A reader asked me for my opinion on this article which is scheduled to appear in the Annual Review of Sociology.
After reading through the article, I concluded that I don’t really have much to add on the topic. A few years ago, I argued that racism is a framework, not a theory. I analogized racism to Freudian psychiatry, Marxism, and neoclassical economics, as examples of ways of looking at the world that can explain anything:
Philosopher of science Karl Popper and others have criticized such theories as being nonscientific because they are non-refutable, but I prefer to think of them as frameworks for doing science. As such, Freudianism or Marxism or rational choice or racism are not theories that make falsifiable predictions but rather approaches to scientific inquiry. Taking some poetic license, one might make an analogy where these frameworks are operating systems, while scientific theories are programs. . . .
I’m not saying that racist theories can’t be scientifically tested and falsified. For example, a race-based model could be used to make a prediction about the comparative future economic performance of different groups, and then this prediction could be evaluated. Similarly, Freudian theories can be used to make testable, falsifiable predictions. The Popperian point is that, although they can be used to make falsifiable statements, these frameworks can retroactively explain anything and thus are unfalsifiable in that larger sense.
I have a similar take on the scientific study of structural racism. It’s a framework, and it can be a useful framework, in the same way that rational choice analysis or Marxist economics or Freudian psychiatry can be tools that help people understand the world. Frameworks are important.
The issue is that I don’t know enough about this framework to say anything useful. I don’t think rational choice is a perfect framework, but I understand it well enough to have written some papers on the topic (see here and here), but I don’t feel confident about saying anything about Marxist or Freudian frameworks (despite what some people seem to think).
So, for example I can’t really get anything out of this passage:
To effectively operationalize structural racism as a multilevel phenomenon, research must move beyond single-level analyses to examine how dynamics across macro, meso, and micro levels interact to sustain racial inequities. Multilevel approaches are critical for uncovering how structural racism operates synergistically, identifying which levels have the most deleterious effects, and understanding how these effects vary across contexts and outcomes. Such analyses provide insights into whetherstructural racism at different levels has additive or compounding effects and inform which levels of intervention might be most impactful.
They say this, “researchers should prioritize data and methods suited for multilevel analyses, such as hierarchical modeling, that can account for cross-level interactions and feedback loops,” which I kinda like because it mentions multilevel modeling, but how this would work would depend on particular examples. I’d be wary of trying to use statistical models to understand “feedback loops”: multilevel modeling would be the least of my concerns in such a setting.
The authors seem to be aware of such concerns. For example, they write:
Despite robust theoretical foundations and indirect evidence supporting the notion that structural racism functions as a system, direct empirical evidence on its mechanisms remains scarce. A significant gap persists between theoretical advancements and methodological approaches for studying structural racism as a system, with measurement and modeling lagging far behind.
Measurement is important! Also, to the extent that this framework motivated careful measurements relating to these social problems, such data can be analyzed by others using other frameworks.
” Once ideas about racial subordination are tied to policies, customs, bureaucratic rules, or laws, even individuals opposed to racial inequality may nonetheless participate in its mundane reproduction(Ray et al. 2023).” This reeks of post-modernism, complete with less than clear English. I agree that like other flavors of post-modernism, structural racism it is a framework, not a theory. I suppose it is possible to do good work within the framework, but that would be despite it, not because of it.
I don’t know why you think they are using unclear English. It seems perfectly clear.
As for the broader point, years ago when I taught AP World History, I scheduled a study session for Saturday. I did not choose Sunday because it is “normal” for that to be a day for family, church, etc. But no one showed up, because most of the kids were Chinese-American and had Chinese school on Saturdays. That is a rather trivial example, and I don’t know that it is productive to call it a form of racism, but it does demonstrate that unthinking conformity with assumptions held by the majority can sometimes have ill effects.
One could say the same thing about grading class participation highly, which could also penalize Asian American or immigrant students. Now, perhaps it is nonetheless sound policy overall, but one should at least think about how any policy might negatively affect members of nonmajority groups. The failure to do so can be harmful, and the extent to which that occurs is deserving of study.
