Substack cometh, and lo it is good. (Pricing)

Against the WEIRD!

Charles Freeman, author of The Closing of the Western Mind: The Rise of Faith and the Fall of Reason (and in the UK, The Awakening: A History of the Western Mind AD 500 – AD 1700), sent me a long email critiquing Joe Henrich’s The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous. Freeman is a bit concerned about the adulation for Henrich’s work when he believes it is fundamentally flawed.

Why? Read his review on Amazon, it’s titled “2.0 out of 5 stars Henrich’s central argument and its offshoots are not supported by historical fact.” Read the whole thing.

All that said, I do want to put in a word for Henrich and his ambition. Lots of people on Twitter and elsewhere have pointed out “did you notice this whopper on page x! The guy doesn’t know what he’s talking about!?!?” Of course, he’ll make errors, The WEIRDest People in the World, and Henrich’s current research program, is broad, interdisciplinary, and it “breaks things.” This isn’t to excuse sloppiness, errors should be fixed. But when you take big risks, it is to be expected that picayune errors will slip in.

Nevertheless, what Freeman alludes to is not a small error. It’s a deep philosophical chasm, and the issues he brings up bothered me too. How truly effective was the ideology promoted by the Western Christian Church? Quite often there is one thing that is promoted officially, amongst elites, and another thing that is practiced. Freeman points out the reality that most people entered into common-law marriages, and that the religious authorities and government only took an interest in elite high-status marriages due to matters of property. Even when it comes to the elite exceptions and workarounds were common. Ideals were there to be broken.

There needs to be more exploration of the details of Henrich’s conjectures, though I think Freeman is too hard on him. After all, elite emulation is a real practice, and Greg Clark’s work (amongst others) shows that elite overproduction means that the ancestors of modern people may mostly be elites of the past.

Addendum: Some of Freeman’s questions are going to be answered soon. For example:  “How many people did actually marry their cousins in pre-Christian society? How does one define a cousin marriage as against a non-cousin marriage? How could one find and assess the evidence when most marriage/ cohabitation arrangements were unrecorded and then map, as Henrich does the decline in percentages of cousin marriages in Europe?” The Krause group at Max Planck is crushing through a lot of Late Antique DNA, and will answer these cousins in regards to endogamy definitively.

20 thoughts on “Against the WEIRD!

  1. That sounds unconvincing to me.
    My country, Argentina, in colonial times, was only half the size it is today, but it was still a huge place with a small sparse population. Around 500.000 in 1800 and 1.500.000 in 1860, and it had a big population of nomadic cattle herders called gauchos.

    There were few priests and some towns would only get to see mass in the local church 1, 2 or 3 times per year. Because of that many marriages were post facto, people would hook up, become stable couples, the women would get pregnant, and they would get married when it came to be that a traveling priest on horseback visited the town. But people still didn’t marry their cousins.

    What matters isn’t the ability of the church to enforce its rules, but the ability of the church to instill a taboo in the population.
    Even these days, most catholics don’t go to mass, they use condoms and contraception, but the cousin marriage taboo still remains as healthy as ever as something only a s*ck f*ck would do. The Catholic rural Cletuses may lose their virginity with a cousin (and the women may have only anal sex thinking it preserves their virginity), but they would never marry or have kids with a cousin.

  2. I’m halfway through WEIRD. I think I bought it because you wrote about it (maybe?). Definitely an excellent book. It has grand propositions so I’m sure many of its arguments will be attacked, some to be modified, some to be reduced to rubble..

    There is plenty of data on actual cousin marriage, so it’s not all supposition based on what the Catholic Church aimed at.

    For example, figure 7.3 on p.241. Panel A, prevalence of actual cousin marriage in 93 Italian provinces. Panel B, frequency of blood donations in each province. (This latter being a proxy for attitudes towards strangers). Striking negative correlation – very low rates of cousin marriage correlates to high rates of blood donations; high rates of cousin marriage correlates to almost zero blood donations.

