Racism is a framework, not a theory

Awhile ago we had a discussion about racism, in the context of a review of a recent book by science reporter Nicholas Wade that attributed all sorts of social changes and differences between societies to genetics. There is no point in repeating all this, but I did want to bring up here an issue that is relevant to political science, which is how do we think about racism, not as a set of policies or even as a set of political attitudes, but as a way of understanding the world.

Wikipedia refers to “scientific racism” as “the use of scientific techniques and hypotheses to support or justify the belief in racism, racial inferiority, or racial superiority, or alternatively the practice of classifying individuals of different phenotypes into discrete races,” and this image from Wikipedia captures the mix of scientific reasoning, racial classifications, and value judgment that is characteristic of that way of thinking.

As with Freudian psychiatry, Marxism and neoclassical economics, the logic of racism can explain anything; it is unfalsifiable. In his book, Wade looked at economic inequality today and ascribed it to race. The study of differences in societies is interesting and I think Wade finds it interesting too (in his book, he has some conflicting lines, at some points talking about how culture is all-important and at other places disparaging those social scientists who are interested in culture). Cramming everything (including interest rates and, in another book, ping pong) into a racial framework is not so convincing to me, for the reasons I stated in my review of his book.

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. That’s why I wrote that I can’t say that Wade is wrong, just that I don’t find his stories convincing.

Just to be clear: 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.

This can be seen in many popular works of racism including the book by Wade. His model is pretty sophisticated: genes affect culture which affects behavior. But it’s one of those can-explain-any-possible-data sorts of theories. If a group does poorly, it’s either bad genes or bad governance that’s unrelated to genes. If a group succeeds, it could be the good genes revealing themselves, or it could be that the genes themselves changed via adaptation. And if a society is poorly governed, this can have no effect on genes, or it can adapt people to behave in an uncivilized way (as in the Middle East and south Asia) or at can adapt people to behave in a civilized way (as in China). For example, Wade writes:

The Malay, Thai, or Indonesian populations who have prosperous Chinese populations in their midst might envy the Chinese success but are strangely unable to copy it. … If Chinese business success were purely cultural, everyone should find it easy to adopt the same methods. This is not the case because social behavior, of Chinese and others, is genetically shaped.

Wade offers no particular clue on what happened to make Thais and Malays such losers, but he makes it clear that he thinks their lack of economic success demonstrates that it’s their genes that aren’t up to a world-class challenge.

My feeling about Wade’s genetic explanations for economic outcomes is similar to my feeling about other all-encompassing supertheories: I respect the effort to push such theories as far as they can go, but I find them generally less convincing as they move farther from their home base. Similarly with economists’ models: they can make a lot of sense for prices in a fluid market, they can work OK to model negotiation, they seem like a joke when they start trying to model addiction, suicide, etc.

All-encompassing frameworks are different from scientific theories. Both are valuable — frameworks motivate theories and help us interpret scientific results — but I also think it’s important to be clear on the distinction.

P.S. I wrote the above note five years ago but it is now behind a paywall so that’s why I’m posting it again now.

114 thoughts on “Racism is a framework, not a theory

  1. > these frameworks can retroactively explain anything and thus are unfalsifiable in that larger sense.

    Is there an example of a “framework” which does not have that problem? If not, then your complaint is really about the use of frameworks in general. And that may be a reasonable complaint! But it is awfully extreme . . .

    • Problems arise when the two are conflated. Even without a realist perspective, theory holds some truth value (or should). A framework need not. It can still be useful, but if the distinction is not recognized science and society as a whole can run into trouble.

  2. “Taking some poetic license, one might make an analogy where these frameworks are operating systems, while scientific theories are programs.“ – absolutely terrible.

    • Maybe a better analogy would be frameworks as programming languages and theories as programs?

      The framework provides the tools with which to express a theory which we can then “run” to make predictions.

      To the extent that the expressive power of a framework was limited to only producing theories that make wrong predictions, this would justify falsifying that framework. But I think Andrew’s point is that many frameworks are “Turing complete” and thus capable of, perhaps in very convoluted ways, expressing theories that could yield any possible prediction.

      Still, perhaps something like alchemy would be a falsifiable framework?

  3. The timing of this post is highly motivating. Daily, my first order of business is reading your posts. Today, my second was to incorporate comments and finish my paper on the importance of making the distinction between frameworks and theories!!!

  4. I am also confused by the distinction you draw between “neoclassical economics” as a framework and its component theories. Supply-and-demand is, perhaps, the ur-theory of economics. Or is that an unfalsiable framework?

    Instead of casting aspersions, it would be most productive to be specific, to think in terms of predictions. Neoclassical economists will tell you that the result of rent control will be less rental housing of worst quality being available in 10 years. That is hardly unfalsifiable! Similarly, people like Nicholas Wade will make specific predictions about, say, performance in Olympic events and in Columbia graduation honors. If these folks can predict the future — better than the folks who use other frameworks — then surely those frameworks have some value . . .

    • D,

      I don’t recall Nicolas Wade making any predictions in his book, but it was awhile ago that I read it so maybe I’m forgetting something. I guess you could say he’s making an implicit prediction about the economic success of Malaysia, for example. I don’t recall him saying anything about the Olympics, but, again, maybe that’s in his book and I forgot it. In any case, the predictions would have to be more specific before we could say assess what value they have beyond some default predictions based on current data.

      Regarding economic predictions: “neoclassical economics” is a general term and it can include some specific models of how prices and supply and demand will interact in controlled, measurable settings, and also unfalsifiable frameworks such as utility theory which has led leaders in the field to come up with statements such as “most (if not all!) deaths are to some extent ‘suicides’ in the sense that they could have been postponed if more resources had been invested in prolonging life.” You could call that statement falsifiable—indeed, clearly false!—except that its proponents appear to continue to hold this belief in the face of all evidence to the contrary.

      Regarding your general point (“it would be most productive to be specific, to think in terms of predictions”): Sure, predictions are great. The point of my above post was slightly different, though, in that I want to talk about the role of frameworks even when they’re not used to make predictions, or even when their predictions go wrong. These frameworks can still be useful—indeed, maybe they’re necessary, at some level, for us to try to make sense of the world.

      • > I want to talk about the role of frameworks even when they’re not used to make predictions, or even when their predictions go wrong.

        Understood. I recommend that you start by being polite. For example, it is rude (and, worse, unproductive) to refer to proponents of Roe v. Wade as “anti-life,” even though, from the point of view of their opponents, that is exactly what they are. Better is to refer to them by the name they choose from themselves, which is “pro-life.”

        Similarly, it is rude to refer to Wade (and others) as “racist” when they use/prefer different terminology, most commonly as proponents of human biodiversity or, in short, HBD’ers.

        I think that HBD would be an interesting case study of the usefulness (or not) of frameworks, but you need to begin with basic politeness toward its proponents. Otherwise, why would they engage?

        • Agreed. Some of this post just seems like gratuitous snark. Lumping in rational choice with racism? Similarly with the discussion of Becker: ” Similarly with economists’ models: they can make a lot of sense for prices in a fluid market, they can work OK to model negotiation, they seem like a joke when they start trying to model addiction, suicide, etc.” Have you actually read Becker’s serious work, or are you comfortable writing him off as a joke based on one off-the-cuff (quoted!) remark which was almost certainly tongue-in-cheek? He’s done pioneering work on applying the rational choice framework to problems that weren’t traditionally considered the realm of economics (discrimination, crime, etc) — if you want to seriously grapple with his ideas, that might be a better place to start.

        • Anon:

          I don’t think of any of the above post as snark. I mean it all sincerely. And, yes, we’ve had lots of discussion on the blog regarding that suicide remark of Becker’s. You now seem to be invoking a new rule that certain statements by prominent people should not be criticized because they are actually “tongue-in-cheek.” Unfortunately, that suicide comment, whether or not it was tongue-in-cheek, was taken seriously by the influential authors of Freakonomics (follow the links above). So, yeah, if a major media outlet is allowed to take that remark seriously in a positive way, then, yes, I’m allowed to seriously criticize it. Becker may well have done lots of wonderful and important work; that doesn’t mean that this particular idea of his made any sense. But, just to be clear: in criticizing that particular statement from Becker, I’m not at all trying to disparage all or most or even much of his entire body of work, nor do I consider myself well-enough read in this area to do so.

        • > Similarly, it is rude to refer to Wade (and others) as “racist” when they use/prefer different terminology, most commonly as proponents of human biodiversity or, in short, HBD’ers

          So people who literally make the claim that certain ethnicities are genetically inferior to others cannot be called racists so long as they would prefer to not be called racists. We can only make statements about someone that they would personally agree with, and anything else is impolite? This is political correctness gone mad!

        • I’d say “Racist” is someone who thinks that everyone belonging to the group “X” is inferior just because he belongs to “X”.

          If statements that average observable differences between groups are attributable partially by genetic differences (without making statements about inferiority) are racist, then a lot of research about differences in height, cancer rates, diseases rates are racist too.