There is an implied “there” or pause between “policies” and “customs.” The first time I read the sentence, I took policies to be part of a list:” policies, customs, bureaucratic rules, or laws….”
I don’t understand your point.
The setence does seem fairly clear—at least after a few readings. Yet there is another problem with it: it is so vague that it will almost always be true. Yes, once X is the case then Y may happen. You can see how trivially true this is if we replace X and Y:
– Once you start power posing it may increase how powerfully observers perceive you.
– When hurricanes have female names, it may increase deaths.
– Men with larger circumferance arms may be more right wing politically.
All of these are statements are literally true. But they also have no good evidence for them.
“Postmodernism” encompasses an awareness of uncertainty and ambiguity in interpretation of evidence. It may help to contrast it with Modernism with its implied certainty that it held all the answers. I think that is the main reason for the projection of less than clear English on “Postmodernism.” The use of the term “framework” in this context is seemingly an example of less that clear referents to which the term is applied.
Is a framework the same as what Kuhn called a “paradigm”? I think it might be.
FWIW, I think of frameworks/paradigms as continuous with scientific theories; they just are much harder to test and revise. Quine said they are closer to the center of the web of belief, and hence more insulated from empirical revision. But paradigms do get overthrown every once in a while (e.g. Aristotelian physics, classical pantheism, the moral acceptability of slavery, etc.).
The so-called framework is Critical Race Theory, not structural racism. Structural racism is the core concept of CRT.
CRT is not a Kuhnian paradigm. CRT did not supersede a previous paradigms in decline by virtue of better explaining empirical observations that are anomalous to the earlier paradigm. As Gelman points out, CRT is the type of “theory” that can “explain” anything and everything, so it explains nothing. To wit, CRT is tautological: structural racism is both explanans and explanandum. CRT is a negative contribution to sociology, it does not do anything other theories can not do better.
Sounds similar to this:
https://intellectualmathematics.com/blog/consequentia-mirabilis-the-dream-of-reduction-to-logic/
Dmitri:
I kind of see what you’re saying, but I think the overarching nature of these explanations of the world–these “frameworks”–make them different than scientific paradigms. You talk about paradigms being overthrown. I don’t think the framework of racism, or various frameworks of religion, will ever be overthrown, and I think there will also always be a place for neoclassical economics, or whatever is the name of the theory that promulgates the naturalness and desirability of economic and political inequality. Frameworks do go in and out of fashion–at least in the U.S., religious doctrine (as opposed to political religious affiliation) isn’t what it used to be, and a few decades ago it seemed that racism (as a framework for understanding the world, not just a set of political attitudes) was in decline, but ultimately I see these not as theories or even paradigms that can be dismissed for lack of predictive power.
OP: So, for example I can’t really get anything out of this passage:
“To effectively operationalize structural racism as a multilevel phenomenon, research must move beyond single-level analyses to examine how dynamics across macro, meso, and micro levels interact to sustain racial inequities. Multilevel approaches are critical for uncovering how structural racism operates synergistically, identifying which levels have the most deleterious effects, and understanding how these effects vary across contexts and outcomes. Such analyses provide insights into whether structural racism at different levels has additive or compounding effects and inform which levels of intervention might be most impactful.”
I think I understand what it’s trying to say, although its language is almost deliberately obfuscatory. My response would be that structural interconnections that defeat a purely one-at-a-time analytical approach can themselves usually be examined one at a time! If a training program that feeds into a job classification is discriminatory, one can empirically examine how bias is transmitted from one sphere to another. And if there is residential segregation combined with commuting costs, you could add spacial accessibility effects into it as well. To not untangle “structural racism” into concrete interactions between observable institutions or processes is lazy, IMO. In other words, to function as a useful framework — the “dig here” injunction — structural racism has to be much more than a vague placeholder.
Sorry, but I find this writing perfectly clear.
