to us. Am I being too I recently just listened to a podcast about 1988 by George Orwell. So maybe I'm I'm making too much of it. But it seems like it ties into a view of human human being.

[00:38:12]>> Yeah. I No, I don't think you're making too much of it. I mean it's a little bit of a stretch but it's a stretch I can get on board with. I you know in the in the so I'm I'm more familiar with the United States than other I mean I lived here my entire life so I you know if you go back a hundred years like part of the culture was that Americans have a can do attitude you know right that that American knowhow was a thing you know right and and we were pr Americans were we I wasn't even born yet but but Americans were proud of that there were obstacles and Americans could overcome it and you know kids a kid who grows up poor can grow up to be president of the United States which at Lincoln actually did, you know, right? Yeah. So, so >> that was all part of the culture and it seems gone like you don't hear that kind of thing anymore, you know? It's all it's sort of like you were talking about at the beginning of this podcast was, you know, check your privilege. And so, what is the whole privilege discourse is it's not that somebody has accomplished something, it's that they have unearned advantages. And, you know, they're not really mutually exclusive. I mean maybe some of us do have unearned advent to some degree and yet you can also have accomplished something you know that's that's worth accomplishing through you know hard work and sweat and and you know utilizing the skills that you have.

[00:39:30]They're not really mutually exclusive, but the but the you know that can do agency people are capable of in general with lots of exceptions figuring out right and wrong and the best way to do things. It's just gone from the cultural discourse. It's just gone.

[00:39:50]>> Yeah. I I find like your research perhaps like the stereotype accuracy.

[00:39:55]It's sort of what you're saying is not like the the guy in the the the proverbial guy in the bar. He's not correct about all of his biases, but you're basically saying that people aren't stupid. People are people can actually assess the world around them and they get things wrong, >> but they get a lot of things right.

[00:40:12]>> Right. Yes, that is what that's exactly right. And and also the what you said now about this this paper that the the new paper like you have the implicit bias is sort of let's trust the scientist that makes like a weird that's like a new version of a roshack test. So you have to trust the smart psychoanalyst to say that you interpreted this ink plot this way and that means we put you in this retraining camp. But what you what you also said is like the actually like the explicit bias it had a large correlation.

[00:40:45]>> Yeah. And but then people are actually they are they are themselves like rating and they are saying so you you basically what I what I find when I read you is like you're actually sort of uh I'm a little bit uh I've become a little bit tired of being told that I have so much like unconscious implicit stupidity and it's like [laughter] I have to go through like these training courses or I have to be taught by people who know much [laughter] more than me to understand the world and when I read you it feels like maybe I actually I get a lot of maybe as a [laughter] as a little human being I could actually get some things right.

[00:41:21]>> Well, that's going to make my week, you know, that that that anyone has ever gotten that message from my stuff is just >> it's like damn, >> I pulled it off. I mean, because that's true. That's absolutely underlies a lot of what I have done and written about.

[00:41:37]Not all of it, you know, because people are imperfect. I have stuff on bias and stuff also. So, but but that is an underlying theme and and uh I'm really glad that that came across to you actually.

[00:41:49]>> That's great.

[00:41:51][laughter] Let's move into uh uh you wrote you wrote uh a piece on your Substack Unsafe Science and with the headline we tried to warn you and it's it's a discussion about basically I mean Trump and his administration has been going going hard for the DEI programs and also for the universities and it's been written >> a lot in Sweden we we love to write about because uh we love to write that uh uh scientists are fleeing elite institutions to come to Sweden and I'm not sure if if it's true and I'm not sure if they're coming uh if they're going to have if the people who are actually fleeing are the the people we want [laughter] uh in that regard.

[00:42:38]Yeah. But but I mean there's a sort of a it's been like a similar discussions but to a lesser degree also here. But basically like you you're not saying it like this but you've been warning about this for a long time and you've been writing about it for a long time. Not just like the last five years or 10 years. You've been writing about it for much longer than that. And you've been do you not only been uh uh like you've not been an old man yelling at clouds.

[00:43:04]actually like you've actually done a lot of stuff and you started founded hetrodox academy together with Jonen height and uh and and uh started and you've been writing academically. I'm going to I'm going to leave the question to you now like what did you try to warn people about? Do you think it's uh the comp like the worry or the complaining now or the criticism towards Trump? Is it correct or is it misplaced or Yeah, >> it's like yeah all of those things are true. It's both correct and misplaced.

[00:43:35]What we warned and it wasn't only me. I mean that that we tried to warn you post actually goes back over 100 years. And of all things it was the AAP. So the AU is the American Association of University Professors. They are both a national organization but also a union.