        • Szopen:

          1. I’ll use the Merriam-Webster definition of racism that I quoted in my Slate article: “a belief that race is the primary determinant of human traits and capacities and that racial differences produce an inherent superiority of a particular race.”

          So, not “everyone…” but rather “primary determinant.” And, not “observable differences” but “inherent superiority,” which I take to be an average property, not applying to every individual in each group.

          It’s uncontroversial to say that people of European or African descent have different average risks of skin cancer from sun exposure: these are observable differences but this would not be a statement of “superiority” as we usually think of the term.

          2. Your more general point is that racist statements can be true. The point of my post above is that I don’t think it makes sense to think of racism as being true or false; rather, racism is a general framework used by Nicholas Wade and others to make sense of the world, analogously to the way that others might use Freudianism or Marxism. Such frameworks can be valuable as structuring devices, and you can look at ways in which these ways of thinking help explain various phenomena, without being concerned about the truth of the framework as a whole.

        • “The point of my post above is that I don’t think it makes sense to think of racism as being true or false; rather, racism is a general framework used by Nicholas Wade and others to make sense of the world, analogously to the way that others might use Freudianism or Marxism. Such frameworks can be valuable as structuring devices, and you can look at ways in which these ways of thinking help explain various phenomena, without being concerned about the truth of the framework as a whole.”

          But to me, the real question is whether or not “explanations” that are “structured” by a framework reflect external reality, or whether the “structure” influences the results. For example, I would guess that an “explanation” that is “structured” by racism, an “explanation” that its “structured” by Freudianism, and an “explanation” that is structured by Marxism would likely have differences that plausibly would be the effect of the structuring frameworks.

        • D:

          Following the links, here’s what I wrote in my book review:

          Wade does not characterize himself as a racist, writing, “no one has the right or reason to assert superiority over a person of a different race.” But I characterize his book as racist based on the dictionary definition: per Merriam-Webster, “a belief that race is the primary determinant of human traits and capacities and that racial differences produce an inherent superiority of a particular race.” Wade’s repeated comments about creativity, intelligence, tribalism, and so forth seem to me to represent views of superiority and inferiority.

          I agree that other terms could be used, but I think “racist” is accurate. As discussed in the above post, I think of racism as one way of making sense of the world, comparable in some ways to psychoanalysis, Marxism, etc.

        • +1 I think it is dishonest for those who repeatedly want to argue that variation among humans should be primarily explained by genetics and that those genetics track common notions of race to object to the term “racist.” That is just what racism is.

        • +1 I think it is dishonest for those who repeatedly want to argue that abortion should be legal to object to the term “anti-life.” That is just what being pro-abortion is.

        • D:

          That’s a bit different. “Racist” is an existing word with existing meanings that is used in various contexts. “Anti-life” is a political slogan and I don’t think it’s used in any other way.

  5. Regarding Chinese immigrant communities in Southeast Asia, the most straightforward explanation for their success and the inability of their hosts to copy it are the two facts that (a) Chinese IQs are significantly higher than the Southeast Asian average, and (b) IQ is mostly genetic.

    As a descendant of some of these very same Chinese in Southeast Asia, Amy Chua has some excellent writing on the topic: https://en.wikipedia.org/wiki/World_on_Fire_(book)

    She relates the Chinese-in-Southeast-Asia situation to similar ones in other places and times such as Indians in East Africa and pre-WWII Jews in Europe. Given rising ethnic tension in the world currently, the book is extremely valuable in understanding the underlying dynamics and how future racial violence may be avoided.

    • pwyll said,

      “Regarding Chinese immigrant communities in Southeast Asia, the most straightforward explanation for their success and the inability of their hosts to copy it are the two facts that (a) Chinese IQs are significantly higher than the Southeast Asian average, and (b) IQ is mostly genetic.”

      There are a couple of confounding factors that this statement does not take into account:
      First, immigrant communities bring their cultural backgrounds as well as their IQ’s with them when they immigrate. Second, immigrant communities are not necessarily representative of the countries from which they came — because the choice to immigrate usually depends on individual traits of personality and intelligence.

      • Yes, how about selection bias. The Chinese diaspora in South East China did just get randomly put in Indonesia or Malaysia. They got there, as anyone interested in the facts will know, because they were merchants. They were merchants who left China to do trade in places like Indonesia, Singapore, Malaysia. My Chinese wife’s grandmother born to a family of Chinese merchants who lived in Indonesia immigrated back to China to be sold as a wife. Guess who stayed in those countries. Merchants who were successful. So, you have a selection bias for migrants who are merchants, educated, and who have successful businesses, and what happens a couple of generations later. Their descendants are on average more successful than the native population that include a large segment who descended from peasants. Astonishing! The only possible explanation must be genetics. Really?

        • As you state, selection bias could be some mix of selection for more-mercantile-sucess genes, or more-mercantile-sucess culture.

          If it’s more-mercantile genes, then that’s just a more-extreme version of my initial statement regarding average Chinese IQ being higher than average Southeast Asian IQ. So, I agree?

          If it’s more-mercantile culture, that would be a reason for *initial* Chinese diasporan success, but if it persists (which it apparently has) then we’re back to the quotation from Wade in the post: “If Chinese business success were purely cultural, everyone should find it easy to adopt the same methods.”

  6. This post seems to follow “Popper0”. Every theory is unfalsifiable in practice because you can only falsify the conjunction of the theory and a bunch of auxiliary assumptions. Ie !(T & A) = !T | !A. So the theory can always be saved.

    So the lack of falsifiability isn’t really a valid critique.

    Instead the problem you are describing is that these “frameworks” fail to make any surprising predictions that turn out to be accurate, instead the theories lag the observations. They are always following behind and explaining stuff retroactively. That’s Lakatos’ degenerating research programme.

    For another perfect modern example look at GR + dark matter. Galaxy brighter than predicted -> dark matter, galaxy dimmer than predicted -> dark matter. Galaxy exactly as predicted -> hey that proves dark matter exists too.

    • It should be about better (and hence sometimes surprising) prediction than existing theories. Einstein’s theory of general relativity comes to mind in terms of surprising predictions it made that turned out to be true.

      • It’s just applying Bayes’ rule. Surprising just means the prediction is unlikely given other theories are correct. If dozens of theories all accurately predict the same thing with equal precision then that doesn’t help much to distinguish between them.

        Of course, a surprising prediction that is wrong will lead to a low likelihood, so you need it to be both accurate and surprising.

  7. Wagner had a ‘theory’ which explained that Jews copy or imitate the forms of the societies in which they burrow. So for example Mendelssohn, despite being Protestant by choice, wasn’t capable of actual Germanic music because he was, at best, a Jew pretending to be a German. I wonder what he’d have thought of gay, Jewish, New York composer Andrew Copland’s ‘western’ music, like the wonderful part that became the ‘Beef, it’s what’s for dinner’ ad. Now, one can argue about the merits of Mendelssohn or Copland versus others, but you really can’t argue that they’re at least pretty darned good. So imitation works. Just like it did when decoding with Enigma. in fact, all math is imitation: the representation of physical realities using numbers and symbols that indicate values like probable position, momentum, etc. And statistics is an often crappy imitation, both because it is so often inexact and because it’s often so poorly understood and done.

    Let’s take as given that imitation works. Because all of science depends on imitation. Why then would it be ‘genetic’ that Malays don’t act enough like overseas Chinese (per the person quoted, not by me)? I need to use something more familiar. I believe the GDP per head for Israel is about the same as the UK’s, so Western. This is often used to say that Palestine has been devastated, because the GDP for Palestine (West Bank & Gaza) is something like $2700-2800 per head – or about $3200-3300 per head if aid per head is included. Huge obvious difference. Except the GDP per head in Egypt is about the same, actually less when you add in the aid. And the GDP for Jordan is a little over $4k per head. And if you look at the West Bank and Gaza separately, you see that if Gaza weren’t ridiculously low – under $1k per head without aid – then Palestine would be the same as Jordan.

    Egypt is the 2nd largest grain importer in the world. Their water is becoming saline. Israel is the learner in water technology. They’re now planning to pump desalinated water into the Sea of Galilee to keep it reasonably full – there’s been a terrible drought and there will be more – which I gather is the first ‘refilling’ of a fresh water body from a saline one. One could easily apply a racial framework and say things like the Arabs are genetically incapable. But the history of the Jews demonstrates the untruth of that because Jews go into places like Germany or the UK or Israel or the US and prosper BUT other groups don’t adopt their methods. Instead, other groups blame Jews for their success, as if the world would be better with less success. For a variety of reasons, other groups don’t emulate Jews. Or rather, they attribute a ton of bad actions to Jews which they then emulate, but that’s a different issue.

