Demographics [S. Worthen

I went to a talk on the history of Walthamstow tonight. The speaker said that the sewage pumps, installed at the end of the nineteenth century, reduced infant mortality (before the age of one) from 15% to 7%. The audience was impressed. Then he contextualized it: the infant mortality rate at the time in Poplar was 35-40%.

I've spent a lot of the past day thinking about demographics and how little I know about them.

geesepalace observed that, contrary to my sarcasm, it's true that life expectancy in the Middle Ages was about 25. And he's right. If you don't take infant mortality into account, the number of children dying young brings the average age of death way down. (Although not as far down as it did late nineteenth-century Poplar, I'd bet.) Equally, however, it's disingenuous and misleading to do so. It gives the impression that the average person could expect to die in their 20s, which isn't really true. Anyone who survived their first few years in the period stood a good chance of living to 50+ years.

That's the standard line I've been using for as long as I've been having discussions about medieval life expectancy, but I couldn't tell you where the figures come from. What's the research on which it's based? Are they any better than the Black Death mortality rate estimates, which vary by at least 40% between the higher and lower estimates? Our statistics are only as good as our data, our use of them, and our intentions. The data's not great, but improving with every new census record and with many archeological digs.

So: medieval demographics. What are the major sources and who are the major authors? Joel Rosenthal has published several relevant articles and Old Age in Late Medieval England. I've found Josiah Russell's Medieval Demography: Essays. I'm intending to read Peter Biller's historiography work on the subject, The Measure of Multitude.

While, I'm asking, here's a related question from

geesepalace:

On the other hand I've decided, based on no evidence whatsoever except my own retrospection, that most of the vikings and others who shared in what (I gather) was a sort of northern-european warrior death-cult were around 15 to 25. Not of course that they died out then, just that by their 30s fewer of them might have subscribed so whole-heartedly to the belief, unless of course they had risen to positions of power, where the cult would have served them well. For some reason contemporary artists who depict ravaging vikings rarely show them in their teens or early twenties. Do you know whether anyone's done much work on the age of the average viking thug?

October

Sun	Mon	Tue	Wed	Thu	Fri	Sat
1	2	3	4	5	6	7
8	9	10 1	11	12	13	14
15	16	17	18	19	20	21
22	23	24	25	26	27	28
29	30	31

Sun

Mon

Tue

Wed

Thu

Fri

Sat

Patrick Bronte commissioned a report on the parish of Haworth in about 1850, not long after the deaths of three of his children. It was pretty damning, but one statistic is haunting - over 41% of children died before their sixth year.

[link] [reply]

I'm reading Three Cups of Tea and learned that this statistic is (or was in the early 1990s) pretty much the same in rural Pakistan due to lack of good sanitation. In our western perspective, we forget that this is not just a thing of the past.

[link] [parent] [reply]

Currently, the peak of child mortality rates by the age of 5 is about 27% (Sierra Leone) from a quick look around online.

For Pakistan, this article cites about 10% child mortality (by age 5), rising to 16% for Belochistan. In the '90s, it was more like 13% - it's been coming down. (PDF from WHO with graph) In contrast, in 2002-2005, Afghanistan had a 25% child mortality rate (Article).

This are all terrible numbers, but no where near as bad as 40% would be.

(In contrast, the child mortality rate for the UK is 0.65% and for the US, 0.8%, according to this article.)

Edited Date: 2009-03-13 11:56 pm (UTC)

Of course, these are averages, and you were specifically talking about rural areas which could be bringing the averages up.

Here, we go: "In the village of Korphe, Pakistan, the child mortality rate is one out of three die before the age of one."

33% vs 40%...not desirable either way.

Those warrior death cults were common I read in something a while ago, when I was researching military bonds. The real Hussars--the Russians--expected to be dead by thirty, or you were a coward. If there wasn't a war, you dueled. Musketeers, same. It served kings to have their hot heads fighting each other if he had no convenient war draining them off.

Honestly, 25 is not life expectancy. If you got out of childhood, you'd probably live to be in your 50s or early 60s. An average of the high and low is not really expectancy, which should be calculated on a median, not a mean. IIRC.

It is technically *A* way of calculating life expectancy. It's not a GOOD or responsible way of doing so. It is, however, the way a lot of older articles on life-expectancy have calculated the numbers. I spent a while last night trawling through a random selection of JSTOR articles, and there are lots of charts of archeologists giving the average life expectancy in the 25-30 year range, based on the average age of the skeletons they're excavating. So there's plenty of precedent in the scholarly literature for modern commentators to abuse the figures this way.