I’m not sure that’s what the authors mean with ”multilevel”. If I understand your examples correctly, they all seem to be on the same level. A toy example of a multilevel approach would be to study how X works within individuals, communities, and nations. This is an old discussion in sociology/early social science in general, where sociologists have criticised psychologists for focusing on the individual (micro) level, arguing that this is futile because so much of behaviour is determined at higher levels.
A commenter above said that a quoted sentence ”reeks of postmodernism”. There is nothing postmodern about the passages quoted here.
While I think there is reason to be critical of contemporary sociology, I think its important to build from a position of earnest curiosity. (I’m not implying that Peter Dorman isn’t). In a similar vein, I think there are lots of important critiques of economics to be made; however, us psychologists consistently fail in making them because we are not taking enough care to understand the field we are criticising. This is holding us back.
Mattias, Sociology at one time provided a valuable alternative perspective to psych. Pls see statement from Roy Baumeister (number 9)
https://www.ihatesociology.com/faculty-testimonials
I think in general if you are going to read academic language from a discipline you are not part of it will generally be more challenging. If I read out of my discipline I also fee like “I think I understand what it’s trying to say.” To me it reads pretty straightforward. If anything, saying something has micro, meso and macro elements (e.g. individual, network, society/country) is obvious and true for most social constructs.
If you take the Miasma theory of disease and tack on the grift that {insert your group of choice} is the source of the “bad air” causing all disease and therefore owe you compensation, you get a “framework” with equal standing to the one discussed in this post.
Well originally germ theory was in stark contrast to miasma theory, and only infection by close contacts was accepted. Over time the airborne and superspreader evidence has grown (and the deniers have retired and so on), so miasma is looking more like a (somewhat misinterpreted) consequence of germ theory.
The standard reproductive number calculations (and hence herd immunity) don’t account for this either. Given the insufficient data generally available, the models that do account for it need extra many parameters so you can fit anything. That is even if we can work out the math today using MCMC/etc.
The word “theory” in the English language is fraught with conflicting meanings. For most people, “theory” is contrasted negatively with data. As in, “I have a theory” which means a contention, a hunch, a feeling, etc., but not much in the way of so-called “hard data.”
On the other hand, in science the term “theory” is the highest accolade than can be bestowed–e.g.,quantum theory, electromagnetic theory, theory of evolution, heliocentrism. A theory explains what has happened in the past and successfully predicts what would happen in the future.
Lookin it up, going back to ~ 600 BC greek “theory” meant going on a trip to see something interesting. Eg, an oracle or festival, or the Olympics.
Then Plato used it to mean a mental “journey”, followed by Aristotle who used it to mean *totally useless* mental sightseeing. Where uselessness was the most exalted virtue.
https://www.jstor.org/stable/20163840
I’d define theory as a collection of premises, rules of logic, and the consequences that follow from those premises when applying the rules.
This seems off:
“Despite robust theoretical foundations and indirect evidence supporting the notion that structural racism functions as a system, direct empirical evidence on its mechanisms remains scarce. A significant gap persists between theoretical advancements and methodological approaches for studying structural racism as a system, with measurement and modeling lagging far behind.”
How can your theoretical foundations be “robust” when the empirical evidence is either indirect or scarce, and how can theoretical advances far outpace measurement and modeling? How does one know these are true advances without supporting data?
This recent paper may interest you: https://link.springer.com/article/10.1007/s42113-024-00214-8
In many scientific fields, sparseness and indirectness of empirical evidence pose fundamental challenges to theory development.
…
From examples of theoretical and empirical work in this domain, we distill five virtuous practices that scientists could aim to satisfy when evidence is sparse or indirect: (i) making assumptions explicit, (ii) making alternative theories explicit, (iii) pursuing computational and formal modelling, (iv) seeking external consistency with theories of related phenomena, and (v) triangulating across different forms and sources of evidence. Thus, rather than inhibiting theory development, sparseness or indirectness of evidence can catalyze it.
