Calcification of the merit castes

By Razib Khan | June 23, 2012 11:38 pm

After ragging on Chris Hayes for a week I decided to check out the conversation above between Hayes and Mike Konzal about his new book, Twilight of the Elites: America After Meritocracy. As Mike suggested the book does seem more nuanced in its take than the piece in The Nation which highlighted the role of high-stakes testing at Hunter College High School. In the conversation above Hayes supports his suppositions that test-prep was excluding blacks and Latinos by asserting that that is what the teachers themselves believe. I wouldn’t dismiss this out of hand, but it certainly isn’t enough to make me accept that portion of Hayes’ argument. People have all sorts of weird misconceptions.

 That being said, there are many descriptive and positive elements of Hayes’ narrative which I can agree with. By this, I mean that I do agree that phenomena such as the “iron law of oligarchy” do exist, and are pervasive. Additionally, I also accept that merit-based systems eventually tend toward corruption, as the measuring-sticks become not the means toward ascertaining productivity, but the ends toward which one optimizes. I’m not a Leftist, so my prescriptions would be different from those of Hayes, but I see much of the same reality he does (I am not an egalitarian in the way that any progressive, liberal, or Left-of-center person would recognize, first of all).

But there’s one aspect which I’m always wondering about, which I think gets neglected: the fact that the nature of the human capital stock itself is shifting. For example, in Hayes’ book there is one reference to genetics, where he is quoting someone else (thanks Amazon “look inside”!). A point I have made repeatedly is that in an efficiently operating meritocracy where the sorting process (selection) is effective you would eventually see a convergence toward a static equilibrium. As smart people marry other smart people, their children will inherit their smarts. As one transitions into a genuine meritocracy the first generations will see a great deal of class church, social mobility if you will. But eventually the smart lower class people will have descendants who are in the upper class, while the shiftless upper class individuals of earlier generations will have descendants who move down the class hierarchy after exhausting their inheritances. At that point the amount of churn will be due to stochastic processes, rather than expected shifts in individual life status due to mismatch between human capital and social status.

Assortative mating is a major driver of this process. I recently read that inter-class is rarer today than it was in the past. Why? Modern communication and mobility allows for greater self-segregation, but one factor may be the educational and professional advances of women. With a larger pool of educated female professionals it may no longer do for male professionals to marry their secretaries or nurses, as was not uncommon a few generations ago. But in the process males may now be narrowing their mating pools.


Comments (12)

  1. David

    I have the anecdotal impression that men still have the social permission to marry their secretaries and waitresses, but the quality of women has improved so much that educated men can typically do even better. I thought it was pretty well documented that educated women are having a hard time finding equally educated men to pair with, and there’s less social room for them to marry their plumber or waiter. This surplus of educated women has left educated men lacking a sense of urgency about marriage, since they don’t need it to get their companionship, fun and sex from educated women. I think the effect is that females, not males, are narrowing their mating pools.

    About the smarts sorting equilibrium you mentioned, I think you overestimate just how much smartness was valued by smart men of previous generations. Sure, they preferentially married women with some qualities that were loosely correlated with smartness, like hotness and wealth, but also qualities that were anti-correlated, like obedience, meekness, conformity and piety. Smartness in women is much more valuable now, it seems to me, which makes me think that the smartness sorting will accelerate.

  2. Darkseid

    i guess libs are more focused on the fair *outcome* and cons are more concerned about the fair playing field. i actually tend to agree mostly with Hayes as it’s obviously dangerous to have concentrated power but i get off before most lefties make it to the point where they’re inventing all sorts of ways to “prove” that there aren’t that many inherent differences at the outset. i’m not sure if they are really that aware of even the obvious differences in capabilities from human to human because it’s not something they like to believe or emphasize. ….which seems weird because it seems you’d have to acknowledge there’s a problem in order to solve it.
    for example, here’s the top voted comment from your post on Saturday. this kind of blanket denial is not at all uncommon to see. of course, one downside of having a more diffuse power structure is that it’s hard to actually get anything done.

  3. Murray and Herrnstein pointed out the increase in assortative mating back when in The Bell Curve.

    However, even with assortative mating, there are still plenty of smart, poor or middle income, whites, Hispanics, blacks, and Asians. In my opinion, the move *away* from test scores as a primary indicator and *towards* grades and “soft” indicators , which are extremely unreliable, is one way in which the established elites maintain their edge.

    Thanks for the mention, by the way.

    Darkseid, that series of comments was a tad mindboggling. I particularly liked the guy who said “well, what was weird about 1997?” as if it was just some weird anomaly.

    The “poor whites do better than non-poor blacks” is an extremely robust finding that shows up without any need for a study. It’s all over the NAEP data, and you can find it in every year of the California standardized test results. Poor whites do about as well as non-poor Hispanics, too.

  4. Lucretia

    IIRC, Murray’s latest book cites the decreasing number of college grads marrying partners without a college degree as a sign of increasingly assortative mating – without ever raising the question of how college education has changed on the whole in the past fifty years, or even mentioning the vastly different forms of educational experience and validation that all fit under the banner of a college degree.

    If I look at retired faculty members where I teach, it seems that most of them married women of above average intelligence. A common pairing I observe there and among my contemporaries is a +2SD with a +1SD. Interestingly, in talking with some of these faculty wives of yore, I have noticed that they are not uncommonly very well-rounded culturally, more so than most faculty members I see around me. The time they did not dedicate to developing high-level careers was often spent on worthwhile cultural pursuits – instead of wasting time in committee meetings like their daughters do now, they would sit at home reading Tolstoy or Proust.