Life expectancy at the moment of birth != life expectancy when you're 20.

Ah! This may be an area where the historians have demonstrated a failure to understand how these things are handled by epidemiologists - I'm a little more familiar with epidemiology than with this sort of history, to be honest - my best friend is an epidemiologist. Averaging infants with old dudes is not a way to find out the usual life expectancy!

I think it's just poor stats literacy. It *is* an average(d) life expectancy after all. But it doesn't follow a normal distribution. (And things don't always do so, which affects the way you can use that sort of data, etc.) Although I think in some cases of population mortality you'd expect a normal curve, in a case without significant environmental factors (endemic disease, cultural dangers like violence for young men). Hrm, demography has been a while for me.

In general, it's also a matter of technical expertise in fields, I think. I know that some climate scientists use text data in a way that (should) make a historian of the time blush. (Or in my case grumble a lot about the contextual variation of what people call "earthquakes" and "drought," or even "excessive rain" which is sometimes "the head honcho wants a picnic, and just look at it outside! Look at it!" in my experience--not always, of course.) And on the other side, when historians want to give some general context on What Life Was Like Back Then, then just using averages from other scholars without really looking too deeply into how that average played out; or taking Greenland ice core samples to talk about weather trends in China....

In general, I think as scholars we have to trust as sources the studies that made it through the scholarly process in different fields as well. (But verify.) The problem is the technical and unspoken knowledge that goes into those assumptions. Which is not just historians, but since the defining element of history is "stuff that happened in the past" as opposed to "stuff involving cell mechanics," I think sometimes we have more of a risk of that sort of thing.

The evidence I was looking at was from archeologists. At least one historian was more specific. I don't feel like I've looked at enough of these articles to generalize further. I don't know any epidemiologists.

My friend studies morbidity and mortality in a specific American population, and I've done some proofreading of her work... But still, averaging the highest and the lowest ages is naturally not going to give a real idea of "average", right?

Just one of the many things wrong with the mean as a measure of central tendency. I see this all the time at work: millions of pounds of budget decided on a statistical technique we were taught in primary school.

And it's not even as if the alternatives were scary college maths: the better measures are ones we were also taught in primary school, or secondary school at the latest, but for some reason business people don't like medians... and don't even try to get them to like quartiles.

My rule is, always show the distribution if you can, not just the average.

You recall incorrectly.

Life expectancy should be a mean not a median -- formerly an "expectation" in the statistical sense. It is the sum over i of i p_i where p_i is the proportion of the population likely to die at age i -- the average age of death.

Life expectancy goes up with age so if you want to calculate the life expectancy of someone aged n you sum over i from n upwards and normalise p_i by making it the proportion of people who die at age i given they lived to at least age n.

Of course calculating a forward life expectancy for a current population is much harder because you need to know how this is going to change. (The proportion of people born in 1970, say, who live to be 80 or older is very likely to be higher than the proportion of people born in 1929 who have done so).

This is certainly more specific! Now we all havethe algorithm ;)

Now all you need to do is to get a large number of skeletons from the period you are interested in and a technique for calculating their age at death.

Fortunately for me, I do intellectual history and mostly work in archives. This involves counting dead body parts and calculating their age, but only of animals.

And this is where I come in again with Theya Molleson's class! This is very much what we were looking at. She found with the Spitalfields project that estimates of age based on closure of skull sutures used previously were misleading.

Up to the mid 20s we could get a fairly good estimate of age, but after that it's all tooth wear (grit from the millstones that gets into the bread sanding down the teeth) and skull sutures, and people originally thought to be about 40 turned out to be more like 60.

Try the National History Museum; that's where Dr Molleson was based.

Oh! The Spitalfields Report is fantastic. Sorry - was searching for stats on death rates in the Viking Age, just so I could criticise the commonly accepted idea that all people in the early Middle Ages died young and ran across this thread.

I'm an archaeologist and I hate to admit it, but many archaeologists tend to stick with the average rather uncritically. Drives me nuts.

...late nineteenth-century Poplar...
One of my great great grandmothers was born in mid-nineteenth-century Poplar, as it happens, although sometime in the late 1860s she moved to Clerkenwell and started a family with (and, when his first wife died, subsequently married) my great great grandfather.

My great grandfather had seven children between about 1900 and 1920, of whom three died in childhood - one at the age of 15, in Tooting Bec Asylum from "inanition caused by imbecility", one at 13 months from pneumonia and pertussis (whooping cough), and one at the age of ten from tuberculosis (which disease also killed my great grandmother, his first wife).