[00:43:53]So a lot of college not I don't know what proportion but often professors are unionized. They're one of the big unions. Um, the AUP, like many modern unions, is now like an engine for progressive political activism, especially in academia. But it wasn't that a 100 years ago. And it it made in 1915 and then in the 1940s it had two very very good nuanced statements about academic freedom and what's necessary what academic freedom means and what's necessary to protect it and keep it. And in the 1915 statement, it explicitly warned about warned academics not to become an engine for partisan politics because if it becomes an engine for partisan politics, it invites intervention from the outside and that intervention is likely to be far more destructive than any self-discipline academics would impose on themselves to avoid partisan politics. That's a reasonable paraphrase of what they wrote.

[00:45:02]>> It's not a direct quote >> and that they, you know, a hundred years later that's where we are.

[00:45:08]>> Prophetic.

[00:45:09]>> It was it was absolutely prophetic that that academia is not only overweed and I'm most speaking about, you know, the two or 300 elite uh research scholarship producing institutions. So, I'm not talking about every little community college or even necessarily every small barely struggling to survive liberal arts college. You know, there's teaching issues there, but it's the the the ones that produce the knowledge, the ideas that get sort of distributed, diffused through the wider culture like implicit bias. You know, implicit bias didn't come out of a community college. It came out of one of these 200 elite research universities. you know the the two originators two most um widely published famous originators one one is at Harvard and the other was at University of Washington so you know these aren't coming out of Podon College so anyway those institutions are overwhelmingly populated by faculty on the left I mean it's like there's different ways to assess this like party political party membership there's political party contributions there's self-reported ideology it doesn't matter how you do it and when I say overwhelmingly depending on the survey and the criteria maybe 70 to 95% again also varies by field to some degree identify as some form of left there's another 10 or 20% in the middle maybe five to 10% at the most who would identify on the right it's probably even less than that and it's certainly less than that in social sciences and humanities which are the fields that are most likely to deal with politicized topics not that the natural sciences don't know if you climate science that's become politicized. So there's some of that in the natural sciences, but not as much. And the political skew is worse in the social sciences and humanities than it is in the natural sciences. So there's a skew, but that doesn't mean necessarily what they produce is wrong, distorted, or unjustified. But empirically, in fact, a lot of what they produce is wrong, distorted, or unjustified. We've just talked about some of it. We've talked about the stereotype inaccuracy nonsense. We've talked about the implicit bias nonsense. I you know if the fields were you know even 40% people on the left and 25% people on the right and another whatever 25% in the middle of whatever it would be 30 I I think there would have been enough people not on the left who would have reacted more skeptically to these now debunked claims that have emerged from the academic left that would have made it those those critic criticisms would have been trenchant. I mean, the few that have gotten in have been trenchant, but if there were five times as many people, there would have been at least three or four times as many of these criticisms.

[00:48:01]It would have been real. The problems would have been realized earlier. The nonsense would have been debunked earlier and the fields on these politicized topics would have proceeded more effectively. Okay. So that that's like the scientific part. But instead academia the the real world in here now the American ac academia is largely I mean it's not largely much of it majority probably not a majority but a lot of it is an engine for progressive politics. And so what we've been war you know the AUP warned this about this 100 years ago.

[00:48:36]I've been warning about it for like 15 years. Um, and you know, in in in at least two or three publications, you know, I'm just paraphrasing this, I wrote something like, you know, if Republicans eventually figure out what's going on, >> what what they are likely to do is to defund these fields, you know, and sure enough, Trump comes in and he's like defunding the the word is that the I don't know the status of this plan, but they were that they were going to completely eliminate the social and behavioral science wing of the National Science Foundation. So, I don't know the status of that. If they mean it, maybe it's done already. If it if it's done, if they're going to do it, that's it.

[00:49:20]There's no funding for social and behavioral science. Now, maybe you can wing it if you know, you hook up with a biomed researcher and you do, you know, psychology and health or something like that. There might be some funding. uh but but straight up funding for things like implicit bias, stereotypes and prejudice or romantic relationships, you know, basic political attitude work in social if they cut that. I mean, there are other potential sources of funding, but then none but none of that will be coming from NSF.

[00:49:49]>> And I mean, >> so here we are. We warn, you know, we warned it. You know, it's like, well, what did you think was going to happen?