    Inside Israel, Arab Israelis – who are now arguably the most successful Arab group in the world (when you don’t count oil money as achievement) – have to deal with issues of Arab identity versus Jewish, that fitting in with Jews somehow means becoming Jewish. When you step across the Israeli border, you have people who associate ‘becoming Jewish’ with evil even though that ‘evil’ is historically ‘successful’. There is in human beings a fear that if you think like someone then you lose yourself. I’ve seen this a lot over the years with male reactions to gay men: that if you allow the concept that a male finds you attractive as a male, if you open that door a crack, then suddenly you are no longer a ‘man’. I assume the ’that’s gay’ taunting still exists.

    I want to switch to math but to complete the point: acceptance of another means revelation of that other’s existence within yourself. You have to be able to imagine the other to accept the other and that imagination is the construction of a mental model in which that other’s existence is acceptable. As I just noted, Arabs have a problem with very idea that other groups, not only but certainly Jews, are in any way palatable, so they block any mental model that constructs an image.*

    Back to Malays: I assume they have strong aversions to thinking they’re Chinese. You might be able to map out how that restricts their mental model and thus the choices available to them, but that would be a lot of work. The concept that this connects to a genetic construct of ‘race’ is not only unhelpful but doesn’t make sense because each culture sees its choices within its mental model. I noted that the West Bank’s GDP per head is similar to Jordan’s in a region where low GDP per head is normal – even among states with oil like Algeria – and that GDP per head isn’t even a rational figure for much of the region (e.g., Syria, Libya). In that context, it’s rational for the Palestinians to continue on the course they’re on, and it’s particularly rational to continue to demand aid. In the context of the region, that Israel succeeds only proves it’s a Western implant, because success is considered an implant.

    As an aside, the recent BBC commissioned survey of Arab attitudes – which was misinterpreted in general – shows that 20-50% of the populations of each country are considering leaving. That’s a level of dissatisfaction, right? Why now? Because the nature of success has changed and people around the world are now visibly connected so nearly everyone can see how people live where there is actual opportunity and actual jobs. (I read a piece in Tablet about a gay activist from Philly who moved to NYC because he saw on Susskind that gay men lived there. In the 1960’s, you were genuinely isolated in your little worlds.) So now people see that some are successful. In typical human fashion, this attracts a lot of blame: you’re the reason we aren’t successful! It’s never our choices that have been bad; it’s all things that have been done to us. This is common in blaming the US for every thing wrong with Central and Southern America, and it reaches the insane level when you find people blaming Jews for pretty much everything from droughts to the existence of gay people. I wish I were making that up, but it’s common in the Arab world to read that Jews plant homosexuality, that they make daughters ‘sexual’, though there aren’t any Jews in that country. You can explain this as terminal stupidity rooted in genes or, better, as intense rejection of a mental model in which it’s your fault that things aren’t going the way you insist in your mental model that they should go.

    As to the math, the issue becomes one of the most troubling of all: how do you know you’re right when you don’t know you’re right? How do you reach a level of understanding without going through the process of reaching that level of understanding? How do you solve a problem when you need to find a mental model that you can’t find until you find that mental model? (That gets toward P ≠ NP.) How do you admit into your mental model what you need to ‘succeed’ when you can’t admit into your mental model what you need to succeed … until you do? I doubt that Saul of Tarsus thought about math when he was enlightened on the road to Damascus but the same question is embedded within: you only get it when you get it. These are all versions of fundamental counting problems like ‘how do you get to understanding when getting to understanding means letting go of a prior understanding?’ (Again, gets to P ≠ NP, Halting, etc., because, as the Arab attitudes about Jews show, you can disappear into an infinity of absurd condemnations without recognizing that they’re absurd, meaning you can’t solve the problem because you can’t realize the algorithm ain’t working because you can’t change the algorithm without changing yourself. I find the relation between behavior and conceptions of ‘time’ particularly interesting.)

    A lot of what you write about is how people impose endpoints on their work, even implicitly, and how that affects their analysis and findings. When I was kid, my expectation was that it was the quality of the tools: we didn’t have very good data, were limited in ability to manipulate figures, etc., so that meant lots of bad guesses and bad analysis passing itself off as good because humans talk a lot. (Twitter remains the best named thing ever: humans tweet like birds, with most of it just noisy chatter.) As the tools have improved, it’s becoming more obvious that it’s the mental models that are lacking. This is the same in the world at large. In the 1970’s, N.Korea was as rich as S.Korea and the non-oil Arabs were richer than the Asians – and India suffered famines, etc. Now, Lebanon leads the non-oil Arab world with a GDP per head of under $10k. It was sensible to believe that ‘this works’ because, bluntly, it kind of did if you had relatively low expectations to begin with. Now? The gap between the rich and poor is not just what you can take by force or extract from the ground but what you can do with your mind. How do you adjust your mind to fit those needs when you can’t accept into your mind that you need to change your mind?

    The world regularly finds ways to reinforce refusing to give up mental models. You point out some absurdities. The social sciences are overrun with people who find creative ways to justify absurd thoughts. The entire concept of ‘pinkwashing’, for example, is that Jews are so manipulatively deceitful that they pretend to care about LGBTQ rights just to cover up their crimes. That you can make a bluntly anti-Semitic argument and publish it in academic journals is not surprising because that’s a repetition of the limitations of human awareness. I could go on about how deeply stupid this is: it’s not like Israel has no LGBTQ people or that they bring some in for a parade, so the argument is that Jews not only tolerate but actually are LGBTQ to cover up their crimes, which is a monumental level of self-deception. You must know that Jews trying to not kill people are actually trying to show they have the power to maim them, so not killing people is actually worse than killing people if Jews are doing it. A person making that argument has a deeply entrenched mental model.

    To get back to the ‘genetic’ and ‘race’ model, what I find intellectually offensive is the concept that mental models directly become immutable, and yet the history of the inability of other groups to accept other groups is demonstrable, and this occurs within ‘successful’ and ‘unsuccessful’ societies. This goes beyond tolerance. Like the Tom Lehrer sense of National Brotherhood Week: you can tolerate him if you try! To reverse Wagner’s 5th rate thinking, it’s not imitation but acceptance: that Jews do well because they accept within themselves crucial elements of a culture while retaining their Jewish core. I grew up in Detroit in a largely Christian, generally Catholic neighborhood, went to school originally established for choir boys – not kidding – and then Yale. It’s not that I ‘became’ or ‘imitated’ Christian or WASP culture but that I accepted that it’s part of my identity, the same as being male or Jewish. All identity is multilevel – I prefer multi-dimensional – so the conception that Malays can’t imitate Chinese because of genes is fucking stupid.

    *This is why the new anti-Semitism is so awful: the construction of the acceptable Jew is now ‘the Jew who agrees with me and not with other Jews’, who is willing to turn on other Jews in order to be acceptable in my mental model. That this is so prominent among ‘intellectuals’ says terrible things about the ability of the human race to think.

    • 1. I think you mean “Aaron Copland”, not “Andrew Copland”

      2. I grew up in Detroit, too. A couple of relevant memories regarding interactions of religions in Detroit when I was young:
      My grandfather was a protestant minister (first Methodist, then Congregational). I recall him saying, “I may not have a drop of Jewish blood in me, but my heritage is largely Jewish”. (He had a Hebrew scholarship when he was in theological seminary, where one learned to look up the original language version of the scripture on which one’s sermon was based.)
      Religious tensions in Detroit tended to be protestant/catholic more than christian/jewish. For example, protestant mothers considered it unthinkable for their children to date catholics, since the catholics would try to convert the protestant, and worse yet, your grandchildren might be raised catholic! By contrast, protestant/jewish dating was not such a big deal — since it was assumed that whichever mother found out about it first would contact the other mother, and they would cordially agree that each would talk to her child to persuade them that they should date only within their own religion.

      • In Toronto, my mother told her father when she was getting married (in 1950) that she was going roller skating.

        Her family was protestant and my father’s was catholic. Members of both sides of the family would not have anything to do with them for about five years.

        The upside was my parents understood being discriminated against in some depth. My mother’s funeral reception looked like a meeting of the United Nations with most there to pay respects for assistance in dealing with being taken advantage of.

        This sort of thing (protestant/catholic being a “mixed marriage”) is easy to forget about.

        • +1

          To give an interracial example in my generation: One of my second cousins, who lived in Hawaii, fell in love with a woman of Japanese ancestry, and they became engaged. Her father forbade the marriage, and locked her in her room on the date for her wedding. My cousin and his friends got a ladder and “rescued” her to go to the wedding. Her father than forbade her mother from talking to or seeing the daughter. However, the daughter would phone her mother at times when she expected her father to not to be home. She told her mother that she would not ask her to talk, but that she (the daughter) would just give her news, so the mother could obey her husband but still learn about her daughter’s life and later her grandchild’s birth and development. Finally the father died, and the mother decided that she no longer was obliged to obey his command not to talk to or see her daughter, so she could then get to know her grandchild.

    • Thanks. Very interesting write up. The SRY switch I found especially interesting.