Even though averaging the ages at death is the normal and traditional way for figuring average life expectancy* (after all, the higher life expectancy that we see in Western countries today is as much a lessening of infant and child mortality as it is that we are healthier and living longer), it's not really that useful when teaching the middle ages.

And then, one of the things that makes most sense to students is to talk about how it's an average. And then to show them numbers for some of the elites we know about -- people living easily into their 50s, some into their mid- 70s or longer. It's a wonderful way to remind them of how different our world is, and how many things they take for granted that just didn't exist in their grandparents' day.

After the elites, you talk about infant and child mortality, and why, then about things like hunting and farming and building accidents, and compound-complex fractures that don't heal or get infected, other infections, diseases like new strains of 'flu and the diseases for which we now have vaccines ...

And all of a sudden, the start to think not only about how things must have been, but how important things like medical care are. And in the meanwhile, they start to get that it's not that the MA were particularly barbaric (in their understanding), or even ignorant -- but that the scientific breakthroughs of vaccines and antibiotics are really phenomenal.

Sorry -- I just really love the whole way you can take something like this and make it real and useful -- and fortunately, you don't need to know all that much about demographics to talk about the average life expectancy.

*and that's the first important part of the lesson -- it's called that because it's figured on average. You could also figure mean or median, but that's not what is normally used, for whatever reason.

It's a great lesson: but how do YOU know the figures? What are your sources for them?

You know, I think that there's stuff in some of the canon -- especially in teh economic history stuff, and it makes its way into the textbooks. But I am pretty sure that there are figures in Duby's Early Growth of the European Economy and Doehaard, and want to say that most of the LA/EM archaeology-based stuff talks about it. And I think I've run into it in most of Wickham's stuff that I've read.

ETA:

chilperic is really a good person to ask -- he started as an archæology person -- or maybe some of Patrick Perrin's work?

Edited Date: 2009-03-13 03:59 pm (UTC)

I haven't looked into the subject very much, but a good starting point would be John Munro's bibliography of medieval economic history.

Again without direct evidence but...

Plenty of evidence in many societies of "scceptable" violent outlets for young men, especially younger sons. It's natural in an agricultural society where land is scarce. Younger sons have to find an heiress, get killed or make their fortune some other way.

One book on that period (and the centuries before) draws a lot of comparisons with football hooligans, even making the point that praise songs were more sophisticated in structure, but very similar in content to football chants.

This is something we looked at a lot in the classes I went to on "Human Skeletal Remains in Archaeology".

One of the things our tutor had noticed is that individuals tend to be "survivors" or "non-survivors". Non-survivors not only tended to die younger, but skeletally they seemed older than they actually were. Survivors seemed younger than they were. After her work on the remains in the Spitalfield crypt, she realised that as a result of this, many ages from cemeteries in the past were under-estimates, and people who had been considered to be 40 were in fact more likely to have been 60 or so. This threw all the demographic assumptions right out.

The peak ages of death that we found were: up to the age of 5 (a child who got past that age usually survived childhood); childbearing years for women; late teens to late 20s for men (mostly warfare but could include fighting on the local level). The best life expectancy of all was in religious houses; no childbirth for the nuns, no fighting for the monks, and better food than was usual even for nobles.

The person to check out on Google is Theya Molleson.

My godmother was a mediaeval historian. She used to get very annoyed by her students talking about the 'mortality rate' and would annotate their essays with 'mortality rate same then as now, 100%'.

I had no idea there was a lecture on Walthamstow history last night. When and where?

Here in town, with the local history society's monthly second-Thursday-of-the-month meeting. The speaker was from the Walthamstow local history group (possibly its chair/president?).

It was an overview talk, so if you're interested in local history there, you may know it all already. It's not a Roman town. It is a medieval one (evidence from over 1000 years back); the church isn't as old as it claimed to be when celebrating its 900th anniversary last year; its heyday was the 16th and 17th centuries; various architectural highlights; the coming of the railway and its effects; sanitation. That sort of thing.

For some reason contemporary artists who depict ravaging vikings rarely show them in their teens or early twenties. Do you know whether anyone's done much work on the age of the average viking thug?

I can't think of anyone who's done anything on this. The presumption is that they were mostly young, but it's only for a very few prominent individuals that there's anything approaching reliable evidence of their dates of birth.