  5. Miguel Madeira

    “A point I have made repeatedly is that in an efficiently operating meritocracy where the sorting process (selection) is effective you would eventually see a convergence toward a static equilibrium.”

    I think that this reasoning only is correct if we assume that the effects of genes over social class are monotonic.

    Or put it in another way – if there is a kind of “social heterezigotic fitness” (where upper classes are made of people with both A and B genes, where people with only A or only B genes end up in lower classes), you will have social mobility even in a perfect meritocracy.

  6. #5, segregation load itself is probably going to be an issue in allowing for overdominance to maintain churn. rather, more plausible to me is frequency dependence and/or fluctuation fitness landscapes.

  7. Karl Zimmerman

    A few points somewhat related:

    1. I’ve been reading, off and on, the new Kim Stanley Robinson book. Although he doesn’t talk about it directly, it strikes me that if offworld colonies ever become feasible (meaning positive fertility), the resulting group will have been subject to the most severe incidental selection to boost intelligence in human history. This is because the first space colonists would likely all have STEM backgrounds (or, at worse, have a leavening of above-average intelligence military types as well). Unless technology got a lot more idiot-proof, and launch costs got a lot cheaper, it’s hard to see why random people with no technical training would ever be sent into space, meaning space settlements would probably be the first human cultures with literally no dumb people.

    2. I have heard (correct me if I’m wrong, I quite likely am), that while IQ is strongly correlated with education and overall life success, it’s only correlated with income up to a certain point. Essentially that income peaks for IQs somewhere in the 120s-130s. However, geniuses, while often upper middle class, have a tendency to eschew out-and-out wealth, with many born to wealthy families essentially giving up on running the family business and going into a career in science just because they enjoy it. If this is true, I do wonder if future gene therapy allows for boosted intelligence the rich are going to be in for a nasty surprise when their superkids don’t show much interest in furthering their family legacies.

    3. While I find most of the social arguments used regarding the black-white test score gap flimsy, I wonder what you think about the hypothesis of stereotype threat. Given everything I know about cognition, it’s clear that priming people really affects their performance on an unconscious level. While of course I cannot say (and the originators of the term don’t claim) 100% of the gap is due to this, it could very well make test score gaps far larger than actual base differences in intelligence.

  8. Balaji

    Perhaps it was Theodosius Dobzhansky in his 1973 book “Genetic Diversity and Human Equality” who first suggested that a meritocratic society that practised assortative mating would result in a stratification of society by intelligence due to the heritability of intelligence.

  9. Florida resident

    Comment on #8.
    Richard Herrnstein has published in 1973 the book “I.Q. in the meritocracy”, essentially with the same statement about stratification. I have actually read this book in full.

    Respectfully yours, F. r.

  10. Florida resident

    Addendum to comment # 9.
    R. Herrnstein, “I.Q. in the Meritocracy” (loc. cit.), Notes to Chapter Five, (p.223):
    […] A brief paper by T. Dobzhansky, “Genetics and the Social Sciences” (in GENETICS, edited by D.C. Glass. New York: Rockefeller University Press and Russell Sage Foundation, 1968) makes the general point that social movement often carries genetic material with it. … .
    The September 1971 article “I.Q.” by R. Herrnstein in “Atlantic Monthly” warned (according to Herrnstein himself) that progress toward equalization of opportunity is, in fact, progress toward a _hereditary_ “meritocracy”.
    I (F. r.) was not able to find the said 1971 article by Herrnstein anywhere on the web.
    With respectful greetings to Mr. Khan and to Balaji,
    your F. r.

  11. If merit and social class get too far out of line you get situations like the acid bath that WWI class driven military interactions in Europe gave rise to in the European class system. Another example is the huge surge in union power until the people who would otherwise have become union leaders were co-opted into the ruling class via higher education. A third example would be the wave of corporate insurgent law firms that arose when Jews were excluded for non-merit reasons from silk stocking New York law firms.

    Any class system that lacks of means to co-opt the most able members of the lower classes develops a disgruntled and capable underclass that could be the seed of its own destruction.

    Then again, if society isn’t changing very fast, then the innovation that smart people can bring about may not be very valueable relative to the status quo of following tradition. Meritocracy may be more important in interesting times than dull ones.

  12. Mathme

    Darkseid, I think it depends on what you think is an “even playing field.” Do you think that giving someone dyslexia accommodations on a history exam is not even? Because, as someone who considers himself liberal and an educator, that’s the sort of thing that I consider to be an even playing field. Otherwise, we don’t know what we’re testing and we’ve compromised the validity of the history exam by also conflating it with decoding ability. I want the outcomes to be as fair as they can be in that we are testing what we suppose that we’re testing. I don’t want to test your math ability and find out that because you’re an ESL student, you failed because the story problems were difficult to comprehend. I’m testing your ability to understand English instead of your ability to comprehend mathematics. It’s not about being lovey-dovey to everyone, it’s about making sure that our tests are valid measures of the ability in the subject that the test is supposed to be over.

    Sure, we can argue about English language learning or special education, but if you’re good or bad at my subject, I want to be sure that I’m getting the most accurate measures as I can of your ability. If that’s not what you’re talking about with the “even outcomes” or whatever, then please clarify. Do we want everyone to learn the material and succeed? Yes. Would we like to help students who are struggling if we can help them learn the material better? Yes again. I honestly don’t know what you’re getting at.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar