Social Evolution Forum
FIND sef:
The Tall-but-Poor ‘Anomaly’

In my previous blog on War Before Civilization I used the paintings by a French artist Jacques Le Moyne de Morgues. Although there was some controversy on the authenticity of his depictions of life in Southeastern North America in sixteenth century, I believe that at this point the verdict is that they are a good ethnographic source. In any case, in my previous blog I used Le Moyne’s paintings as illustrations (and for their shock value, I admit). The main argument about the intensity of warfare in Native American societies rests on very solid archaeological data, and doesn’t depend on Le Moyne’s veracity.

Le Moyne was part of an ill-fated French expedition to Florida in 1564–5. The expedition was a disaster, first running afoul of hostile Indians, and eventually being wiped out by fellow Europeans (the Spanish). Le Moyne was lucky to survive and get back to Europe. We are also lucky that he survived because he brought with him a wealth of ethnographic observations.

But as I was looking through Le Moyne’s illustrations, I couldn’t help noting, first subconsciously, and then in a more aware manner, just how tall the Native Americans were compared to the Europeans:

Sign up for our newsletters

I wish to receive updates from:



I am a big fan of using anthropometric data to measure people’s well-being. Average population height is a very sensitive indicator of changing standard of living (see for example my blog on this topic).

So this observation immediately reminded me of an article by Richard Steckel and Joseph Prince, Tallest in the World: Native Americans of the Great Plains in the Nineteenth Century. These authors analyzed the data collected in the nineteenth century by the students of the famous German-American anthropologist Franz Boas. Steckel and Prince found that the Plains Indians were astoundingly tall. Here’s their data for the average heights of men:

Tribe Height, cm
Assiniboin 169.6
Blackfeet 172.0
Crow 173.6
Sioux 172.8
Arapaho 174.3
Cheyenne 176.7
Kiowa 170.4
Comanche 168.0
Total sample 172.6

Some, like the Cheyenne, were as tall as the Americans today. All, with the exception of the Comanche, were as tall as the contemporary white Americans, and most were taller than them. This was not too difficult, because during the second half of the nineteenth century the heights of native-born white Americans were declining. The average height of American males born in 1850 was 171 cm, and 40 years later it fell down to 169 cm.

I am not even speaking of the sixteenth century French. As you can see in the last Le Moyne’s painting, the European leader of the French expedition is a full head shorter than the Indian chief:


One of the founders of anthropometrics, John Komlos, refers to the observation that the Plains Indians were the tallest in the world in the nineteenth century as the “Tall-but-Poor Anomaly.” But there is no anomaly here. It just shows that GDP per capita is a very poor measure of well-being. For example, between 1850 and 1890 GDP per capita, in inflation-adjusted dollars, increased by 130 percent, but the height of Americans fell by 2 cm. It’s not that Americans were becoming shorter as they were becoming richer. It was the top 1 percent who were becoming richer, while the 99 percent were becoming shorter.


So the Indians were nominally poor, but they lived in a way that only rich people can afford today. They exercised (riding them horses was a pretty good exercise!), ate grass-fed bison, supplemented by roots and berries (that’s paleo diet!), breathed fresh air, and drank uncontaminated water.

Today this kind of living is only within the reach of the very wealthy. Such as Mariel Hemingway and her boyfriend, the stuntman and actor Bobby Williams. Here’s Williams on the benefits of eating buffalo meat: “The stronger the animal, it’s going to be energetically better for you,”


OK, I don’t want to overdo how wonderful the life of Plains Indians was. Yes, they ate a healthy Paleo diet, and breathed fresh air. But they were also at a high risk of being killed in one of the innumerable clashes with hostile neighbors. And quite a few were captured and slowly tortured to death.


Join the discussion


  1. More specifically, potential height is specified pretty tightly by genetics, but actual height depends very heavily on muscle and bone growth, which in turn depend on lots of protein, calcium, and vitamin D (necessary to metabolize the calcium). Native Americans, Scandinavians, and some eastern central Asians usually had lots of all that in the old days (dairy products from a semi-nomadic lifestyle or at least lots of meat in the Asia case). On the other hand, the Native Americans of central and south Mexico were pretty short–living on maize and vegetables–not much body-building stuff.

    • Peter Turchin says:

      Gene, on individual basis height has very high heritability, but on a population basis it’s a very sensitive indicator of changing environment, primarily the balance of nutrition versus disease environment (plus overwork)

      • tward says:

        Mr. Turchin,
        I don’t think Gene is contradicting that. I liked your post but I really do think it missed Gene’s valid point, namely that individuals have a genetic height potential. This point is relevant because the individuals are part of a gene pool. So we would expect median height potential to be different between semi-isolated groups. Thus, *within* a group over time we can use height as a proxy for well-being, but absolute differences in height between groups may tell us little, though relative changes between groups can still be revealing.

        I’m not a geneticist, but I would expect potential height to be selected based on *prior* conditions during long periods of time. I would speculate that a warrior culture – one in which nearly every citizen (male) participated in war – without advanced weapons but with regular access to good nutrition would select for high heights. A culture with sporadic access to good nutrition without need for regular hand to hand combat (perhaps disease had been the real killer) may select small. That difference in genetic *potential* could explain much of the difference between two groups for many generations after the initial conditions disappear.

        Further, once half of the population has reached genetic potential the median height signal is cancelled. Also, I’m not sure whether height changes in linearly in proportion to changes in well-being across all heights. In particular, there may be rapid increases from a base of very poor nutrition, but less height response to additional factors.

        Finally, in particular, we would want to know about immigration patterns in the U.S. during the period when heights changed. And influx of smaller potential height groups would reduce measured heights.

  2. I expect your points about diet and environment are the main reasons for the height difference, but I also have to wonder about the impact of the physical violence. Was being tall a bigger contributor to survival in warfare in pre-gunpowder days? On the other hand, maybe being a bit shorter is not so bad when your main danger on the battlefield is not an axe or spear, but a bullet? The Europeans and Native Americans had been subject to somewhat different battlefield selection pressures for the previous few generations.

    Not sure if there’s any way to test this possibility; are there examples of peoples with equally good (or bad) diet and environment, with height data pre- and post-gunpowder?

    • Igor Demić says:

      My 2 cents: width is more of a disadvantage than height when firearms are in question. If a warrior is tall and thin he would stand a better chance against a musquet or a winchester rifle than a bit shorter and more corpulent man (short and thin combination is the best of course, as the tall and plump is the most disadvantageous, but those are probably the extremes on the bell curve we wouldn’t expect to change the outcome of war). But this scenario is actually quite unrealistic even in some (epi)gene-culture selectionist model: firearms are not the weapon that killed the most native Americans. My point is: selective pressure of firearms probably wasn’t so strong that it would affect the distribution of height in male native Americans. Being that social evolution works in mysterious ways I wouldn’t be surprised if I got proven utterly wrong.

    • Peter Turchin says:

      Folks, to tell the truth I think your discussion on heights versus success in battle is completely beside the point 🙂

  3. I wouldn’t expect de Morgues to be scientifically accurate. One of his pictures shows an alligator hunt; the animal is about twice as big as the largest alligator on record 🙂

    • O.Voron says:

      That might be, since “Le Moyne’s original drawings were reportedly destroyed in the Spanish attack on Fort Caroline” and he reproduced them from memory years later, according to Wiki.
      But why make a Frenchman so much smaller than an Indian every time? My guess is that was how he remembered it. After all, his botanical pictures are very detailed and botanically correct.

    • Peter Turchin says:

      That’s true! It’s a veritable dragon. Here’s the link for those who want to enjoy it:

      link to

  4. O.Voron says:

    Mariel Hemingway and Bobby Williams look fantastic! They are on paleo too, you say?

    • Peter Turchin says:

      According to the NYT article, they live on raw eggs, avocado, salad, and bison meet. Sound pretty Paleo to me

    • tward says:

      Perhaps they are on paleo. But a few things are nearly certain: Top notch professional photographers with perfect lighting and angles. Probable access to excellent personal fitness coaching. Careers that demand physical attractiveness. High probability of professional photo touch up.

  5. radek says:

    It’s not that Americans were becoming shorter as they were becoming richer.

    I’m not sure if I’d lay the decrease in height 1850-1890 at the feet of falling incomes of the non-rich. I’m not so sure that the incomes of the non-rich did fall during this time although this is a subject of some debate; the disagreement is more between “stagnated” and “grew” rather than “grew” and “fell”. Outside the South, nonfarm wages most likely increased (maybe even substantially), although farm wages probably stagnated. After the Reconstruction the South entered a period of stagnation/falling wages but it’s doubtful that this was substantial enough to make the America-wide change negative overall (it may also be the case that the decreases in height were concentrated/more pronounced in the South)

    If overall avg income increased by 130% then for the avg income of the bottom 99% to just stagnate, assuming an income share for the 1% of 15% (purposefully high, Peter Lindert estimates something around 10% in 1850), the increase in the 1% income would have to been something like 1430%. Even splitting it up top 10% and bottom 90% gives very high numbers

    Taking some data from Sam Williamson that I have laying around on the hard disk, between 1850 and 1880 the nominal wage of unskilled workers increased by about 63%. For production workers, the increase was about 117%. Both in nominal terms, so assuming about a 1% annual inflation rate, subtract off about 50% off that. This suggests that while income inequality did increase substantially over this period, the lower classes’ incomes did also grow.

