Thread by @lymanstoneky

Your quirky demographic history fact of the day:

here's the historic *marital completed fertility* for five well-documented Chinese lineages going back WAY before modern birth control.

Source: etheses.lse.ac.uk/4223/

There's absolutely no way you get these numbers without having widespread *marital fertility control*.

Folks the Zhou, Zha, and Gu lines here are marital TFRs around 3 or 4. That's not that much higher than many modern post-contraceptive marital fertility rates!

Especially once you account for different child mortality. Probably if MCFR was 3.5, the share surviving to puberty would be just 2.5 or so, whereas today an MTFR of 2.5 yields surviving fertility of like 2.4 or 2.45, and we commonly see MTFR of 2-2.5 in western societies.

Demographic history is not the baby-breeding-factory you have been told! It contains extraordinary diversity!

Note that the Zha and Gu lineages here are relatively "elite" lineages, whereas the other 3 are essentially "commoner" lineages. So commoner fertility was higher than elite fertility.... but even some commoners like the Zhou family had clear marital fertility control!

Because China spent so much of its history resting so near its Malthusian limits, which were so extraordinarily high already per square mile due to rice cultivation, it's possible low-fertility social norms were more adaptive in China than elsewhere.

The other very-low premodern TFR I know of offhand is Sri Lanka: also rice agriculture, and also Buddhist-influenced. Mongolia is fairly low too: not rice, but does have Buddhism.

The suggested mechanisms are several: historic China had long breastfeeding intervals, and "minor marriages" were very infecund for many reasons, and Confucianism regulated even marital sexual activity.

But the big driver is polygyny and concubinage.

The specific norms of Chinese polygyny meant that adding extra wives/concubines did not tend to greatly increase male fertility. Husbands tended still to sleep with only one partner in a period of time.

As a result, when a smaller share of men monopolize a growing share of women, as was most common in *elite* households, the birth rate per woman fell even if the birth rate per man was stable or rose somewhat.

So this points to the complexity of historic demography. Virtually no formal demographic model of historic fertility includes formal accounting for polygyny, not least because these models all arose from Western demographers studying more-or-less monogamous societies.

But polygyny (or much rarer, polyandry) can throw a huge wrench in those models. This paper found NO EVIDENCE of intentional parity-stopping-behavior or birth spacing beyond long breastfeeding intervals.... and yet fertility rates are far lower than in Europe of the same time!

Why?

Because many women were 2nd wives and more wives per husband tended to reduce the likelihood a given wife conceived a child, because Chinese husbands tended to be "serially monogamous within polygyny."

Another paper in the same thesis shows that number of marriages rose strongly as male status increased: the average highest-status man had 2 marriages, the average lowest-status man had less than 1 marriage!

High-status men also tended to have more sons than low status men, *entirely* because they had more wives.

Finally, this separate paper is more narrative-historical than quantitative, but it argues persuasively that Roman families had *intentional marital fertility control* characterized by family size targets. tandfonline.com/doi/full/10.10…

What's interesting is "Roman elite fails to produce enough heirs" goes back to the days of the Roman *monarchy*, which raises the question of whether Roman small-family norms had a possibly very-early origin, and might explain the rise of non-familial power transfers.

In this case the argument would be that although low fertility norms might be individually maladaptive, in group selection terms having elites in particular have low fertility norms could give rise to non-hereditary political structures, i.e. the Republic and then the empire.

Of course whether the empire had a hereditary power transfer system is much debated. Sometimes yes, sometimes no. It was less hereditary than many true monarchies but more than the Republican period. There's a nice book called "Byzantine Republic" extended this arg.