the median/average age at marriage for women ranged usually from about 18 to 25, not 16 to 18.
I disagree with that summary. In your article, you say for ancient Egypt there was a median of 18, in Greece it was 17 to 19, in China it was ~15, and in Babylon it was 23, but only for slaves.. So, in antiquity, the median age of female marriage for non-slaves was in the teens across all societies observed.
Now for medieval England, you show data consistent with a median of 19 or 20, with peasants being a bit higher (20 or even 21) due to their poverty, and wealthy people being around 19. So this is still teen marriage. You write:
Kowaleski (1998, Table A1) also provides more data from three groups of elites. She refers to these as the “Bristol elite”, “London merchant elite”, and “London elite”, with sample sizes of 18, 41, and 17, respectively, and with averages ages at marriage ranging from 16.8 to 19.1. The weighted average of these figures is 17.42 years.
For Germany:
Among the patricians of the town of Frankfurt, the average age of 43 brides was 18.8 years.
Italy:
The most major study was published by Herlihy & Klpaisch-Zuber (1985) in their analysis of the 1427 Catasto (a type of census). Their figures show a mean ages of girls at marriage of 18.86 and 19.28 in the urban and rural populations of Tuscany, respectively (pp. 205-206).
France:
“French data suggest that in medieval towns and cities, women generally married around age 20”. … In five cases, a woman’s age at marriage can be roughly calculated, typically with some assumptions, based on how the woman was identified in court records (e.g., married for x years, aged y years). Karras calculated a lower bound estimate of 17.25 and an upper bound of 20.75 years (fn. 89) … In ten cases, a woman’s age at (premarital) deflowerment is known. This can be an indicator for age at marriage because this was typically done under the intention of marrying soon after. The ten ages range from 16 to 25, with an average of 17.5 years.
So for every society you cover up to the year 1500, all of them display median ages of female marriages consistent with the number 18. Broadly the reasonable confidence interval here is 17 to 19. The simplest model is that every society had a default median age of female first marriage of 18, plus or minus a year or two based on contingent factors. This model fits every society observed except for Babylonian slaves; this indicates mid-twenties marriage is a part of slavery by historic standards. Free people have a natural right to marry 16 to 20 year olds by ancient and medieval standards.
As for early modern societies, most did not marry in the 20s. Only MWEMP societies did. Aka Hajnal line societies, a minority of the globe. See my article here.
In England, the legal age at which a girl could be betrothed by her father would be 7, at which she could consent to sex, 10 (or 12, it varied), and at which she could choose her own spouse, 20 (I think). Even on the topic of 12 year olds consenting, though, the law was somewhat schizophrenic. For example, this is the 1275 Westminster statute against rape:
21 is when she could choose her own spouse, if she wasn’t already married. 12 was the age at which she could marry with her father’s approval.
But, anyway, the point is that this age is mostly unrelated to the age at which marriage becomes legal.
No, it was about the same, and that makes sense, because shotgun marriage was a response to alleged rapes, especially when the girl defended her “rapist.”
That would either be much younger (if you include betrothal by your father)
Betrothal was just a contract for future marriage. It could be broken pretty easily. The girl would not have any contact with the groom until she was of marriageable age. Often the groom would be another child, or a young teenager, and the betrothal would be arranged by two fathers.
Of course, a marriage that occurs at age 7 would not be consummated until much later.
A marriage would not have occurred at age 7. A betrothal could have. The marriage would be in the girl’s teens and would be consummated immediately upon the fulfillment of the betrothal contract.
it makes little sense to base a specific age on this.
I agree, I think it should be replaced with scientific standards using physical and mental examination plus rape charges. So if someone has sex with a 10 year old, it should be pretty easy to show a jury at trial that it occurred because the girl’s hymen was broken, that the girl was not old enough for it physically because she has not finished puberty, which usually happens at 15, give or take a year or so depending on the person, and that the girl could not consent mentally because her WAIS scores put her in the retard bracket in the adult range (eg the average 10 year old has an IQ of 70 or so on the adult scale). This is the kind of evidence I expect to convict someone of a capital crime like rape, not just defining a number out of thin air, which becomes extremely unrealistic as we have seen in the last 100 years when they make it 18 and prosecute a 21 year old for sleeping with a 17 year old, for example.