      So counting how much genetic difference there is across individuals can be very misleading. The phenotype is not like a soup where a small change in the amount of one ingredient (like, say, salt) makes very little difference to the final product. Rather, it is like a car where slight differences (is the ignition on or off?) can make very large differences.

      I had vaguely suspected that genetic mechanics had to be much more complicated than the soup analogy, or the analogy to a blueprint where every detail is specified precisely. Those mechanics are much too weak and inefficient. I suspected that adding some structure would make the genome much more powerful and efficient. Your discussion of the mechanics of the genetic program (the green squares) was, therefore, very interesting.

      Is this viewpoint widespread?

    • I found this very interesting:

      “Discerning population structure from genetic variation works best, however, when analyses are restricted to variation that is not under selection, the >90% of the genome that is “junk,” i.e., has no influence on the phenotype (non-functional DNA). The genetic basis of “Race,” in this case, cannot have anything to do with racial traits (and there are critiques of the genetic clustering algorithms, e.g., Lawson et al. 2018).”

      So the measure of genetic distance between two populations is based almost completely on non-coding DNA. The distance is really a measure of random mutations accumulated in the part of genome that has no effect on phenotype. This seems to me a very important point. Am I right that it is then possible for two separated populations to drift far apart (measured by genotype) from each other even without any adaptive changes in phenotype?

      I am not a biologist, but I am surprised that I haven’t seen this point made until now.

      • “Am I right that it is then possible for two separated populations to drift far apart (measured by genotype) from each other even without any adaptive changes in phenotype?”

        In principle, yes. I wish I’d spent more time discussing this point in the post.

        Quibble about terminology: coding DNA: codes for protein. Non-coding DNA: often termed ‘regulatory DNA’; doesn’t code for protein but has huge effect on the phenotype. Most of the differences between us and chimps are in non-coding, regulatory DNA. Junk DNA: non-functional. No effect on phenotype.

        Another quibble: most mutations are maladaptive. Thus, a lot of the between-population genetic variation in functional DNA is maladaptive, not adaptive (i.e., differences in disease-related alleles).

        • Thanks. I should have recalled from a course in phylogenetic inference I took a couple of years ago that the distance calculation must be largely based on junk DNA, as the mutations there can accrue at a steady rate without being weeded out by selection.

        • Well, the very first sentence stopped me in my tracks:

          Ten years on from the finishing of the human reference genome sequence

          The human genome was never complete, they just “did their best”. This got some press coverage a few years ago, and now apparently NIH has since removed this info from their FAQ:

          FAQs from the National Institutes of Health refer to the sequence’s “essential completion,” and to the question, “Is the human genome completely sequenced?” they answer, “Yes,” with the caveat — that it’s “as complete as it can be” given available technology.

          https://www.statnews.com/2017/06/20/human-genome-not-fully-sequenced/

          AFAIK no one even really knows how much remains unsequenced since those regions have a high concentration of repetitive elements that are impossible to align correctly. Essentially, the chromosomes from a bunch of cells are mixed together, then they get cut up into smaller fragments to be sequenced. Finally, they look for overlapping ends of the fragments with the same sequence to “stitch” the full chromosome back together. If the same sequence is repeating all over the place how do you distinguish between:

          1) multiple copies of the same region of the chromosome
          2) the same sequence from multiple regions of the chromosome

          Then there is the issue that the “human genome” is an “average genome” that is probably not found in any actual human cell. It is also unlikely that more than a few cells in the same person even have the same sequence.

          I’ll take a closer look at that paper later but if they are starting from the incorrect premise that they are working with the complete genome sequence it seems unlikely to be accurate.

        • Here’s another paper using a different method that reaches a similar conclusion that less than 10% of the genome is functional:

          https://www.nature.com/articles/ng.3196

          And yes, there is sequence that is not in the reference genome, e.g., 300 Mb in this study, which is about 10% of the 3.1Gb in the ref:

          https://www.nature.com/articles/s41588-018-0273-y

          But that’s a bit of a non-sequitur. If 90% or so of the reference genome is non-functional, than non-functional DNA is a thing.

        • But that’s a bit of a non-sequitur. If 90% or so of the reference genome is non-functional, than non-functional DNA is a thing.

          That opening line indicates they are unfamiliar with the data they are working with. It is just a heuristic. I’m really surprised people are still thinking in terms of “junk DNA” so will take a closer look.

        • “Interesting, this does not seem at all paradoxical to me… I would expect the opposite of whoever is calling that a paradox: Orders of magnitude more code to accomplish a similar task = less advanced/refined/stable code.”

          That’s one of the hypotheses. I’m not an expert on this stuff, but my understanding is that there is quite a bit of evidence for the alternative selfish-element hypothesis:

          “You can forget the polarizing term [junk DNA] so long as you remember the data it stands for: astonishing genome size variation, mutational load, a small fraction of conserved DNA, and the large fraction of eukaryotic genomes that is composed of neutrally decaying transposon relics. These data support a view that eukaryotic genomes contain a substantial fraction of DNA that serves little useful purpose for the organism, much of which has originated from the replication of transposable (selfish) elements.

          Sequence conservation analyses, including ENCODE’s, consistently indicate that only around 5–20% of the human genome is under detectable selective pressure. Some additional fraction of sequences has probably evolved new human-specific regulatory functions that are not conserved with other closely related species, but ENCODE’s publicized interpretation would require that such nonconserved regulatory sequences account for 80–95% of the genome, far outnumbering evolutionarily conserved regulatory sequences. Given the C-value paradox, mutational load, and the massive impact of transposons, the data remain consistent with the view that the nonconserved 80–95% of the human genome is mostly composed of nonfunctional decaying transposons: ‘junk’.”

          https://www.cell.com/current-biology/pdf/S0960-9822(12)01154-2.pdf

        • That’s one of the hypotheses. I’m not an expert on this stuff, but my understanding is that there is quite a bit of evidence for the alternative selfish-element hypothesis:

          I’m thinking about it differently. Every codebase I’ve ever developed/maintained goes through phases of expansion and then contraction. This corresponds to new features implemented in the easiest to code/understand but inefficient way, then refactoring where duplicate functions are removed, etc resulting in a reduction in the number of lines of code.

          And the refactoring doesn’t happen randomly either. It happens when the resources for that are available, there is an advantage for doing it, etc.

          So I do not share the expectation that natural selection should lead to “bigger genome -> more complex result” at all.

        • I have a problem with the code analogy since code is written buy a conscious entity with a goal in mind. Under those circumstances, yes, I would expect wild variations.

          But, the two lizard DNAs came from some common ancestor, so the default assumption is they are similar and the surprise is that something made one change drastically.

          It seems like the hallmark of the selfish hypothesis would, therefore, be a sudden change in size.

        • Terry said:

          “… the two lizard DNAs came from some common ancestor, so the default assumption is they are similar ”

          The “common ancestor” could have been way, way, back, to a most recent common ancestor that was nothing like a lizard. (Ever heard of “convergent evolution?” ) So your comment is similar to saying, “the default assumption is that lizard and human DNA’s are similar”. ( How similar? Maybe at a level which is virtually meaningless.)

        • Thanks for the link. The following from the start of the “Introduction” seems especially relevant to readers of this blog:

          ““What proportion of the human genome is functional?” remains a contentious question [1]–[3]. In great part this reflects the use of definitions of ‘function’ that differ from the traditional definition that is based on fitness and selection (see e.g. [4] for a discussion). For instance, equating functionality with annotation by at least one of the ENCODE consortium’s biochemical assays [5] results in approximately 80% of the human genome being labeled as functional [1], [6]. While this approach has the advantage of being empirical, it makes the definition of functionality dependent on the choice of experiments and details such as P value cutoffs. It is also questionable whether, for instance, introns should be classified as functional based merely on their transcription.” Sounds like the Garadn of Forking Paths.

          I would guess that there is a lot of the human genome that encodes for things like having freckles, or what color freckles, or where on the body, or whether they fade in winter, etc.

        • I suspect that large quantities of DNA are involved in subtle regulatory effects, like you need a bunch of bases between point A and point B such that the DNA bends in a certain way in the presence of some enzyme so that it can rapidly shut down transcription of something when that enzyme is present or whatever. it may not matter so much what the sequence is as long as it’s maybe long enough and has some statistical properties like a certain fraction of G bases or something

        • Martha,

          Here’s a bit of background. If you took a biology class in the 1970’s and maybe even into the 1980’s, you would probably have learned that 99% of the genome is junk. This was because many biologists only considered protein coding sequences to be functional, and they’re only about 1% of the genome. It then became widely recognized that non-coding regulatory DNA played a huge role in the phenotype, comprising 5-10x as much sequence as coding DNA. Thus, much more of the genome was functional than many biologists had thought, though it was still a small fraction of the entire genome.

          Many comparisons of DNA sequences between species also indicated that only a relatively small fraction of the genome showed any evidence of purifying selection, suggesting that most of the genome had no role in the survival and reproduction of the organism. The Rands et al. paper is an example.

