le parallélo- gramme

i promised myself that i wouldn’t post any more about france until i’d finished reading robb’s The Discovery of France (and some other materials on that nation), but i’m too impatient, so here goes.

here from Fréquence et répartition des mariages consanguins en France is a map of consanguineous marriages among catholics in france between 1926 and 1945 (this map made a previous appearance on the blog in this post):

france - consanguineous marriages - roman catholics - 1926-1945

last week i posted a couple of maps showing how the distributions of these historic cousin marriage rates in france and the various regions in which different crops are grown are largely congruent — historically there was (prolly still is) a greater avoidance of cousin marriage in the wheat growing areas of the country versus the grass covered areas of the pastoralists (and even those areas inhabited by olive and grape growers!). this is undoubtedly a legacy of medieval manorialism since, as mitterauer has convincingly argued, manorialism was all tied up with wheat/grain growing AND the institution also helped to promote the avoidance of cousin marriage.

normally i don’t like to use a population’s modern cousin marriage rates to try to guess what their past rates might’ve been — it’s dangerous and one shouldn’t make assumptions. mating patterns change. however, in this case, based upon what i know about the history of medieval france, especially the franks and their adoption of christianity, and the patterns of manorialism in northern europe, i think it’s probably safe to assume that the regional differences in the cousin marriage rates on the map above probably do reflect cirumstances on the ground in france for the last few hundred years — perhaps even one thousand. note that i’m not saying that the cousin marriage rates were the same in france in the past as in the early twentieth century, just that these same regional differences probably existed — i.e. that those areas with lower cousin marriage rates in the 1920s-1940s probably had lower rates than the rest of the country for a very long time, etc. going forward, this will be my working assumption for france, but please keep in mind that it is an assumption. could be wrong. if i come across any data contradicting — or supporting! — this assumption, i’ll let you know!

something robb says early on in The Discovery of France [pg. 12] caught my eye:

“Tales of isolation and ignorance tend to be associated with spectacular exceptions and with regions that lie beyond what some French historians have termed ‘an enlarged Paris Basin’, which accounts for more than one-third of the country — an enormous parallelogram [sic] stretching from Lille to Clermont-Ferrard and from Lyon to Le Mans, where ‘men, ideas and merchandise’, all identifiably and self-consciously French, had supposedly been pumping through the system since the Ancien Regime.”

if we map that…

le parallélogramme

…sacrebleu! that’s not far off…

france - consanguineous marriages - roman catholics - 1926-1945 + le parallélogramme

and here overlaid onto todd’s family systems (as best i could =/ )…

todd - traditional family systems of europe + le parallélogramme

my guess is that robb’s paralleogram — the “enlarged paris basin” — represents the most manoralized, most oubred region of france. (i guess, too, that it prolly can be extended a bit to the east). this is “core” france, and the peripheral regions like brittany (where the le pen family is from) and the massif central area further south have experienced more inbreeding (or less outbreeding, depending on how you want to look at it) and so those subpopulations will be more clannish than the population originating from inside le parallélogramme. in other words, brittany and the massif central areas should be thought of as france’s scottish highlands or english borderlands.

indeed, a report from transparency international seems to indicate that, looking away from paris which has no doubt attracted all sorts, there is more corruption in peripheral france than in core france. (i know that it’s also difficult to say much about southern coastal france since there are so many immigrants there.) [source]:

france - regional corruption

also, i previously found, using the world values survey data, that the population in the area officially categorized as “paris east” is the most civic in france. part of paris east falls within le parallélogramme, but much of it lies further to the east, perhaps indicating that robb’s parallelogram should also be extended further to the east. the cousin marriage rates certainly suggest that. we shall see.

and, as we’ve already seen, there are some pretty clannish sounding populations in peripheral france in places such as the auvergne and the greater roquecezière metropolitan region. (~_^) still, Further Research is RequiredTM.

btw, the ancestors of french canadians came mostly from regions bordering on or outside of le parallélogramme and acadians (cajuns) originated entirely from outside this “core” france (see here).

vive la france! (^_^)

previously: meanwhile, in france… and mating patterns in france and topography (and history) and crops and cousin marriage in france and civicness in france by region and the auvergnat pashtuns and the battle of roquecezière and big summary post on the hajnal line and what’s up with french canadians?

(note: comments do not require an email. l’hexagone.)



  1. Greater urban centers in the past was sacerdotal centers too. This explain why “core” people tend to be more inbred. Probably, muslims who live in urban centers tend to less inbred than “saddam house’s niches”.

    Islam have rules about marriage patterns??

    France have greater immigration since half of XIX or earlier. Today, 9% of french population have italian ancestry, for example. France is the most americanized european country, paradoxically, because the greater european diversity similarity. A lot of polish ( Chopin), russians, italians, spaniards, portugueses, englishmen, greeks, etc, since XIX. Other note, cultural note. Catholicism is most universal than protestantism. This explain why Only 4% of white americans have detectable non- White bloods but at least 5% of portuguese, a european people. This is a latin “cultural” trait. Then, french culture, non-core, tend to value language more than ethnicity.


  2. That’s not a parallellogram. If you squint a bit, it’s almost a trapezoid, but really just an undefined quadrilateral. Otherwise an interesting article


  3. Good grief! All the quantitative stuff here, and someone confuses a quadrilateral with a parallelogram.


  4. @fossegrimen and bob – i know. but i can’t very well rewrite graham robb’s book! =P i guess i can add a [sic] to the post, tho….


  5. all makes sense


    very cool. i can’t get over the Polish split on that map.


  6. In the 26-45 map there seems to be an ethnic/historical signal.
    On the one hand, the division separating the Oc languages (from pig latin) apart from the Oil languages (from Church Latin.)
    Also although Brittany speaks an Insular Celtic language, and has migration age Alanic roots (see Bachrach), the majority of the population is descended from the Latin-speaking Romanised Gauls who fled to the peninsula.
    As such, a higher frequency of consanguineous marriage hints at stronger Roman heritage.


  7. @jayman: when? If you drew the same map for, say, 1956, Scotland would be predominately blue. Do you think there has been massive ethnic change there in that time?


  8. ”interesting! thanks! (^_^)”

    For nothing, ;)

    About map of political parties, look to Romania, wow, older territory??? Romanians and hungarian-romanians, as water and oil.


  9. Very interesting! Here are two more maps possible of interest: Poverty and Violent Crime in 2010. On the left is poverty rate (darker is poorer); on the right is violent crime rate (darker is more criminal). Looking at the Massif Central made me think a bit of Appalachia, a mountainous place where poverty is high but crime is pretty low. Brittany, on the other hand, looks to be relatively low-poverty and low-crime.

    And yes, it’s very hard to find immigrant % since they don’t take ethnic censuses. Some possible proxies are the sickle-cell testing of newborns map (they don’t test babies of two indigenous white French parents), or the number of mosques in a given locale (map on left).

    Also, ‘Discovery of France’ sounds very good, I will have to check it out.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s