  3. “WEIRD”. Yeah…..I’m not convinced This name is about honest research, to me it seems to be about sensationalism.

  4. I think that’s actually a better critique of Freeman’s own thesis. How much does his idealized conception of the rational/scientific minded Greeks and Romans (before Christianity supposedly came and ruined everything) apply to anyone other than a few elite, rarefied philosophers? I find it easier to believe that cousin marriage taboos proliferated throughout society and had significant effects than the rediscovery of Aristotelianism by a handful of scholars.

  5. I’m hopeful adna will be able to further close out whether large scale cousin marriage declines or increases happened in particular, e.g. from between 40% or 20%…

    But then we kind of already know that large declines from high (double digit) rates to very low rates (less than single digit) are highly likely to have never happened in Europe from the sparse medieval adna and methods to test for cousin marriage frequency we already have.

    Rather tho., Henrich’s arguments (in published papers) are presented in log % cousin marriage terms – e.g. in his large scale country correlation, the systematic differences between 2.5% and 40% are supposedly as large on “individualistic psychology” as between 2.5% and 0.25%.

    Will we really get enough adna and reliable adna samples to infer that cousin marriage in medieval Western Europe was 0.25% or 2.5% (for ex), in a way that’s immune to sampling bias claims? Or prove that it’s decisively due to medieval church policies rather than simply technology, large scale markets, states, etc doing their thing and mixing up population structure?

    Maybe very large scale modern dna will allow for some more inference about cousin marriage rate change over centuries (although there’s a limit to that as RoH broken up); we already know that large scale samples show some direct change on the margin within last 100 years (between older and younger).

    Henrich makes some other arguments on differences of “kinship intensity”, that don’t rely on cousin marriage rate, but these seem a bit less quantitative, at any rate not through much that can be looked at through adna. A matter of combining how experts have coded societies into an index – e.g. China is coded with intensive kinship, Thailand with unintensive kinship, which seems possibly largely due to the difference in the Theravada Buddhist vs largely Confucian state ideology (where latter tends more to idealize strong kin bonds between parent and child).

  6. (To go on a bit, all Henrich’s plots are based on log cousin marriage; blood donation, conformity-obediance, etc.

    So it’s really important for his thesis that large scale samples can generally evidence the cousin marriage rate at low end over time to within probably 0.1%, and certainly it seems these correlations are likely collapse into a heap if we’re not actually sure if a place had 1.5% or 2.5%, or 0.5% or 2.5%.

    Another thing; Since some countries obviously change on this over time – see data from consang.net – it’s really important that all the estimates are from *exactly* the same time if possible, or at any rate certainly aren’t separated by 30-40 years…

    I’m hopeful big scale biobanks and population samples can do this).

  7. On Freeman, I have no real idea if his specific objection holds water, but, historians of antiquity (and classicists in general) who focus on how people lived and thought, from the evidence we have, are exactly the sort of people I would like to see looking at Henrich’s thesis.

    Some of the talk around Henrich’s ideas I’ve seen has tended to simplify to: “The early Catholic Church destroyed cousin marriage, this created a new type of psychology, unknown in history, which in turn created representative government, unknown in history”.

    Which seems really bizarre to me because it completely removes both classical representative government and classical democracy from the picture.

    It seems good then to have people who study the classical world to comment on those ideas, to put them into some more perspective. Does the world before Christianity really look like Henrich seems to kind of be describing it as; one of omnipresent intensive kinship which undermined collective institutions and this itself made scientific advance and a modern economy, beyond the state of the classical world, impossible?

