i happened to find the book with all the italian inbreeding data“Consanguinity, inbreeding, and genetic drift in Italy” by cavalli-sforza, et. al. — on google books.

and i found something real interesting in it!

remember how i was wondering what the inbreeding numbers for italy were before 1910? well, cavalli-sforza, et. al., have data stretching back to 1780 (cool!) and this is what they look like (the solid line):

before i go any further, this chart doesn’t represent the actual numbers (no label on the y-axis, see?). instead, it’s a model to (as the caption reads): “explain the peculiar trend of consanguinity in italy.” that peculiar trend was low consanguinity at the start of the period (1780); then rising, at first gradually, but then with increasing speed until hitting a peak around the end of wwi; and then a falling off again.

the authors write [pg. 240]:

“Figure 9.13 is an attempt at a simple, semiquantitative explanation of the whole process. The curves assumed to generate the phenomenon are two logistic curves, an ascending one that describes the trend of increasing population and a descending one for the breakdown of isolates. In Italy this phenomenon occurred later than in northern Europe, and was assumed to reach the 50% point shortly after 1925. The two logistics have been given slopes that differ as 1:3 to approximate the ratio of the slopes of the initial increase and the later decrease in consanguinity. No attempt was made at fitting the real data, which are too irregular to be accurately fitted. The solid curve is the product of the two logistics and is expected to describe approxi….”

and then, gosh durnit, i can’t read any more ’cause, of course, there’s only a preview on google books and i don’t have a hard-copy of the book. (*argh!*)

i think i understand the model, tho. i’ve mentioned before somewhere that there’s a theory out there that high population numbers actually lead to higher consanguinity rates (in a population given to inbreeding) simply because there are more cousins available to marry. that’s why cavalli-sforza, et. al., have got a “population size” line on their chart — increasing consanguinity rates coincide with increasing population rates.

on the other hand, consanguinity rates are often high in remote places, ’cause who else is there available to marry except for your cousins? starting in the late-1800s in italy, tho, the population became more mobile (better transport, moving to cities for work, etc.) — and so consanguinity rates go down because the “isolates” are broken down. that’s the other line on the model-chart.

so, that’s their theory for why there’s a roller-coastery bump in the middle of italian consanguinity rates for the last couple of centuries. and it’s a good one.

i wonder, tho, if the increase in inbreeding in italy over the nineteenth century could be connected to il risorgimento — the italian unification process.

take a look at these maps:

in the late-eighteenth century, “italy” amounts to a bunch of states, the number of which decreases progressively over time as they are joined into one state. right up to the end of wwi.

maybe the italian populace was reacting to this push for unity, which came from a bunch of intellectual nationalists (like elsewhere in europe around the same time), by choosing to inbreed more. perhaps this was another response to a stressful situation like after the world wars. a way to close ranks. i’m sure a lot of italians didn’t actually favor a unified italy, which is still evident today (or in recent history) in organizations like the lega nord and the sicilian independence movement.

then, however, the greater mobility of the people thing happened. (maybe.)

anyway — just some thoughts. in any case, this is a good illustration that you can’t assume anything from one generation’s inbreeding rates — the previous or the next generation’s could be higher or lower.

edit: boilerplate and boilerplate 2.0

(note: comments do not require an email. cool!)