          These findings were combined with the so-called “c-value paradox”: the size of the genome varies tremendously among even closely related organisms. For example, there is 120-fold variation from frogs to salamanders. Among the angiosperms, there is 1000 fold variation. There is an amoeba that has a genome that is about 200x as big as ours. The standard and still widely accepted explanation is that that there are a variety of selfish-DNA mechanisms that, if anything, are dysfunctional for the organisms rather than functional, but that can dramatically increase the size of the genome (Daniel is correct, though, that sometimes such DNA can play a structural role, even if the precise sequence is irrelevant).

          Anyway, in 2012 the ENCODE project published a paper in Nature claiming that 80% of the human genome was functional. This set off a firestorm of protest. You can read one response here:

          IIs junk DNA bunk? A critique of ENCODE

          https://www.pnas.org/content/pnas/110/14/5294.full.pdf

          As an outsider, I find the “pro-Junk” arguments more persuasive than the “80% functional” arguments.

        • These findings were combined with the so-called “c-value paradox”: the size of the genome varies tremendously among even closely related organisms. For example, there is 120-fold variation from frogs to salamanders. Among the angiosperms, there is 1000 fold variation. There is an amoeba that has a genome that is about 200x as big as ours. The standard and still widely accepted explanation is that that there are a variety of selfish-DNA mechanisms that, if anything, are dysfunctional for the organisms rather than functional, but that can dramatically increase the size of the genome (Daniel is correct, though, that sometimes such DNA can play a structural role, even if the precise sequence is irrelevant).

          Interesting, this does not seem at all paradoxical to me… I would expect the opposite of whoever is calling that a paradox: Orders of magnitude more code to accomplish a similar task = less advanced/refined/stable code.

        • Ed,
          Thanks for the additional links. The second one only gave me an abstract and references. I skimmed the first one. The following quote from it seems to make the (unsurprising to me) point that (as in many other situations) differing definitions of concepts, and taking theories as more deterministic than warranted, can lead to unwarranted certainty:
          “genomic anthropocentrism, unacknowledged conflation of possible meanings of “function,” questionable null hypotheses, and unrecognized panadaptationism are behind this most recent attempt to junk “junk.”… My aim here is to remind readers of the structure of some earlier arguments in defense of the junk concept (10) that remain compelling, despite the obvious success of ENCODE in mapping the subtle and complex human genomic landscape. Also, I will suggest that we need as biologists to defend traditional understandings of function: the publicity surrounding ENCODE reveals the extent to which these understandings have been eroded. However, theoretical expansion in other directions, reconceptualizing junk, might be advisable.”

          PS: Re your comment “If you took a biology class in the 1970’s and maybe even into the 1980’s”: My biology background is pretty atypical. I took high school biology in the late 1950’s and don’t recall DNA being discussed at all. Two or three years later I took a high school third-semester- physics course that had an introduction to ideas of quantum mechanics and relativity that was a strong influence on a lot of my subsequent thinking in many areas. My next encounter with biology started in the early 2000’s, when I had become involved in statistics and worked with some biology grad students in conservation biology and some in phylogenetics. (I also attended the phylogenetics seminar for several years, and tried to keep the math and statistics honest.)

  8. There are two comment-able themes at play here, theory vs framework and what-is-racism. A few words on each.

    I find myself somewhere in the middle (muddle) on theories vs frameworks. It is true that there are intellectual postures (interlocking systems of entities or forces thought to be central to explanation) that resist being tested the way theories can be because they are adaptable to a wide range of empirical outcomes. That is true of neoclassical economics (atomistic rational choice theory with ancillary assumptions about how markets equilibrate), where it’s well known that every graduate student should be able to write a model that explains the data you’ve already seen. I once had a conversation with a Marxist about the “tendency” for the profit rate to fall. I said there was no such thing, logically or empirically. He said tendencies are not governed by such criteria; they are like permanent background forces that are always pushing, but there may be other, temporary forces pushing against them, albeit for a very long time. This position struck me as nearly theological.

    But even frameworks are not universal and impervious to evidence. Neoclassical economics can definitely be falsified; the famous ultimatum game tests are examples of this. I think the combination of real world evidence and the Okishio theorem contradicts Marx’s falling profit rate hypothesis.

    Incidentally, one reason supply and demand analysis are so adaptable is that there is actually no theory at all to them beyond the sign of their slope, and not always even that. In practice economists *always* extrapolate from elasticity estimates using constant elasticity of substitution. If that doesn’t work, you just say there are nonlinearities, and that puts you conveniently in the world of epiphenomena.

    As for racism, two things. First, racism is fundamentally not about biological categories. It is a reduction of genetic diversity to a single hierarchical scale with at most a handful of “races”, to which it attributes a high level of explanatory power in individual cases. That is, it posits just a few genetic groupings, places them in a rank order, and believes between-group diversity is substantial relative to in-group. All the stuff about genetic markers for disease, etc., which is real, is more or less orthogonal to racism. If you accept my definition of racism, it is definitely falsifiable, although Andrew is right that there is also a lot of wriggle room for true believers.

    I have one other thought to share. My view, which reflects I suppose my political economy background, is that racism in the sense I describe is a hermeneutic aspect of colonial domination; it’s how colonialists and in many instances also the colonized understand the “naturalness” of their situation. If this is true (and it could be falsifiable), modern racism, which places Europeans and the European diaspora at the top of a compact racial hierarchy, is a product of European global domination and is destined to recede as non-European peoples gain parity or even superiority in economic and political terms. (My prediction is the rise of Japan and now, currently, China should elevate the ranking of “East Asians” or perhaps generate a new “racial” distinction between the successful East Asians and Asian and Pacific Island populations who remain subject to domination.) I also suspect that earlier periods of colonial domination spawned their own racisms—Greeks and later Romans vs barbarians, Chinese vs the bordering communities they dominated, etc. All of this is pure speculation and may be completely wrong.

  9. Most racists seem to be entirely ignorant of history. Thus they can make any idiotic claim that they like, totally oblivious to the fact that there are multiple counter–examples.

    If they were not so dangerous, they would be pathetic.

  10. When we constantly compare racial groups on outcomes, a racist framework seems hard to avoid. Also, does it even have to be about genetic differences to be considered racist these days? I think a lot of people consider cultural difference explanation to be racist too.

  11. I think there’s a strangely magical quality about genetic explanations where people turn their brains off and stop thinking critical once “genes” are a covariate in a study.

    If I regress lung cancer risk on an indicator for automobile form factors, people can very easily identify the problem with assigning a causal interpretation. Maybe people with hummers are more likely to smoke.

    If I regress a linear additive model for lung cancer on some fat subset of alleles, people go “oh hey, it’s the genes.” But people from the U.S. have different genomes on average from people from continental Europe, and people in the U.S. are less likely to smoke than people from Europe because of our heavy taxes and a strong anti-smoking education campaign. It’s just as confoundable of a causal identification strategy.

    In addition, there’s some abominably misleading terminology like “heritability” and “explained” (though the latter is borrowed from statistics), unresolved open questions on additivity assumptions, and an intractably wide space of “possibles”. The result is a perverse situation where laypeople (and yes, Murray and Wade and Rushton are laypeople to genetic statistics) are totally confident in sweeping genetic theories of intelligence, while experts are reluctant to make strong claims about seemingly basic relationships.

    Of course, they can point to any expert that agrees with their strong claims as support and any expert with skepticism as the “politically correct” orthodoxy, again falling into the trap of an unfalsifiable theory that can explain any result. Because laypeople making strong statements on a still-nascent field with so much fundamental disagreement between experts formed their conclusions first, and use the science as a rhetorical strategy.

    • +1. Exactly, and racist trolls have come to blogs like this to mock commenters for using nuanced and statistically correct definitions of the basic terms. They think (or pretend to think) that “liberals” are prevaricating in the face of what seems blindingly obvious to them, i.e. that there are better and worse races.

      • Kyle:

        I think this is a general problem in life, that people who believe X often think that, deep down, people who say they believe not-X are insincere, just saying it for political reasons, etc.

        We discussed this a few years ago in the context of this quote from social critic Charles Murray: “try bringing up the issue of single women having babies at your next dinner party, and see how many of your companions are willing to say, even in a private gathering of friends, that it is morally wrong for a woman to bring a baby into the world knowing that it will not have a father, and morally wrong for a man to impregnate a woman knowing he will not be a father to the child.” Murray recommended to upper-class Americans that they “stop being nonjudgmental in public about moral principles that you hold in private.” And he seemed to just take for granted that everyone agreed with him that it was a moral principle that single women should not have babies.

        Of course you can find lots of examples from the other side of the political spectrum too, where people on the left do not want to accept that others legitimately hold disagreeing views.