It's probably not an unreasonable supposition that going viking is the act of men on the make: certainly there is reason to think that a typical viking life would often eventually lead to settling down and marrying, if, of course, it didn't just lead to death. Either way, it's reasonable to assume that the majority of the vikings would have been young men.

Work has been done on the early Anglo-Saxons, who seem to have had a similar social setup. Fighting men were much the same age as WWII fighter pilots: mostly under 24, and 23 was seen as getting on a bit. After that they usually wanted to settle down with their own bit of land and get married.

Interesting. It's always reassuring to know there are (endlessly) major projects left for future researchers to explore. I'll pass this on to

geesepalace.

There's an interesting grave from Orkney that relates to this. Viking-style boat-burial, strontium isotope analysis suggests the man was an immigrant from Northern Norway, stack of weapons (shield boss was heavily damaged before burial), other grave-goods, and a bunch of arrowheads in various parts of the body. The man was likely to have been in his 50s (though see the comments above re: the problems with determining age at death of older individuals).

So, not a teenage warrior.

It may not be as academic a source as you're looking for, but I just starting reading "A Time Traveller's Guide to Medieval England" by Ian Mortimer. a sort of guidebook to the 14th Century, written in an accessible style but with tons of descriptions about the details of daily life, how towns and cities were laid out, architecture, farming etc and also stuff about people and life expectancy; very nice overview and has a list of sources at the back for further exploration. I'm loving it thus far, so many small details: it might not be of use for this particular query but worth a read nonetheless - I don't *think* its one you've mentioned in the past here, but if so, apologies for unremembering :)

Edited Date: 2009-03-13 05:38 pm (UTC)

This was in the Guardian today - Vikings who _weren't_ raiders and warriors, but farmers, craftsmen, merchants etc.

http://www.guardian.co.uk/uk/2009/mar/13/2

This article is also interesting. There are fairly detailed nineteenth-century health records for American civil war soldiers and veterans, and they show that huge numbers of them suffered from chronic conditions and died young.

I often get annoyed at the idea that in medieval times (or some other benighted age) 25 years old were old people and had only few years to live. I think that is largely false/a gross exaggeration. However my sense from trying to figure out the truth is that as has been pointed out people apparently suffer from infirmity and death with greater frequency earlier even after they made it to adulthood.

I once thought about going through the Dictionary of Scientific Biography to try and get death stats for those who made it to adulthood. Although I guess ages are uncertain in many cases. I think my one attempt confirmed the sense that while people did live to be 70 or 80 plenty also died in there 40s. I'm still tempted to try it... If I do I'll share the results.

Since I was rereading this today.

[on the time-frame for climatic changes] "Few major climatic events lay within the span of generational memory, and were thus quickly forgotten in times when life expectancy everywhere was little more than thirty years."

Ignoring (1) mediated or institutional memory. And (2) what "average lifespan" means (the above conversation). (Of course, he's not really doing anything with this particular statement, but it's the statements that he throws off and doesn't do much with that bother me the most.)

I just found this article ( http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1949148 ) which makes the following interesting claim: "Even when life was much shorter than today there was a reasonable supply of older males: among hunter gatherers with an expected lifespan of 33.5 yrs, the ratio of 70 yr olds to 30 yr olds was about 0.32 and to 40 yr olds was about 0.37."

If they moved inside, we were holding welcomed by way of a man named Jesus christ. Could it be your ex? Each one besides [url=http://jswingsforsale.webs.com]Jeremy Scott Shoes[/url] learns about one, although exactly what Christ understands all of all of them. The meaning lesson trained during this striking video is usually that we are able to produce alternatives who seem to i will serve and also invest our endless everyday living within the final..
"It's really psychological and yes it increases high desires the particular children specially. It implies much to [url=http://jswings20onsale.webs.com]jeremy scott wings[/url] glorify the people who died for their liberty and those that misplaced his or her everyday life, like the 29 with my children as well as family members. their particular spirit is just what allows myself to obtain in place every single day and in my small shoes and teach in order to honor all of them,Inches said Marial..
Gripman Walter Scott [url=http://cheapjswingsshoes.webs.com]Jeremy Scott Shoes[/url] III muscled this chest-high material result in that functions the leading stop for the 99-year-old motor vehicle. This individual properly creaked our 9-ton insert on the slope. Thankfully, for this nice, the actual tires held, and not one of the passengers obtained sideswiped [url=http://jeremyscottwings20shoesforsale.webs.com]jeremy scott wings 2.0[/url] off in to traffic.

S. Worthen | Weblog

October