Monday 25 April 2016

What does the medieval family look like?


Rick Stump's recent article about village and family demographics, and in particular the graph on there, really got me thinking about what the medieval village and family looked like.

The graph shows population before the first demographic transition which is yet another reason that, despite enjoying medieval game worlds, I really don't want to live there. In simple terms, people are dying as fast as they're being born, mainly from the old standbys - disease, famine, war and famine caused by war.

From the starting points in Rick's article I thought I'd try and visualise what the population of a village really might look like but then, deciding that particular rabbit hole was a bit too deep, thought I'd instead try to get a picture of just one family. I'm sure there's been all sorts of scholarly study on this but only a few interesting data points came to light from my initial hunting. This gives me enough to hang some ideas on to while seeming to me at least to be plausible, however wrong those ideas may in fact be!
  • Högberg et al, Maternal Deaths in Medieval Sweden: An Osteological and Life Table Analysis has all sorts of interesting information, but of most use is the life expectancy graph.
  • There's a widely quoted figure of 14.4 maternal deaths for every 1,000 births in 15th century Florence. This initially seemed a bit low to me given the gap in life expectancy for men and women in Högberg's paper, until you remember Rick's figure of 6 children per family and hence a 9% chance of dying in childbirth for the average mother...
  • The life of Lucrezia Borgia is interesting anyway, but especially so in this context, giving the stark reminder that a limiting factor in the size of a family wasn't the parents deciding not to have more children, but one of the parents dying! Again from the Högberg's paper it can be seen that the average life expectancy for a woman was around 33, for a man around 40 (although interestingly if they can reach age 50, life expectancy is actually slightly higher for women - presumably due to the risk from childbirth being removed).
From there I had enough (probably dodgy) data points to get started from -
  1. Given that 33 year life expectancy and an average family size of 6 children, in hand-waving terms I called that a 1/3 chance of a child being born per year from when the mother is age 16 until age 40
  2. Rick furnished figures for death in infancy and up to mid-teens
  3. The Högberg's paper allows me to generate life expectancy for those who make it to the advanced age of 15!
With some random number generation and the WFRP name generator I came up with the following for a family on a farm somewhere in Reikland in 2504 -

Klemens runs a farm where his recently deceased wife's grandfather still lives, her father having died just a few years before. They have four surviving young children (although the random number generator says Gustaf will be dead within the year). Brother-in-law Friebald and his wife Esmeralda are an invaluable help as is his sister-in-law Carlotta, although she'll probably be married off soon. Rigo (aged 13) and Eldred (aged 11) are both expected to do a full adult's share of the work as well.

I found generating the above an oddly poignant process, as the random number generator killed off children that various stages of infancy, or mothers with families, and that was just for entries on a page with labels like "Son A3-2"! But it did bring home those bare statistics about mortality rates in a way that will hopefully give my in-game villages a bit more soul in the future. 

There's are a few things I might give some more thought to when I get the chance, such as making couples more varied in age, especially if I ever tackle the "village" problem - the (fairly obvious) thought being that the smaller the community the more it comes to who's available, or simply not related to the suitor, rather than any thoughts of romance. Passing adventurers who show any sign of success and who have all of their limbs and some of their teeth suddenly move from "we don't like their sort" to "eligible"...!

The other thing to think about is to have the deaths cluster more - although certainly people would die from isolated incidents and accidents, such as the "minor wound becomes infected" theme that I've started to notice from certain historical and dark fantasy authors, at least some would be clustered from disease, famine or violence. It'd be interesting to know what the balance should be between the isolated and the clustered - maybe 3 of the 4 deaths across 2502 / 3 / 4 could actually have been caused by the same event? 

Meanwhile I go back to being thankful for epidemiology and all the rest of our modern wonders.