        • “I think this is a general problem in life, that people who believe X often think that, deep down, people who say they believe not-X are insincere, just saying it for political reasons, etc”

          I’ve grappled with this issue for a long, long time. Why don’t people change their minds more often? This question is actually a recurring theme on this website. This blog is abnormally open-minded and its commenters abnormally skeptical and willing to question authorities, especially published authorities. So why do so few people on this website change their minds in the face of intelligent and forceful counter-evidence?

          I would be very interested in stories from readers who have changed their minds about various things. I am open to being proved wrong about this.

          Currently, I find it disheartening that most people’s positions can usually be predicted by what they want to be true and arguments and evidence have little explanatory power. Whatever actually goes on in people’s minds (the insincerity and political reasons you write about) their behavior is observationally indistinguishable from a model where they reverse-engineer their arguments to come out the way they want them to come out. Moreover, people are often quite explicit about their reverse-engineering.

        • Terry said: “I find it disheartening that most people’s positions can usually be predicted by what they want to be true and arguments and evidence have little explanatory power. ”

          There may be a confounder here: namely, what kinds of arguments and evidence are convincing to which people.

        • This is true to some extent. Some people respond to emotional appeals, some tend to like economic explanations.

          But there is still a lot of reverse-engineering of very stuff that has nothing to do with attitudes or morals or world-views. I believe there are papers about this. People are much more attuned to logical flaws in opponents’ arguments while blind to exactly the same flaws in their own.

        • Terry:

          There like is some reporting bias here, its not likely every time some one learns something contrary to what they thought is going to comment that they have. In particular, sometime it takes a while for the change to sink in.

          In particular, I was wrong to over look the value of prior predictive distributions over or at least in addition to marginal prior plots. Where should I comment on that?

          Now, I have always wondered when someone is going to to a social network analysis of comments on this blog – some of us have been commenting for more than 10 years and there are 100,000s of comments.

        • I have recently began to wonder how many “unique views” (or what’s the appropriate term) an “average” post on this blog gets per day? Or what the total amount of “unique views” are per day concerning all the blogs on this site.

          If i am not mistaken the twitter account on which the blog posts get mentioned when they have been published has about 25000 followers, but these people may of course not read every post, every day.

          Is anyone aware of a (ballpark) number concerning the “average” views of the “average” blogpost on this blog? Or the total number of “unique views” concerning the entire site (all the blogposts) per day?

          I find it very hard to guess the amount of unique views. It wouldn’t surprise me if it’s “only” 100 and it wouldn’t surprise me if it’s 2000. I think everything above 2000 i would find surprising.

        • “sometime it takes a while for the change to sink in”

          This is VERY true. I try to make a point of admitting mistakes quickly to counter this, but usually it just takes time to realize you are wrong about something.

        • Quote from above: “Currently, I find it disheartening that most people’s positions can usually be predicted by what they want to be true and arguments and evidence have little explanatory power.”

          It’s one of the reasons i don’t want to participate in academia anymore. I feel the level of reasoning and argumentation might be sub-optimal at best, but most importantly i fear its importance is often not even realized and/or acknowledged. It sometimes seems to me that people just say stuff, and/or “debate” like they are a student at a highschool debate club, or a lawyer, or a politician.

          This is also why i think it could be crucial to educate students in (logical) reasoning and related stuff. I personally never received any courses on it at university, which still baffles me to this day. I think there could be a direct connection between the possible “poor” state of (parts of) academia/science and the possible “poor” state of the quality of reasoning and the teaching (or lack of) it.

          I view (logical) reasoning as kind of like math in a way, and to a certain extent: you can use it to backtrack certain “steps”, and can pinpoint where possible faults lie. Or at least pinpoint, and subsequently make clear, where there is a “fork in the road” that possibly warrents going back for a bit and/or examining if the correct “path” has been taken.

          I also think (logical) raesoning should be taught in highschool, as i reason it is useful for many things in life. At least i think it’s more useful than most of the stuff i was taught in highschool that also still baffles me to this day (i am talking to you physics, chemistry, advanced math, economics, music, physical education, etc.).

          (As a side note: I agree with how i interpret your expressed view concerning this particular blog. More specifically, that a lot of people on it may be abnormally open-minded and its commenters abnormally skeptical and willing to question authorities, especially published authorities. It’s one of the reasons i like this blog, and why i participate.

          To answer your question related to it, i think people on this blog may very well change their minds but that it could be possible that they do not subsequently share this and/or reply to make clear that someone wrote something that made them question something)

        • “I also think (logical) reasoning should be taught in high school”

          I have often thought this. But lately I have come to think that it is important to teach the ubiquity of much more egregious dishonesty. Logic is fun (for me), but it is bloodless for many. Everyone enjoys catching bad people in a lie, though.

          Every first-year college student should take a course where each week you hand out four articles and ask “what is wrong with this article”. Many articles will suffer from logical flaws, but many more will contain just flat out lies or omit critical information or be content-free sneers. Examples should be pulled from across the political spectrum to make very clear that lying is ubiquitous.

        • “I have often thought this. But lately I have come to think that it is important to teach the ubiquity of much more egregious dishonesty. Logic is fun (for me), but it is bloodless for many. Everyone enjoys catching bad people in a lie, though.’

          I used the words “(logical) resoning and related stuff” to indicate several (related?) things. I don’t even know which words to optimally use, because i never received any education in this stuff.

          Your comment about catching bad people in a lie could very well be a “hip” way to teach this to (highschool) students. You could have examples from politics, sports, science, etc.

          You could leave out whether or not someone is purposely lying though in my opinion. I think it’s a good thing to make students aware of this possiblity, but you can then mainly focus on the “rules” of reasoning and argumentation, and mistakes made concerning those things.

        • Anonymous:

          We largely agree. I would just teach the logic as the third or fourth step in the process. It would give students vocabulary and a framework.

        • There is a vast majority of our society who view the teaching of what we here would call “logical” reasoning as the last thing on earth (second only to perhaps “sex education”) they want their children exposed to in school. Many parents feel the two main purposes of schooling are to teach students a laundry list of useful task-oriented skills and to teach them to do what they are told. Logical thinking quite often leads to the opposite of doing what one is told.

          In my opinion it’s simply another way in which society is segregating itself over the generations. There’s the high achiever track for students whose parents want them to be able to reason, question, come to their own conclusions and synthesize unique perspectives. Then there’s the shut up and do your job track for those whose parents want them to be able to get a job but not be taught that every question is open to debate and reexamination.

        • Quote from above: “In my opinion it’s simply another way in which society is segregating itself over the generations”

          Interesting comment in light of discussion i had with someone concerning the idea of teaching highschool students (logical) reasoning.

          That person said something along the lines of “not everyone can think that way” or “not everyone is smart enough to understand this”. My reaction to that has always been that i think (logical) reasoning can be taught at different levels, just like math for instance.

          Just like i think math is useful for everyone to learn to a certain extent, and tailored to the capabilities of the student, so do i think (logical) reasoning is useful for everyone to learn to a certain extent, and tailored to the capabilities of the student.

        • Anon said:
          “…That person said something along the lines of “not everyone can think that way” or “not everyone is smart enough to understand this”. My reaction to that has always been that i think (logical) reasoning can be taught at different levels, just like math for instance.”

          To some extent I agree that logical reasoning can be taught at different levels. But I would augment that with the idea that teaching at a lower level shouldn’t be thought of as necessarily “terminal,” but should be aimed at developing skills that can be built upon to develop further skills.

        • Anonymous said, “I view (logical) reasoning as kind of like math in a way”.

          I regard it as more than “kind of like math in a way”. High school geometry was where I leaned logical reasoning. Back then high school geometry was mostly about proofs. Nowadays it covers a lot of other stuff (much of which is indeed worthwhile), but has little if anything about proofs.

        • “Of course you can find lots of examples from the other side of the political spectrum too, where people on the left do not want to accept that others legitimately hold disagreeing views.”

          I would be genuinely interested to know which views are held on the left which could be legitimately disagreed with. There are so many examples on the right!

        • Howdy there stranger. I don’t believe I’ve seen you round these parts before.

          What’s your motivation for popping up so sudden like with such a broad question?

          (I’m honestly interested in your motivation. I have become interested in what people are thinking when they post the things they do. An honest answer would be appreciated.)

        • My motivation is based on your specious claim that the left holds views which can legitimately be disagreed with. I don’t think that’s true and I think your point is wrong but I am willing to hear evidence hence my question.

        • Quote from above: “My motivation is based on your specious claim that the left holds views which can legitimately be disagreed with. I don’t think that’s true and I think your point is wrong but I am willing to hear evidence hence my question.”

          I hate politics, but am i understanding your point correctly that you think that “the left” (whatever that means) holds no views that can be legitimately disagreed with? Does this mean “the left” is some sort of wise, and allknowing, thing that sees everything how it is, and makes no mistakes, and stuff like that. This seems unlikely to me…

          Also see the following (perhaps you can spot some possible views “the left” has, or has had, that can legitimately be disagreed with):

          http://theconversation.com/how-the-politics-of-the-left-lost-its-way-86427

          “Of course, people that consider themselves as Left-leaning are not obliged to follow the ideas of the original Left. But it is important to understand how strains of Left thinking have twisted and turned from their original source. And recognise that alternatives are possible – particularly when the language of politics today is so broken. George Orwell wrote in 1946:

          One … ought to recognise that the present political chaos is connected with the decay of language, and that one can probably bring about some improvement by starting at the verbal end.