    (On a bit more of a further tangent, in the same vein we can often have people who occasionally burble on about how Confucius created the idea of a meritocratic bureaucracy that would later be realized and hold true in China… but we see exactly the same pattern in ancient Egypt, two millennia before the birth of Confucius; a hereditary autocrat, surrounded by a bureaucracy appointed on merit, not blood – https://www.britannica.com/place/ancient-Egypt/The-king-and-ideology-administration-art-and-writing

    “The king’s position changed gradually from that of an absolute monarch at the centre of a small ruling group made up mostly of his kin to that of the head of a bureaucratic state—in which his rule was still absolute—based on officeholding and, in theory, on free competition and merit … The elite of administrative officeholders received their positions and commissions from the king, whose general role as judge over humanity they put into effect. They commemorated their own justice and concern for others, especially their inferiors, and recorded their own exploits and ideal conduct of life in inscriptions for others to see. … According to royal ideology, the king appointed the elite on the basis of merit, and in ancient conditions of high mortality the elite had to be open to recruits from outside. There was, however, also an ideal that a son should succeed his father. In periods of weak central control this principle predominated, and in the Late period the whole society became more rigid and stratified.”

    An autocrat with no peers in a troublesome blood aristocracy backed by a bureaucratic machine of commoners with “merit” and no complicated and challenging independent dynastic basis of their own, is deeply attractive to the rulers of long-standing strongly centralized empires. The same reasons “Enlightened Absolutists” in Europe much later tended to patronize them. Thus will tend to prevail in increasingly centralized long-standing monarchies over time. The ideas of Master Kong on “merit” probably had nothing to do with it, as much more than a justification for a trend. Meritocratic bureaucracy was absolutely nothing new (and it didn’t seem to particularly push forward living standards in Egypt over Mesopotamia, just as it never did later in China over clan ruled Japan).

    But we’d only know this if we thought about ancient Egypt as another example of a heavily centralized dynastic monarchy with a developed bureaucracy, and tried to draw it in. Otherwise we’d easily fall for ideas that this sort meritocratic bureaucratic form is a “uniquely Chinese characteristic” invented as a result of Confucius’ thought… So deep and comparative historical knowledge is critical to test theories).

  8. I am pleased that there is discussion on this one and thank you, Razib, for initiating one. It was not so much the issues about cousin marriages that I had problems with (though I am pleased to know that some research on this will be published soon), it was that Henrich argues that the bans were so restrictive that people were ‘forced’ to seek mates outside their ethnic and tribal groups. If you know of evidence of this happening, outside dynastic marriages, I would be pleased to know.
    My main point, with evidence to support it , is that the Church simply did not control marriages before the later Middle Ages. And the later evidence from the Florentine archives shows that kinship was really important in the politically volatile cities of northern Italy.
    I may have been too harsh on Henrich, the ensuing discussion will tell. I have to admit that I had really looked forward to this book, as it tied in with my recent book The Awakening and I really felt I should read it, and I paid full price. I was really disappointed which may be reflected in my review.
    The main thing is to go beyond the adulatory reviews to probe more deeply.

  9. I shall need to study Freeman’s take. But, I have objected to Heinrich’s thesis before.

    The commercial and industrial revolutions begin in Protestant northwestern Europe that had cast off the Roman Church and its rules a couple of centuries earlier. England, in particular, practiced cousin marriage. It is a subject of the domestic novels of Austin and Trollope and is never objected to on any moral or genetic grounds. Charles Darwin, of all people, married his first cousin.

  10. Is there anywhere in The Weirdest People that Henrich defines what degree of relationship between a couple counts as a ‘cousin marriage’? I may have missed it.
    As you will see from my review (which you now need to press on ‘All Reviews’ to find on Amazon), it is not clear how Henrich knows that ‘cousin marriages’ (however he defines them) declined at the steady rate he describes. The vast majority of marriages were unrecorded .Does DNA analysis of bones from burials help us on this one?
    Then, and this is crucial, how does the decline in cousin marriages lead to psychological individualism? I have a problem here. Henrich suggests that the effect was so dramatic that the only way of marrying someone as to seek out a partner from another tribal or ethnic group. Evidence?
    There are a myriad of issues raised by Henrich’s book. I feel that many reviewers have been overwhelmed with the breadth of his evidence and the panache with which he presents it. I feel that we now need a second phase when there is intensive discussion of his arguments. Historians will have much to contribute as these issues have been argued over at a sophisticated level for decades. ( I remember studying the reasons for British industrialisation in my economic history course at Cambridge in the late sixties!)

