Cousin Marriage Rates Have Varied Over Human History

The new paper, linked below, has made it possible to tease inbreeding rates out from incomplete ancient genomes, which has added an accurate measure of a key society level culture practice from prehistoric periods to the arsenal of anthropologists. Previously, ancient DNA could only show the relatedness of people in shared burials.

[C]ousin-marriage was far less common in the prehistoric and historic past than it is in some modern societies. In particular, it is far less common than it is today in the Islamic world and in India. That being said, the r.o.h values [runs-of-homozygosity (to measure inbreeding)] do decrease with agriculture from hunter-gatherer periods, indicating that large farming societies were more exogamous…before a recent shift in some areas to endogamy. . . .

The authors, for example, had transects from Pakistan, and in the past people in this region were not nearly as inbred. The dynamic that is important to remember is the confluence between population growth and the development of what Samo Burja calls “social technology”. In a modern world where societies are undergoing demographic transition and in the earlier stages of that process, so subject to massive growth, there will be lots of cousins to marry. In Malthusian societies, where families are just replacing themselves, there will not be as many cousins to marry. In other words, when there is a strong ideology of cousin-marriage, the limitation is going to be the number of cousins.

As many Islamic societies undergo demographic transition, I predict cousin-marriage will decrease as a phenomenon simply due to the reality that smaller families produce smaller kindreds from which one can select a mate.

But the second aspect here is social technology. In David Reich’s group’s work on India, it seems clear that strong endogamy as we see today did not really crystallize until ~1,500 years ago. . . . The practices of these societies are not from time immemorial but develop due to the exigencies of history and social evolution. Practices that we see as “conservative,” such as arranged-marriage, are actually innovations.

From here citing a table from Human Parental Relatedness through Time – Detecting Runs of Homozygosity in Ancient DNA.

* Previous analysis of cousin marriage, the frequency of which which varies greatly from culture to culture today, can be found at sister blog Wash Park Prophet, which noted that:

For the United States as a whole, the cousin marriage rate is about 0.2% (so there are about 250,000 people in these relationships), based mostly on old and not completely reliable data that are still unlikely to be wildly wrong, but "if you take a global perspective, consanguinity is not rare at all. Of the 70 countries studied, only 18 have consanguineous relationships as less than 1 percent of all marriages. In five countries, more than 50 percent of all marriages are between people who are second cousins or closer, and in Burkina Faso, it’s estimated that two of every three marriages are consanguineous."

A source for that post contained this map illustrating rates of consanguineous marriages globally:

The patterns of cousin marriage shown in the map above largely emerged around 500-1000 CE, in connection with the solidification of jati caste endogamy in South Asia and with the spread of Islam in the Middle East and North Africa, while the emergence of Christianity as the main source of marriage norms in the European dark ages, following the demise of the Roman Empire, reduced endogamy there.

A low rate of cousin marriage is one of the factors that some scholars believe drive some of the culturally distinctive aspects of WEIRD populations. See, e.g., Joseph Henrich, "The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous" (September 8, 2020).

In contrast, high rates of cousin marriage are associated with corrupt and weak governments, since obligations to one's extended family clan and even more extended family tribe can tend to override loyalty to a central state. See, e.g., Mahsa Akbai, et al., "Kinship, Fractionalization and Corruption" (October 3, 2016). Clan based societies seem to be the default "state of nature" in the absence of strong non-kinship related government, often overlapping with "cultures of honor" which are also associated with weak governmental power.

* Consanguinity measured by ROH is also related in endogamy rates. In modern India, as of 2011, the jati outmarriage rate is still just 5%. Jati are much more specific than varna or racial or macro-religious denomination type classifications. 

By comparison, native born Asian-American and Hispanic women in the U.S. have an outmarriage rate of almost 50%. Previous discussion of interracial marriage in the U.S. can be found at this post in sister blog Wash Park Prophet. It notes that:

About 15% of all new marriages in the United States in 2010 were between spouses of a different race or ethnicity from one another, more than double the share in 1980 (6.7%). Among all newlyweds in 2010, 9% of whites, 17% of blacks, 26% of Hispanics and 28% of Asians married out. Looking at all married couples in 2010, regardless of when they married, the share of intermarriages reached an all-time high of 8.4%. In 1980, that share was just 3.2%. 

Gender patterns in intermarriage vary widely. About 24% of all black male newlyweds in 2010 married outside their race, compared with just 9% of black female newlyweds. Among Asians, the gender pattern runs the other way. About 36% of Asian female newlyweds married outside their race in 2010, compared with just 17% of Asian male newlyweds. Intermarriage rates among white and Hispanic newlyweds do not vary by gender. . . . 

About one-in-five (22%) of all newlyweds in Western states married someone of a different race or ethnicity between 2008 and 2010, compared with 14% in the South, 13% in the Northeast and 11% in the Midwest. At the state level, more than four-in-ten (42%) newlyweds in Hawaii between 2008 and 2010 were intermarried; the other states with an intermarriage rate of 20% or more are all west of the Mississippi River. . . . 

The overall out marriage rate in the absence of any in marriage preference would be something on the order of 35-40% of newlyweds rather than the current 15%, although geographic considerations and social class considerations might trim the race neutral out marriage rate to closer to perhaps 30%. 

There is no race-gender combination in a person who is more likely to marry a person of another race than they would be if random chance determined pairings, and for almost all minority groups the difference between random pairings and actual out marriage rates is considerable. Even the people most prone to outmarry (for example, certain subcategories of U.S. born Asian and Hispanic women) have outmarriage rates on the order of 50%-60%, in a society where outmarriage rates would be closer to 80%+ in the absence of some endogamy effects. Once one controls for geography at the scale where people actively interact with potential spouses and for social class, quite a bit of this endogamy tendency is diluted, but any way you measure it, it is still there. As a general rule, people tend to marry people who are as similar to them as possible without being closely related. . .  

Native-born Hispanics were nearly three times as likely as their foreign-born counterparts to marry a non-Hispanic in 2010 [36.2% v. 14.2%]. 

The disparity among native- and foreign-born Asians is not as great, but still significant: Nearly four-in-ten native-born Asians (38%) and nearly a quarter (24%) of foreign-born Asians married a non-Asian in 2010.

Among Asian newlyweds, the intermarriage gap between native and the foreign born is much bigger for Asian men than for Asian women. In 2010, native-born Asian male newlyweds were about three times as likely as the foreign born to marry out (32% vs. 11%). Among newlywed Asian women, the gap between native and foreign born is much smaller (43% vs. 34%). The gender differences are not significant among Hispanic native- and foreign-born newlyweds. . . . 

[I]nterracial marriages that are most vulnerable to divorce involve white females and non-White males (with the exception of white females/ Hispanic white males) relative to white/white couples. Conversely, there is little or no difference in divorce rates among white men/non-white women couples, and white men/black women couples are actually substantially less likely than white/white couples to divorce by the 10th year of marriage.

Another point of comparison would be outmarriage rates for religious groups that make up a small share of the total U.S. population (Jews, Mormons and Muslims):

87% of Mormons and more than eight-in-ten Muslims (84%) in the United States are married to people with the same religion. 

Rates of intermarriage vary considerably among the major U.S. Jewish movements or denominations. Virtually all Orthodox respondents who are married have a Jewish spouse (98%), and most married Conservative Jews also have Jewish spouses (73%). Half of Reform Jews who are married have a Jewish spouse. Among married Jews who have no denominational affiliation, 31% have a Jewish spouse.

* Another way of measuring genetic segregation and divergence between populations is with an F(ST) statistic. Some populations in ancient times had genetic distances from each other that were as large as those between Europeans and East Asians today. The modern distinctions are summarized in this table:

From a 2003 paper (modern DNA sampling was possible long before good samples of ancient DNA were available).

For example, this was approximately the amount of genetic division between European hunter-gatherers and the first farmers of Europe, as illustrated by comparisons using, for example, ancient DNA:

The FST test indicated that European hunter-gatherer groups (hg_ce, hg_sca and hg_cant) did not show statistically significant differences between them, but they are significantly different from any population compared. . . . On the other hand, it was noted some differences within the European Neolithic groups, with the Neolithic group in Central Europe (neo_ce) showing the highest number of statistically significant differences in FST test, whereas the Neolithics from France and Catalonia (neo_fr and neo_cat) showed the lowest number of statistically significant differences.

Similarly, a 2010 ancient DNA study found that "early farmer and hunter-gatherers were from two well-differentiated populations (FST = 0.163; P<10-6)." 

Similarly, this was approximately the amount of genetic division between the hunter-gatherers the Levant and the hunter-gatherers of the Caucasus mountains and the highlands of what is now called Iran, a genetic division that carried over into the first farmer populations of these areas which was derived from local hunter-gatherers.

We computed squared allele frequency differentiation between all pairs of ancient West Eurasians, and found that the populations at the four corners of the quadrangle had differentiation of FST=0.08-0.15, comparable to the value of 0.09-0.13 seen between present-day West Eurasians and East Asians (Han). In contrast, by the Bronze Age, genetic differentiation between pairs of West Eurasian populations had reached its present-day low levels: today, FST is ≤0.025 for 95% of the pairs of West Eurasian populations and ≤0.046 for all pairs. These results point to a demographic process that established high differentiation across West Eurasia and then reduced this differentiation over time. 

Our data document continuity across the hunter-gatherer / farming transition, separately in the southern Levant and in the southern Caucasus-Iran highlands. The qualitative evidence for this is that PCA, ADMIXTURE, and outgroup f3 analysis cluster Levantine hunter-gatherers (Natufians) with Levantine farmers, and Iranian and Caucasus Hunter Gatherers with Iranian farmers. We confirm this in the Levant by showing that its early farmers share significantly more alleles with Natufians than with the early farmers of Iran: the statistic f4(Levant_N, Chimp; Natufian, Iran_N) is significantly positive (Z=13.6). The early farmers of the Caucasus-Iran highlands similarly share significantly more alleles with the hunter-gatherers of this region than with the early farmers from the Levant: the statistic f4(Iran_N, Chimp; Caucasus or Iran highland hunter-gatherers, Levant_N) is significantly positive (Z>6).

From Iosif Lazaridis, "Genomic insights into the origin of farming in the ancient Near East," 536(7617) Nature. 419–424. (August 25, 2016). doi: 10.1038/nature19310

A few years later, the genetic sources of Anatolian first farmers was established with ancient DNA:

Anatolia was home to some of the earliest farming communities. It has been long debated whether a migration of farming groups introduced agriculture to central Anatolia. Here, we report the first genome-wide data from a 15,000-year-old Anatolian hunter-gatherer and from seven Anatolian and Levantine early farmers. We find high genetic continuity (~80–90%) between the hunter-gatherers and early farmers of Anatolia and detect two distinct incoming ancestries: an early Iranian/Caucasus related one and a later one linked to the ancient Levant. Finally, we observe a genetic link between southern Europe and the Near East predating 15,000 years ago. Our results suggest a limited role of human migration in the emergence of agriculture in central Anatolia.

Feldman, M., et al. "Late Pleistocene human genome suggests a local origin for the first farmers of central Anatolia." 10 Nat Commun 1218 (March 192019). https://doi.org/10.1038/s41467-019-09209-7

In contrast, the first farmers of most of the rest of Europe (also here and here), North Africa and South Asia were predominantly migrants from elsewhere.

From this March 2011 paper.

There was also significant genetic distance between the first farmers of Europe and the Corded Ware and Bell Beaker people derived populations that partially replaced them in the early Bronze Age. For example, a 2015 paper explains that:

The Corded Ware are genetically closest to the Yamnaya, 2,600 km away, as inferred both from PCA and ADMIXTURE and FST (0.0116 0.002). If continuous gene flow from the east, rather than migration, had occurred, we would expect successive cultures in Europe to become increasingly differentiated from the Middle Neolithic, but instead, the Corded Ware are both the earliest and most strongly differentiated from the Middle Neolithic population.

 

As indicated in the table from that paper above, the genetic distance in terms of Fst between the most diverged Middle Neolithic population from the Corded Ware and Bell Beaker populations is 0.042 in the case of the Bell Beaker sample is 0.052 in the case of the Corded Ware population - only about half of the difference between the hunter-gatherers of the Levant from those of the Caucasus mountains, or between European hunter-gatherers and the first farmers of Europe, or between Europeans and the Han Chinese - but about twice as great as between any two West Eurasian populations in existence today.

This suggests these cultures were strongly endogamous with respect to these respective populations despite their sometimes reasonably close geographic proximity to each other.