          The term Left has gone through major changes of meaning in the last two centuries. With this decay there has been a large degree of chaos. Meanwhile parties on the Left around the world are in crisis as a result of ideological fragmentation. If we are to have progressive change in society we need to first reconfigure the political map and no longer be restricted by what has come to define Left and Right.”

        • Given that “the left” is such a large and amorphous conglomerate it should be pretty easy to identify a view held by someone who self-identifies as being of the left that can be legitimately disagreed with. You might then argue that that viewpoint isn’t actually of “the left”, but then if you are of “the left” we would be in disagreement and I might have a legitimate argument that such a view is actually of “the left”. But since I sometimes think of myself as of “the left” that would leave us with the left disagreeing with itself about what it is, which may or may not be a legitimate disagreement depending on your definition of “legitimate” (a definition which I might disagree with).

          I think the blanket statement “GMOs are bad” qualifies as a view that is often associated with the political left. I would disagree with that statement and think I would have legitimate ground for such a disagreement.

        • “My motivation is based on your specious claim that the left holds views which can legitimately be disagreed with.” I think that, technically, Andrew said this, not me. But that’s a trifle.

          You are unaware of no leftist view that later turned out to be wrong or at least debatable? Have you yourself ever held a view that you later changed your mind about or came to doubt?

          Perhaps, if you are quite young, this may be true. I might have said the same thing when I was 17. Can I ask how old you are? (Feel free to not answer this.)

          You might want to consider that many on the left used to be quite friendly to communism. Some were even friendly to Hitler during the brief Hitler-Stalin pact.

        • “The Left” is not a homogeneous group. There is a lot of diversity (thinking here of quality of thinking skills in particular) among individuals in “The Left”, and similarly a lot of diversity in quality of thinking skills among individuals in “The Right”.

        • To elaborate just a little: Many of the differences between Left and Right are matters of values — that is, of what are the “givens”. In other words, the outputs of logical reasoning depend on the inputs, as well as on the reasoning itself.

        • I guess what I am getting at here is what salient policies on the left have no basis in fact? Are there any policies or viewpoints which are completely at odds with the facts and therefore could be legitimately disagreed with, i.e. not just a difference of opinion.

          There are plenty of instances on the right, for example, global warming denial and trickle down economics both viewpoints and enacted policy which go against the established facts.

  12. I have made this point before, but Karl Popper’s theory explaining scientific theories as having to be falsifiable is demonstrably wrong. Carnap showed it to be wrong a long time ago. Any theory that has nested quantifiers (which is all of them) is not falsifiable in the same sense that Hume showed that scientific theories are unconfirmable. In the absolute sense no scientific theory is verifiable or falsifiable. Popper’s theory is not helpful. Under it, the theory of evolution is not a scientific theory but a framework. Newton’s first law is un-falsifiable. If a philosophy of science has problems with two of the best contributions to human knowledge ever made, it isn’t worth the time applying it. The Quine-Durheim thesis points to much the same phenomena. When a theory is confronted with evidence that it is false, the scientist has a choice to challenge the evidence or revise the theory. The choice is under-determined by the evidence. Classical empiricism is false. And so is Popper’s project. There is no demarcation between science and pseudo-science in some absolute sense. It is not the theory that is unscientific. It is the behavior or rather the values of the individuals and community engaged in the inquiry that demarcate good science from bad or terrible science. People like Wade don’t really want to confront the evidence that they may be wrong. There is no genuine effort to weigh the evidence against their views and no genuine effort to state the lack of certainty that should ascribed to the evidence and attempt to get to the truth. And, of course, this distinction is also a matter of degree not absolute. Racists and sexists don’t really confront the evidence. How can we have geniuses who are women or of African descent? How can we see massive changes in the educational attainment of women and African-Americans in a time span that is totally inconsistent with a genetic explanation but consistent with an explanation about historical repression.

  13. Andrew, when you write “My feeling about Wade’s genetic explanations for economic outcomes is similar to my feeling about other all-encompassing supertheories: I respect the effort to push such theories as far as they can go, but I find them generally less convincing as they move farther from their home base.” Taking into account all the evil and terrible destruction that racism has had in the past, I am saddened that you could say that you “respect the effort to push such theories as far as they can go.” I think it is interesting to describe racism as a framework rather than a theory but I don’t think there is anything to respect about racism.

  14. While things like Freudianism might not be directly and literally falsifiable it’s not true that any theory has that property. As the philosophers of science who came after Popper pointed out even a theory as simple as the earth is a sphere can’t be directly falsified since one can always save it by adopting more and more outlandish supporting propositions, e.g., all the evidence and experiments suggesting the earth is a sphere are the result of induced hallucinations.

    The views that racial differences explain differences in outcomes is similarly subject to disconfirming evidence. True, you can always save the theory by adding more and more implausible excuses but if we find out that people with a given heritage don’t tend to (after adjusting for likely confounders) share any common outcomes across societies that’s strong evidence against the view.

    Personally, I think it’s more appropriate to distinguish between racism as an *inclination* and racial differences as a theory. There is nothing intrinsically wrong with entertaining the possibility that racial differences explain certain differences in outcomes. For instance, it’s a plausible hypothesis that Ashkenazi jews were selected for intelligence in the middle ages explaining both their supposedly higher average IQ and higher than normal incidence of diseases affecting the brain (cost of the mutations). Indeed, it would be wrong for scientists not to at least pursue that potential explanation.

    On the other hand there is something problematic about someone who is always extremely eager to explain everything in terms of race. That’s not just someone who has priors that make this plausible explanation but who seems to actively want this to be the explanation and is inclined to irrationally persist in this belief and ignore evidence to the contrary.

    • “There is nothing intrinsically wrong with entertaining the possibility that racial differences explain certain differences in outcomes. For instance, it’s a plausible hypothesis that Ashkenazi jews….”

      But one of the points of my definition of racism (above) is to differentiate race from ethnicity. There are gazillions of ethnicities with all sorts of internal commonalities and dispersions, but a racist is someone who tries to explain the world in terms of a few human groupings more or less corresponding to visual appearance and geopolitical history. Race is a social construct, ethnicity can be but often isn’t. Are Ashkenazi Jews “white”? Sephardic Jews? That takes us into racism. (And yes, I check off the “white” box on forms because I’m treated by my society as white, and that’s what the question is asking about.)

      • Humans, social animals that we are, find groups to denigrate. There is nothing special about racism versus other forms of discrimination. Violence between ethnic groups has been far more common and deadly than violence between races.

  15. You wouldn’t think I would have to make the point in this forum that it’s important to use language carefully in order to preserve distinctions. To say “X and Y are both bad and can lead to similar outcomes, therefore X and Y are basically the same thing” is to retreat from understanding.

  16. Parts of this reminds me of what i have only very recently began to ponder, also as a result of recent discussions on this blog.

    I wonder if science should stay out of investigating certain things, like racism. If science is about understanding, explaining, etc. things, then i reason the possible knowledge coming from this all can just as well be used to increase racism for instance. Or, perhaps all the good intentions of interventions that are supposed to tackle racism can have the exact opposite effects.

    Even if it’s possible (which i think it isn’t) for scientists to be are aware of this, and try to take this into account when investigating or proposing things, i wonder if they are then still doing science, or whether they would then actually be doing something else?

    I think racism could perhaps be dealt with better via the law, or via other things.

    When i ponder these matters, i always feel glad i am more focused on individuals and their choices and responsibilities, also when it comes to science. I dislike groups, group processes, and “nudging”, and stuff like that. Perhaps this is part of why i dislike a lot of social science as well.

    • “I wonder if science should stay out of investigating certain things, like racism.”

      A reasonable thought. But it assumes that the general public will become aware of the unpleasant results and be influenced by them. That assumes a lot of rationality and honesty on the part of the general public.

      The other option is to allow the research to continue, but keep the results confined to the cognoscenti, and lie about the results to the general population. That seems to be the path we are on now. Lying about scientific results is a booming industry nowadays.

    • I have also recently began to tie the sentiment described in the above comment to (parts of) social science in general.

      I wonder if the possible goals of (some? many?) social scientists may be achieved better if they would focus on, and investigate, totally different things.

      Perhaps it’s kind of like the example of how you can say “don’t think about a pink elephant” to people, but that saying this may not lead to not thinking about a pink elephant and/or may actually increase thinking about a pink elephant.

      I also wonder if the possible goals of (some? many?) social scientists may be achieved better when they would stop what they are doing, go find some other job, and make way for possibly more wise and competent people.