  11. Henrich argues that the bans were so restrictive that people were ‘forced’ to seek mates outside their ethnic and tribal groups; If you know of evidence of this happening, outside dynastic marriages, I would be pleased to know.

    One thing I tell you, though it is much more recent and of course I’m just an anonymous commenter and not some big shot whatsoever.

    70 years ago, Italians in the hinterland of Southern Brazil (which were mostly Venetians) had a very stark prohibition against cousin marriage. People say that their colonies kinda broke apart because, eventually, everyone was related…

    (this is an exaggeration, of course; it was more like every family had 15 children and transmitted the land only to one, then the rest found it better to migrate).

    Meanwhile, in the hinterland in Northeastern Brazil, the descendants of the earliest (Portuguese, though not all of old christian background) colonists married preferentially among cousins. Preferentially. That region is up to now a a place to go if you want to find weird genetic diseases (see here for an example), and is definitely much less developed than the rest of the country.

  12. I am interested in the question why was the Catholic Church against cousin marriage. I don’t see any difficulty with the Catholic dogma.

    Also, prohibition of cousin marriage isn’t unique to Catholics. North Indian Hindus also had it since forever.

  13. The Wikipedia article on ‘cousin marriage’ is very thorough. It does not give much support to Henrich’s rigid thesis. The real question remains how does one know whether a marriage is a cousin marriage, first and second cousins according to how you define them, when most marriages were unrecorded.

  14. A very week point in this theory is that the Protestant world seems much more WEIRD than the Catholic – more individualistic, more cosmopolitan, historically more democratic, more nuclear family (even today, children in Catholic countries usually only leave parents home when they marry – the whole “loser who lives in his parents’ basement” trope does not make any sense in the Catholic world).

    And the stereotype of the inbreeding royal family were the uber-Catholic Spanish/Austrian Habsburgs.

  15. @Mactoul: “I am interested in the question why was the Catholic Church against cousin marriage. I don’t see any difficulty with the Catholic dogma.”

    I think the theory is that the Catholic Church has an interest in making marriage difficult, because unmarried and childless people usually gave their properties to the church as heritage.

  16. @Charles Freeman:

    “I was muddled by Henrich’s assertion that the growth of WEIRD individualism led to more cooperation and trust -I would have thought that it was the opposite. ” (…)

    “How do ‘collective brains interact with individualism??”

    I did not read Henrich’s book, but I imagine that the reasoning is that in a individualistic society, based on voluntary associations, you have to be capable of interact and cooperate with everybody, instead of in a kin-based society, based in groups where you belong from-the-birth, where you only have to cooperate with members of your pre-ascribed group.

  17. Miguel- I have added a Comment to my review, both that on Amazon.com and Amazon.co.uk, to show that kinship groups which comprised several hundred were known in fifteenth century Florence. So Henrich’s argument that they were ‘demolished’ does not hold for one city which had been Catholic for many centuries. As most marriages/ cohabitation agreements were unrecorded, we simply cannot say whether Church marriage laws had much effect. There is evidence that most couples just got on with it. As I understand it, the very first record of a couple getting married in church is ninth century.
    We need specialists in these areas to assess Henrich’s arguments before the adulatory reviews that this is a truly groundbreaking, innovatory book get set in stone. ( I am just a very generalist historian who has, as you will see from googling me, done a lot of work on the early and medieval church but the argument does not make any sort of sense to me.)

  18. @Charles, I may be misunderstanding the questions but some thoughts

    1) on defining cousin marriage prevalence and use of dna, ancient and modern DNA can pretty robustly define whether a sample is the offspring of parents who are biologically related through biological grandparents, great-grandparents, etc and from that can infer a frequency of cousin marriage, assuming vast majority of reproductive events occur within socially accepted marriages and that cousins no more or less likely to have offspring. Of course those ideas are maybe somewhat open to some challenge (are these assumptions all correct?).