You mean like string theory?
I read the abstract and am familiar with the authors and journal. I’ll read the article later. For the time being, keep in mind that the existence of PRESENT structural racism accounting for differential outcomes is taken as an article of faith in sociology. This is why this once relevant discipline has become debased, intellectually corrupted and emphatically dishonest.
Structural racism is one of their unchallengeable core tenets. All debate is shut-off and violators have been censorted (less so today) and attacked. Below are the operative precepts of sociology. (manifestations of wokism – a term that needs to be defined in the context of a subject matter – e.g. education or policing – or explanatory theory and time-stamped – e.g. early Marxism varied from the later Marx).
Most sociologists keep their heads down, afraid to speak out. Here are some who do:
https://www.ihatesociology.com/faculty-testimonials
This article describes what happens to those who violate this taboo (see paragraphs beginning with “Sociologists Forbid Any Conclusion That Strikes Them as a ‘Social Injustice’”.
https://www.ihatesociology.com/todays-sociology
Below are sociology’s tenets – adopted from an article by Professor Anna Krylov (in bibliography section of above site)
Everything (including science and academia) is racist, sexist, hetero, colonial and about power struggle between the oppressors and oppressed.
Existing social inequalities and unequal representation are due to currently existing systemic racism and sexism.
Everything (including science and education) needs to be dismantled and rebuilt to ensure Diversity, Equity and Inclusion (DEI)
The oppressor is defined according to their position in the intersectionality power spectrum.
An oppressor is inherently unable to appreciate information presented by an oppressed victim person.
Those who are not with us are against us and must be punished
Is it that bad? Reminds me of this story:
https://www.businessinsider.com/north-korea-defector-says-columbia-university-reminded-her-kim-regime-2021-6
Interesting comment. On one hand, I believe it is an accurate portrayal of the state of the discipline. It is not my discipline, but in mine (economics) there are similar “truths” that prevail, and disagreeing with these can be career-demolishing. But – on the other hand – there is a reason why such beliefs prevail. In the case of economics, appreciation for market forces comes from disciplinary knowledge that outsiders often don’t possess. I believe sociologists’ belief in privilege and class struggle also stems from their disciplinary training. You may believe that “training” is really “indoctrination” but I think that is an uncharitable interpretation. It can also represent insights from the discipline that others may lack. For example, while sociologists appear to have an almost universal belief in privilege, you will find almost no presence of the topic in economics. Which is closer to the truth? I’m not sure.
This ambiguity shows up in discussions about diversity. Does diversity mean that all views must be equally represented – even if they don’t make sense according to a discipline’s tenets? Must Marxism be taught in all economics departments? Must the world be presented as color-blind if reality (opportunity and history) is not color-blind?
I think we can all agree that no discipline should banish those who believe differently than the mainstream. But it seems equally disturbing to demand that every discipline must equally promote all views, regardless of their merits. I can agree that power structures in academic fields are counter-productive and that dissenting views deserve to be heard, appreciated, and responded to. But the dismissal of sociology on the basis of its central focus on privilege strikes me as misguided. Privilege is too often dismissed (perhaps by everyone other than sociologists) as important or denied as reality. Anyway, that’s my take on many of the Supreme Court decisions. They want all policies to be color-blind while the reality of people’s experience is that color matters.
Dale –
On one hand, I believe it is an accurate portrayal of the state of the discipline.
I disagree. I think far more likely, it’s is a grossly exaggerated depiction of the state of the discipline. That’s not to say that it isn’t rooted in legitimate critique – but the legitimacy is buried deep beneath grievance-mongering, agenda-pushing, victim-claiming hyperbole.
Joshua
While I am inclined to agree with you (that joey overstates the reality), I think you also overstate your position. I have no direct knowledge of the practice of academic sociology, but I find it reasonable to believe that some mainstream views are granted preference over alternatives. Academia rarely is the open marketplace of ideas that we’d like, and I can well believe that views denying privilege or class struggle are met with hostility and some form of discrimination. But it is rarely so extreme as to be met with “Those who are not with us are against us and must be punished.”