      Perhaps (some? many?) social scientists are standing in the way of the exact things they want to achieve themselves…

      Here is a section of a certain translation of chapter 2 of the Tao Te Ching that i find relevant concerning these (type of) thoughts https://www.taoistic.com/taoteching-laotzu/taoteching-02.htm:

      “So, the sage acts by doing nothing,
      Teaches without speaking,
      Attends all things without making claim on them,
      Works for them without making them dependent,
      Demands no honor for his deed.
      Because he demands no honor,
      He will never be dishonored.”

    • I’m pretty sure there’s a UN/WHO directive that forbids using or supporting research that would suggest one genetic makeup is superior to another. I read it years ago but have no idea how to find it again.

      Anyway, I remember being pretty disturbed when I was reading it. Politics/agenda driven research of any type is not a good thing.

      • The UN/WHO should “collaborate” and come up with some sort of world wide “collaboration” for social science!

        Sort of like a “New World Order” of Social Science.

        Think about the things they can achieve by doing that!

  17. Hi, I understand from biologists that there is much more variability in genes within a given “race” than between any two “races”, and so it is unlikely that gene variability between races can explain outcomes. This is not to say that race is irrelevant. I’ve just been reading Darity’s interesting work on “stratification economics”, which proposes that economic disparity between races have inter-generational effects. (I hope I am representing Darity’s work fairly.) But I am sure Darity would not want to argue that there is some gene-based explanation for the disparity that we observe today. Instead, the disparity has its roots in (for example) slavery, racial prejudice, etc – i.e., social/historical roots, not biological roots.

    I haven’t read Wade’s work, and I hope he isn’t arguing that genes is a reason for the racial disparity that we observe today.

    In terms of the Malays and Thais, I would argue that the reason for the economic success of Chinese migrants in SE Asia has something to do with selection (e.g., unsuccessful migrants could return to China) and some kind of “resource curse” effect, where the indigenous populations have weaker incentives to invest in their own human capital since they had privileged access to land or other resources. In any event, I doubt genes had anything to do with the racial disparity between Chinese migrants and the indigenous populations.

    It would be nice if Nicholas Wade were invited to write a post for this blog, so we can all learn the reasons for his views.

    • “I haven’t read Wade’s work, and I hope he isn’t arguing that genes is a reason for the racial disparity that we observe today. ”

      It was a while when I read his book, but I think I can still more or less accurately characterise his arguments as such:

      Step 1: Imagine introvertism/extravertism is real dimension, and it’s influenced by genes.

      Step 2: Introverts/Extroverts prefer different institutions and create different culture, if allowed

      Step 3: Genes do not directly affect behavior, but are moderated by environment. Hence, different institutions and culture influence behaviour and change how genes are expressed

      Step 4: Assume two populations have different initial ratios of introverts/extroverts. Say in one introverts are 55%, and in second they are 45%.

      Step 5: The population with larger number of introverts will skew more into creating culture and environment favourable for introverts, which in turn will influence the society by changing the way genes are expressed

      Because of that, small differences in ratio of the genes in two different populations will produce wildly different environments/cultures (if strip of external influences).

  18. If “framework” = taxonomy then I agree. Before information about something can be collected or disseminated the thing must be named even if simply “that thing over there.” Differentiating between thing and not-thing requires categories that specify those traits necessary for thing-ness and thus our natural inclination to the taxonomical.

    The trouble is that when assessing categories, e.g. Carnivora, we make both objective and subjective assessments. Thus the hyena would hardly be surprised to learn that we have found the cheetah to be the faster of the two, for it is; but if the hyena knew that we found the cheetah to be beautiful, graceful and sleek and the hyena to be ugly and cowardly it might rightly take offense. For our perspective is not the hyena’s; and what it is like to be a hyena we cannot know.

    The lines we draw are never as bright as we imagine. Take Chordata. Having a dorsal bundle of nerves is a helpful way of knowing that I am not a slug. And yet the slug and I share an awful lot of highly conserved genes. And the invertebrate slug has been around a very, very long time. Whatever I and my backbone may think of slugs they long preceded humans, will likely outlive us, and as far as can be determined take no notice of our subjective assessments – which, I think, makes them rather more wise than the typical human.

    Things are otherwise when those being binned perceive the binner’s categories and their definitions. And for the same reason. Just as I cannot know what it is like to be a hyena, I cannot know what it is like to be e.g. Andrew Gelman. So because I cannot know what it is like to be someone else it is wisest to, as my Dad sternly said to me once when I hurt my sister’s feelings “Keep your damn opinions to yourself unless you want to hear what everyone thinks of you.” Just sayin’.

  19. The fundamental problem is apparent in a phrase such as “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.”

    The essential construct of the model, “race” is not well-defined, perhaps not even defined. As a consequence all that follows has no useful meaning.

    Whenever someone speaks of the Chinese, the blacks, the Ashkenazim, the Malay, etc., the listener should be skeptical. The social model assumed by such classifications is often (generally?) false, even were it possible to give sense to them in terms of (generally unavailable) genetic information. For example, in my own family, which I suspect is more typical than not (though I have no idea, and this suspicion is a product of my ideological biases), I have genetically related (if one believes everything the grandparents and aunts and uncles say) cousins of first and second degree living, as native born residents, in four different continents and some outlying islands, who could be classified (evidently not very accurately) as white, Jewish, black, Asian, and Hispanic (some of them even as three or fours of the five!), whatever those terms mean.

    • “Race” is actually extremely *well* defined. Already over a decade ago it was possible to assign people using genetic analysis to common-sense “races” almost perfectly accurately:

      http://www.gnxp.com/blog/2007/01/race-current-consensus.php
      https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1196372/

      …and as I understand since then the process has become even more accurate. (23andme takes in a healthy amount of revenue answering these questions for consumers.)

      Put another way, do you think the word “color” is meaningful, even though there’s no unambiguous dividing line between “red” and “orange”? Because in that sense, the question “is that person black or white”? is even *better* defined that the question “is that color green or blue?”

      As such, to the extent that your extended family actually is as difficult to racially classify as you state, it would seem to be extremely unusual.

  20. This criticism of “racism” framework is equally true as a criticism as “non-racist” framework. If you say genetics can’t possibly explain even part of the variation in outcomes of certain groups, and then you bring either bad governance, colonialism, geography etc it’s exactly the same; you just bring tones of different factors to explain away something unexpected. “Culture explains X” or “environment explains X” does not mean anything, since “culture” and “environment” are so all-encompassing.

    • “This criticism of “racism” framework…”

      I’m not clear if “this criticism” refers to Andrew’s original post, or to Dan F.’s 4.44 am comment above, or to some other comment.

      • I was refering to the sentences starting with “His model is pretty sophisticated: genes affect culture which affects behavior. But it’s one of those can-explain-any-possible-data sorts of theories. If a group does poorly, it’s either bad genes or bad governance that’s unrelated to genes. ” etc. My point is that both “racist” and “anti-racist” (when defined as “genetics explain part of observable effects” or “genetics does not explain even a part of observable effects”) BOTH can be seen as “all-encompansing” theories/frameworks.

  21. How is neoclassical economics unfalsifiable? If most people were regularly donating large portion of their income to charities anonymously that would clearly falsify neoclassical economics/rational choice.

    • Murmur:

      First off, I don’t see why you feel that regular donations of large portions of income falsifies that theory but irregular donations of small portions of income does not.

      In any case, see here regarding the falsifiability of neoclassical economics. There are two ways the economist can argue:

      1. People are rational and respond to incentives. Behavior that looks irrational is actually completely rational once you think like an economist.

      2. People are irrational and they need economists, with their open minds, to show them how to be rational and efficient.

      It’s like Freudian psychology: any behavior fits the theory. Or its opposite. In your example, if most people were regularly donating large portion of their income to charities anonymously, this would not at all falsify neoclassical economics/rational choice. The economists could just say that people get positive utility from making those donations.

      • But people don’t derive as much utility from charities as large parts of their income. In general, utility calculations are difficult. I deliberately chose an example where the difference between utility lost and gained is large. So if people were doing this then it would be against rational choice.

        Karl Popper made similar arguments against the falsifiability of natural selection. However, there’s nothing unfalsifiable about natural selection, it’s just difficult to calculate fitness functions. Similarly, in rational choice theory it’s difficult to calculate the utility function in the general case. That doesn’t render it unfalsifiable or useless. There are a lot of cases where the difference in utility between two choices is clearly very large. Rational choice theory leads to good predictions in those cases.

        • Utility functions are unobservable, it’s not just difficult to calculate them, it’s impossible to calculate them. Or rather, for almost every behavior there is some utility function that would predict that behavior. The existence of such a utility always provides an out for any “failure”. The same is true for fitness.

          What you *can* falsify is the assertion that a utility takes on a particular form, or that a particular fitness function describes the selection pressures.

Leave a Reply to Dan F. Cancel reply

Your email address will not be published. Required fields are marked *