    2) As to how Henrich is defining cousin marriages (which is a different question), as far as I know he sourced data from consang.net (maintained by a health professor and a collaborator) and they stated the following: http://consang.net/index.php/Summary“As a working definition, unions contracted between persons biologically related as second cousins (F ≥ 0.0156) are categorized as consanguineous. This arbitrary limit has been chosen because the genetic influence in marriages between couples related to a lesser degree would usually be expected to differ only slightly from that observed in the general population. Globally, the most common form of consanguineous union contracted is between first cousins, in which the spouses share 1/8 of their genes inherited from a common ancestor, and so their progeny are homozygous (or more correctly autozygous) at 1/16 of all loci.”. This is as far as I can tell; if he uses another source it should ideally mention it in the text of his papers / book. The authors there have used various published data sources to compile their data tables. I am not sure how confident they would be in the accuracy of data bounds in which Henrich puts them to, which require high accuracy to with 0.5% at low and mid bound. Presumably Henrich has consulted with them tho!

    3) As far as I know Henrich does not document changes in cousin marriage in history directly via records, or try to, and the data and methods for inference via ancient DNA were not available to him at the time of writing (more of this will come out over time, as well as more accurate genetic measures for many of the present day regions he discusses). The method for rates of decline of cousin marriage looks to look at the frequencies in present day (from above source), then correlate with “time under Western Church” and infer a constant “rate of decline” under Western Church from this. Adna will test if correct or not; so far his inference suggests that 0 centuries under Western Church should be 15% on average, roughly 1/7 people (that’s the intercept of the correlation). See main figure from Schulz 2019 – “The Church, intensive kinship and global psychological variation”.

    But direct analysis ancient DNA from a Roman transect and medieval samples so far, and it’s still early days, does not seem to support this estimate of frequency (with caveats from 1) about assumptions behind how ancient DNA can inform questions).

    Examples – https://www.biorxiv.org/content/10.1101/2020.05.31.126912v2 – “Human parental relatedness through time – Detecting runs of homozygosity in ancient DNA” (and https://www.biorxiv.org/content/10.1101/2020.09.24.311597v2 – “Human Inbreeding Has Decreased In Time During The Holocene”) though caution these are recent preprints

    The first paper’s analysis of a Italy Roman Imperial transect (150 – 340 CE) found no “close kin” (inc. cousin) offspring among 48 samples. While if we aggregate Imperial and Late Antiquity then there’s 1/73, or 1.3%. (E.g. at a guess about 10x lower than Henrich’s mean estimate). That seems reasonably fair as Henrich’s Western Church measure counts from around 500CE and the last Late Antiquity samples date from 450-550 CE. Likewise 16 Langobard samples from Hungary finds no close kin offspring. More samples will build up more knowledge in terms of rough “big picture” though getting precision could be harder (There is a big set of pre-Christian Viking Age ancient DNA samples that would be interesting to test for another example).

  19. Thanks,Matt. I shall read/ think this through. I am grateful for input from an area where I am not strong. Genetic information from DNA is vital especially to plot migration/ settlement.
    If there were not that many cousin marriages to start with then things are not going to change much, certainly not to the degree to men being ‘forced’ ( his words) out into other ethnic and tribal groups to find mates as Henrich suggests.
    The evidence from Florence suggests that non- cousin marriages were sought to form links with other kinship groups. You need to strengthen your base in whatever way you could. So non-cousin marriages strengthened kinship by enlarging it not ‘demolished’ it. Was this typical?
    My real problem is with Henrich’s assertion, so far as I understand it, that European populations ( though presumably only those subject to the Catholic Church) became individualistic AS A WHOLE. How would this happen economically or socially?
    I would prefer to argue that certain, not all, regions of Europe became OPPORTUNISTIC, in that there were opportunities for individuals or communities to get ahead in a variety of ways. The classic example is the Dutch Republic of the seventeenth century that dominated European trade ( and gathered wealth) for decades even though there was a small population.

Comments are closed.