Dale –
It’s not clear to me what I’ve overstated. I fully agree it’s “reasonable to believe that some mainstream views are granted preference over alternatives.” I also agree that it’s likely not infrequent that “views denying privilege or class struggle are met with hostility and some form of discrimination.”
I think there are legitimate criticisms to be made. That’s why I think joey’s rhetoric, although typical of what’s to be seen online or on popular RW media, is particularly counterproductive – because it obscures useful exchange.
joey –
Existing social inequalities and unequal representation are due to currently existing systemic racism and sexism.
You speak in absolutes.
It seems reasonable to me to conclude that to some extent, existing inequalities and unequal representation are due to existing systemic racism and sexism.
You portray the authors as believing that ALL existing inequalities and unequal representation stem solely from systemic racism and sexism.
Such authors attribute ZERO causality to a legacy of centuries of systemic racism and (perhaps millennia of?) sexism. Those authors also attribute ZERO causality to ANY other factors.
I’m willing to believe that some authors might hold such extreme views. However, my guess is that far more often, these are the beliefs of strawmen, who function more as an impediment to productive engagement than an enhancement.
Sorry – this should have read:
You portray the authors as believing that ALL existing inequalities and unequal representation stem solely from EXISTING systemic racism and sexism.
Thank you for the comments.
Yes, my website is an exaggerated depiction of the discipline. That is the nature of an angry polemic. Yes, there is also grievance-mongering and agenda-pushing, and maybe victim-claiming hyperbole (not entirely sure what that means in this context). It doesn’t make the site wrong or unimportant.
Below is a response from a distinguished sociology with whom I shared this blog post. This person is a ferocious anti-Trumpist and unwavering progressive. Many sociologists share this view (STRUCTURAL RACISM IS MOSTLY GARBAGE) but won’t speak up.
Also, my rhetoric (which I am trying to tone down) is counterproductive from the standpoint of progressives. From my point of view, the cautious suit and tie ‘you’ve gone to far’ approach is not having much impact and is not the right approach for combating the woke insanity (again my opinion). To illustrate, last week the Heterodox Academy had their meetings in Brooklyn. A participant told me it was comprised almost entirely of thoughtful liberals who are blinded (in my view) by their strong anti-Trump sentiment. No one would dare make a defence of what the administration is doing to Harvard (at least some areas like DEI and admissions). I think wokism is a nasty pernicious false ideology which is closer to a civilizational threat than the ‘pendulum going too far’. To combat it we need our own Jerry Rubin (https://www.youtube.com/watch?v=gdE04yrUIg8&ab_channel=DaveWendt)
Here is the aforementioned comment
“I’ll try to read that Annual Review article when I get a chance. The whole issue of structural racism is mostly garbage. There are a couple of elements of the concept that make sense and are worthwhile, but the way the term gets thrown around by most sociologists and others, it tends to have little meaning or is open to all kinds of alternative interpretations.
I agree with Gelman that Marxism, psychoanalysis, and rational choice theory are frameworks rather than theories. I’m not sure that structural racism is even a framework rather than a concept. As an analytical approach it’s related to what Wimmer calls “race-centrism.” Much of it, at least as practiced by Victor Ray and others, is a bastardized form of Marxism in which instead of there being opposing classes with distinct interests, there are opposing races with distinct interests. That’s pretty absurd in the current context. Right or wrong, Marx had a structural basis for his positing of inherent conflict between the dominant and subordinate class- control over the means of production and the need to extract surplus value from workers. There is no analogous mechanism of intrinsic conflict between races, nor do white Americans constitute anything close to a group with a collective shared identity and shared interests. The white population is far too diffuse for that.