[00:49:54]There's two kinds of political activism in scholarship. this blatant political activism where they say quite you know quite forthrightly that you know the purpose of my field my discipline this effort is to change the world. Well that's political act of so there's nothing to argue about. You said that's what you want to do. Well if people don't like the way you're changing the world they're going to they're going to do everything they can to oppose you and if they control the government that includes you know cutting the purse strings. So that's explicit and then often they don't say it explicitly. Um our originators of the IAT didn't say well we're doing this to change the world at least not right away although within 10 years they were publishing things in law journals about how the law needs to change. So it's very very clear that that was also an engine for progressive political activism. And so what did you think was going to happen?

[00:50:47]>> Did you think that you would be allowed to do this forever? You know this is how the government works. people elect representatives to do the things they want. They don't want what you're offering. Why should they fund you? Like this this seems so obvious to me. I mean, it's amazing to me that it didn't happen sooner.

[00:51:06]>> Yeah. I mean, I'm a I'm a huge fan of academic freedom and it sounds like basically like saying I don't like racism. Basically, it's a it it should be self-explanatory like to to say that, but I like the you you recently wrote about what what you brought up like adversarial collaboration and that they usually produce findings that require walking back social justice narratives and I mean we we just talked about implicit bias and but those kinds of adversarial collaborations are of course much harder if you can't find any like adversary which is an academic like so if you have 90% basically or or maybe 70% on on the left more lefting and you have 20% that's basically >> in the middle but uh maybe that means just a little bit to the left and then you have these few sort of dots of denters which with other perspectives and it's going to make the science worse. I also wanted to ask you like that's the we we we went into the I told you so thing, but also I'm a little bit uh sometimes I've seen my arguments quoted and used as an argument to like defund social sciences in Sweden or some gender science or and I'm like no I I don't want these subjects to disappear.

[00:52:33]I want them to to be forced to be more rigorous and have external eyes, but I don't want >> I am not the judge of the evidence here.

[00:52:43]I can't I'm not a scientist.

[00:52:45]>> Yeah. Yeah.

[00:52:46]>> So, but I see my arguments and like you like you you can eas >> totally my ar same argument. It's it's the same idea, right? I don't want I mean I'm a first of I haven't had a federal grant in a very long time. So, but I really don't want the fields defunded. However, I do recognize a credible threat of defunding as a potential motivator to upgrade academic standards. I am willing. So, again, I you know, to me, Trump is just very superficial. He does things for flash.

[00:53:16]He doesn't do them. He does all sorts of things, not just the academic stuff. You know, it's not coincidence that he went after the Ivy League schools because they're big and flashy and everybody knows about them. That's that's what Trump does. So I have no confidence in the constructive value of much of what he's doing except the fear of that this sort of ham-handed attack on academia in general, the social sciences in particular has some potential to put the fear of these continuing attacks, the sustained attacks on academia in ways that and and I think are motivating some reforms.

[00:54:00]There are reform efforts and noises and even policies and practices over the last two or three years that I never thought I would see. Drop in the bucket.

[00:54:12]Total drop in the bucket. You know, academia is fundamentally the same as it was 10 years ago. It's very hard to change because of tenure and people's long-term jobs and which I'm not saying I'm opposing tenure either. That's a different discussion. But nonetheless, descriptively once you get tenure, people can have the job for life unless you know unless they sexually harass somebody or do something totally unethical. So that makes it very hard to genuinely reform academia.

[00:54:39]But there are bonafide efforts underway over the last few years that I never again I never thought I would see. It remains to be seen how effective they're going to be, but I didn't think I'd see these sort of efforts. So, I am convinced that Trump, for all his dysfunctional ham-handedness, is probably he's probably having a mixed effect on academia because some of the direct things he's done are probably just outright bad, you know. So, I don't know the current status of it, but when he first when the Trump administration first expuned diversity stuff, it was things like on biodiversity, uh, you know, things that had nothing to do with DEI. And you know there might be I mean I remember reading reports on biomemed studies that might have looked at different population differences demographic differences in response to some treatment or levels of disease. So that was treated like DEI you know because you're looking at group differences and it's like some I mean I'm sort of inventing this now but life's work on lifethreatening conditions was defunded and stuff. So so that's just bad. It's just just there's nothing good about it about that. And yet [music] this threat, this heavyhanded threat, I think is also having positive effects on academia because it's motivating these efforts to reform. That's not mutually exclusive with it having these other bad effects.

[00:56:03]>> Thank you, Lee, for being part of a pleasure.

[00:56:13]for [music] [music] interest organiz group.

[00:56:45]>> [music] >> for [music] [music] glowing and rock. [music] [music] >> [music] [music]