Addendum on Charles Murray

With graphics, transcriptions & a firecracker

I hear thunder, so let me get my umbrella. Let’s start with this following transcript that I created today with the help of a cool AI, excerpted from Charles Murray’s talk with Glenn Loury. This, is a much clearer bit than the other thing with Tucker Carlson, so there’s no obligatory seriousness here, I’m putting it all as clearly as I think is required. Pay attention to the bold italics as I take them up.

Murray & Loury
MURRAY Are there genetic differences among people who self identify as different ethnicities? Yes. The most clear differences are the continental level. East Asia. Europe, Sub-Saharan Africa. Are these genetic differences to be expected. Yes, absolutely. Somebody put it this way. So the chances that different groups that are separated geographically would have evolved exactly the same is about the same as the chances of dropping a handful of silver dollars and they all land on their edge. That's just not the way it works. If you have geographic boundaries, if you have other kinds of boundaries, And you're talking about tens of thousands of years. There is a lot of room for evolution to have taken place. And by the way, I'll just give you a little more background.

Sorry. I love to talk about this because it's fascinating. All right. But also people have a throwaway wrong impression. People Stephen, Jay Gould is famous for this said the races can't be different because really this first out of Africa is 65,000 years ago. And then evolution by mutation just takes forever.

There hasn't been time for major differences to occur. And at the time he made that point in the 1980s, that was a respectable way to look at it subsequently, especially after the sequencing of the genome, we've learned that you can get much more rapid evolution than that through the variation that is called standing variation.

Okay. I won't go any further into that central point. Number one is. Certainly true. And we already know it's true that there are lots of population differences and in various genetic variants, that is not a controversial statement among geneticists. Do we know what effects those have on a variety of traits?

Not very much. Are we learning a lot more as time goes on? Yes, we are. Are these differences going to prove too scary? No, they, they won't that there will be differences. They will be usually moderate or modest and they all involve overlapping distributions. And at this is this Glenn, I swear I am amazed by how otherwise intelligent people have such a hard time grasping the concept of overlapping distributions, which is to say we've got two bell curves to coin a phrase and they overlap.

Millions of blacks are smarter than millions of whites, despite the existing differences in cognitive scores. And by the way, millions of whites are better coordinated millions of blacks, despite the black dominance in major sports. Yes, you have, you had differences between groups. They do not determine your fate.

They don't lend themselves to superiority or inferiority because they're bundles each. If you think of the bundle of characteristics, you have of a whole bunch of things about you personally that are above average and below average in somewhere near the average. Whether we are talking about beauty or charm or industriousness or whatever they are, you have this very complicated set of attributes.

And the idea of saying, you can say one of these set of attributes is superior and the other's inferior is just an abuse of language.

LOURY Okay. Let me stop you because there's a lot of lot on the table. I want to try to summarize. So we're talking about populations, not about individuals,


The claim that populations which have descended from ancestors who were isolated from one another for long periods of time could have distributions of genetically determined traits that are different from one another is an uncontroversial almost self-evident claim.


That needn't be a big deal.


It certainly needn't be saying that any population is inferior because there's not one trait, there's a lot of different kinds of traits, which are differently distributed between these population groups. And they will manifest with different dimensions of human performance because human performance isn't one thing.


The Persistence of Sociology Amidst Failure of Social Policy
Charles Murray agrees with me, and I am a reasonable man. What's interesting is that I'm not a social scientist. My parents were social scientists and I decided long ago never to walk in my parents' shadow. So there's a kind of grating familiarity I have with this kind of narrow dimensionality studying narrow benchmarks of success that are deeply engrained in the interests of sociologists and the people who believe that sociology can make good use of public policy. I myself am quite skeptical that sociology makes the best use of public policy when the goals are about advancing the status of some group of human beings.

But that's not where Charles Murray agrees with me because he is a social scientist and he hopes that public policy and public political debate can be adjusted towards what is liberal and reasonable. The problem is that we've been allowing irrationality and illiberality simmer for so long in our emotional stilted debates, that the chances are slim that there will be any political manifestations of truth and light. I'm just that skeptical. Murray on the other hand loves talking about it because he finds it fascinating. I’m stoic. I say it is what it is and I don’t care about the emotional component and the wishful thinking. That’s why I’m getting down to the interpretation of genetics.