The closest analysis to what these people are claiming is the neo-Weberian approach used by the late British sociologist Frank Parkin, who talks about race/ethnicity/gender/etc. as mechanisms of exclusion, where members of one group have access to goodies and then identify an arbitrary signifier (race, gender, nationality, language, etc.) to use as a basis for excluding others so that the group can maintain its access to the goodies. See his brilliant 1979 book, Marxism and Class Theory: A Bourgeois Critique.”
Reply ↓
Joey,
By “victim-claiming,” I mean the anti-woke habit of claiming censorship on issues they freely discuss, portraying themselves as silenced victims.
Your rhetoric, though you’re moderating it, hinders meaningful engagement. It fuels partisan divides, leaving both sides feeling vindicated or victimized, without fostering constructive dialogue.We live in different worlds. Angry polemics dominate, and progress is slow—though not absent.
Your Heterodox Academy anecdote, an N=1 observation used to confirm your biases through blanket characterizations, feels selective. Had someone said, “There’s a problem, but we must avoid overgeneralizing,” I doubt you’d highlight it.
You view “wokism” as a pernicious civilizational threat. I see it as a pendulum swing within a broader trend where more people have unprecedented agency to advocate for themselves. I dislike the term—it’s often a group-based identity flag—but I accept it can be used in good faith to mark where the pendulum has swung too far. From a bird’s-eye view, this swing is part of a larger phenomenon: on a societal scale, far more people can now shape their lives’ quality. For every misapplied “racist” or “homophobe” label, countless others gain a voice, unprecedented in history.
Your interlocutor’s comment seems logically inconsistent. It acknowledges “worthwhile” elements in structural racism but dismisses it as meaningless or overly interpretive—how does that square? I agree structural racism may be better described as a concept than a framework, and its ties to distorted Marxism or “race-centrism” deserve scrutiny. But claiming the white population is too diffuse to apply the concept ignores historical realities like Jim Crow, which demand rigorous examination. Dismissing these ideas outright feels unscientific and stifles the structured debate they claim to support. It seems an odd way to respond to authors who seem interested in a more scientific approach.
“Structural Racism” is nothing but half-witted quackery. If you want to see racism just step out your office doors at Columbia, Harvard, MIT, and the rest of the Ivy League dumpster fires.
The reason it’s hard to measure “structural racism” is because it doesn’t exist, just like Jessica’s “bias” in high school grades and test scores. Use your eyeballs and look at the real world. All you’re doing is manufacturing high noise BS to then seeking an explanation within it rather than accepting what’s blatantly obvious and staring you straight in the face. Get a clue.
Anon:
I’ve said this before with other comments of this sort: If there’s a specific thing in our posts that you disagree with, let me know. The comment section of this blog is a perfect place to bring up specific points. General statements such as “Use your eyeballs and look at the real world. All you’re doing is manufacturing high noise BS . . . get a clue” add nothing useful to the conversation.
If you have nothing specific that you disagree with in these posts, you might consider that you’re picking a fight with the wrong people. If you do have specific disagreements, you’re wasting an opportunity by not bringing them up here.
IMO the real problem in social sciences is the purported “scientsts” need to pretend their doing something useful and important and world saving to advance their own careers. L-O-S-E-R-S
Anon:
That’s a problem with some social scientists, for sure! Although I wouldn’t call them “losers,” as some of them get adoring publicity and lots of money from public speaking. Like it or not, they’re winners! I actually think the winners are more of a problem than the losers. The losers are kind of pathetic and don’t have much of an audience. The winners can affect policy.
+1. Sigh.
Some social scientists are so afraid to unjustly discriminate,
that they have stopped thinking why and how things might originate
In doing so, they might largely unjustly sophisticate
and might block those that justly originate
They think they know because of a diploma or certificate,
and they think they have the knowledge, wisdom, and right to administrate
They don’t think about what they might facilitate,
or what they intoxicate, or that they might miscalculate
Perhaps it’s good that some others also communicate,
and once in a while invigorate, and disseminate
For the message is clear and can possibly illustrate,
which may sometimes be all that one needs to state
https://alexanderriley.substack.com/p/the-immoral-lie-of-metaracism