Murray and Loury both agree with me that there are bundles of characteristics in large populations that we don't measure well that overlap in those categories we call races or as Murray explicitly states as ‘self-identified ethnicities’. We agree that it is silly and foolish to suggest that these uncontroversial differences mean anything like inferiority or superiority. Furthermore we agree that it is a kind of innumeracy (my harsh word) that keeps people from understanding the basic math with regard to those variant distributions that we do measure. 

But before I drop what is nothing more than a very small snarky firecracker, let me remind you of why I'm the bomb. That has something to do with whey I'm not of a racially motivated polity, rather that of a scientifically oriented tribe that readily accepts that tribal membership is not zero-sum and can even be contradictory. BUT, the contingent nature of every conjecture of science, and genetics specifically means I'm happy to be proved wrong. AND I am against social engineering as a civil libertarian, even by the most capable of social scientists. FURTHER I have no faith in the narrow interpretations of human history as a tale of conflict between ethnic populations. All that superior and inferior blather is not of interest because it’s wrong and clearly out of scope of this talk. That’s racial baggage projecting or at best social science doing its circle jerking.

So I offer the two graphics as an insight to the dynamism I see in this whole business, and why my panties are not in bunch.  Begin with figure 1.

In the first, I illustrate what Loury and Murray agree are the overlapping regions of average genetic characteristic of three populations. (Note I do not have a bell curve stencil and these distributions are meant to be equal in every way except for their position on the x-axis) These net averages peak along the spectrum of each trait (in a non-binary way, by the way) at lines A, B and C. The A group has the lowest median attribute. The B, the average median and the C has the highest median. How important is the difference? Well it depends upon two things.

The first dependency is how socially important you make of the difference vs how much you emphasized the space under the three bell curves that are common to all three populations. The second is exactly how distant those median peaks are. So even if you decide that the distance between the peaks is fairly static and not amenable to be changed much by social policy you can still decide (as we did during COVID) to make a big deal about the social value of whomever you deem to be 'essential workers'. But then you'd be thinking like me and giving priority to something other than genetics and the heritable of bundles of traits. The question remains as to whether you obsess over these bundles of attributes like your social science profession demands of you. On the other hand, you might decide race must matter.

Is Every Difference Material?
Let me drill into the following point. I challenge anyone to tell me exactly what the precise number at which the distances between median peaks becomes an edge condition over which we should be socially concerned. I do so specifically by asking that question without putting any measured gridlines between the median lines. So I could be talking about median grey hairs per head, median blood pressure or median physical height. All heritable. But yeah I know you want to obsess over race and IQ, so tell me how many IQ points you say makes a race superior or inferior. Go ahead, quantify your own racial theory. I know you want to. 

Oh yeah, the firecracker. If you are a social scientist, AND you agree that public policy must be effective, AND you make racial determinism central with IQ as the indicator, AND you accept the extreme interpretation of Murray as racial superiority THEN you must agree that one of the best policies must be that of miscegenation. That's right. Jungle fever. So get out there and mix it up. Oh by the way, you'll also have to recognize how homosexuals are, by their refusal, extending and persisting this inequality by not participating in the eugenics your policy requires. Bang.

Now let's get into the second chart, also with three population groups. Note that this is a more accurate way from my POV to talk about bundled attributes whether or not they are heritable, but generally not. Note that the arrows indicate the dynamism of the overlap. This does not lend itself to Venn diagrams that we are accustomed to. In fact, keep in mind that we share something like 95% DNA similarity with pigs or some such oversimplified genetic explanation. My point is that humans share 100% DNA with humans and the overlap is much more like this than a Venn diagram. Nevertheless, presuming we can identify three such populations we must take note of the arrows which will indicate evolutionary and/or social pressures that change the amount of overlap in random and immeasurable ways. Keep in mind miscegenation and migration between cultures & societies and all policies and their intended and unintended consequences. 

The other important part of this population mapping is that this dynamism isn't graded as better or worse, like IQ, and is more accurate with regards to our ignorance of how genes map to any particular human behavior. Scientists (even social scientists like Murray) admit that simply don't know, and racists only pretend that they know. 

So, there is the entire tiresome thing spelled out for you, and you should have a good idea of how I think about the distributions of everything among human groups and why I emphasize consistently that we should rescue our humanities by actively considering many more dimensions and attributes of ourselves and others.

One last note is that in the next section, Loury goes on to talk about crime and punishment. I didn’t listen because I have something of an aversion to all conversations that deal with the intersection of race and crime. I happen to think it’s one of the least interesting subjects in all of history and an insulting way to talk about people. Then again, I’m not a social scientist.