    In a somewhat recent paper in European Economic Review (Nov 2012), Pamela Sharpe looks at the short stature of children in 19th century Britain. There’s some striking data points in the paper. The avg height of adolescent convicts arriving in Australia was almost a foot less than the avg height of same age adolescents who were born to the convicts who arrived a generation earlier. The avg height of working class and even middle class early teens in mid 19th century London was at the 3% percentile of those at the end of the 20th century. 3% not 30% so 97% of early teens today are taller than the average child in 19th century. (I’m relating this bit from memory so some inaccuracies may be present).

    A good chunk of the period Sharpe covers may have involved stagnating or even falling wages. Still, I’d argue that it’s possible for heights to go down while people are getting richer and look for the cause of the height decline referenced above somewhere else. That somewhere else being cities, i.e. urbanization, which also accelerated during this time.

    Late 19th century was still a Malthusian world although that was changing. Cities had higher mortality rates due to overcrowding and the resulting disease environment. But by Malthusian logic this meant wages were higher in cities than in country side. This resulted in a continual flow of people from the farm to the city. So mortality was higher but people were richer. And the prevalence of diseases in the cities stunted childhood growth (as in Sharpe) and led to lower heights.

    You can push this a bit further. Assume nutrition is related to income. But nutrition affects infant mortality to a greater extent than it affects height. Height is also affected by childhood disease (rickets, dysentery, bone disease – as well as widespread opium abuse according to Sharpe). By moving from the farm to the city an individual winds up with more surviving (due to higher income) but less healthy and shorter (due to diseases) children. Heights go down even as incomes go up.

    The above story is the flip-side of the so called “quantity-quality tradeoff” argument which seeks to explain the fertility decline that occurred towards the end of 19th/beginning of 20th century in the developed countries. The decrease in number of children/increase in their human capital has received widespread attention in the literature, but if one takes the theory seriously, then there must have been a period when the tradeoff worked in the opposite direction. And that would be the period being discussed.

    • Peter Turchin says:

      Radek, thanks very much for this thoughtful reply, which my off-hand remark doesn’t deserve. However, I have developed these issues at some length in my book draft on the structural-demographic analysis of American history. The stuff is dispersed throughout the book, but the most relevant chapters are 3 and 9. Can I induce you to take a look? I would be very interested in your critique.

      • Anthony says:

        Another big change between 1850 and 1890 in the U.S. was immigration. Many (most?) immigrants to the U.S. were fairly poor, and the likelihood that any given immigrant would have suffered a period of malnutrition severe enough to stunt his (or her) growth would be pretty high. A bunch of 160cm men moving in where the average was 170cm could drop that average significantly. Especially if, in the first generation, their kids were likely to have at least one year poor enough to stunt their growth.

        • Peter Turchin says:

          The data are only for native-born Americans

          • tward says:

            Mr. Turchin,

            Even if the data were only for native-born Americans it would reflect changes in the population’s origin mix. If we accept that genetic potential differs between groups, then a change in mix will impact the native-born numbers. Ideally we’d like to see heights of a static mix of origin. Given the probable changes in mix, we can legitimately ask about the statistical significance of the 2cm change.

  6. radek says:

    Peter, I’d be glad to. I’ll send you an email.

  7. Brel says:

    It’s questionable as to how representative those height measurements of the Plains Indians are. I don’t necessarily doubt their accuracy, but they were taken at a time when much of the Plains population had been decimated by disease that traveled along the trade routes from European-dominated areas of the continent. Those who survived would have been unusually robust specimens, and would have had more food and other resources per person than before the Columbian Exchange. Similarly, the Black Death led to higher wages among surviving Europeans. Also, the arrival of the horse among Plains tribes made it much easier to kill buffalo and obtain supplies than it had been before. Gregory Clark discusses these phenomena in A Farewell to Alms.

    That said, relative egalitarianism among the Indians surely led to better health overall. The situation of short, relatively unhealthy Europeans encountering robust natives has parallels to short, relatively unhealthy Romans encountering robust Germanic tribesmen 1500 years before.

  8. Peter Turchin says:

    I agree with these points. Yet, skeletal data suggest that American Indians were quite tall before the European contact. More generally, hunter-gatherers are typically taller than farmers, especially early farmers who have just taken agriculture.

  9. jesus alfaro says:

    If hunter-gatherers are typically taller than farmers, why is it interesting that american natives were taller than europeans? It does not seem to be the case with south american natives which were all farmers

  10. tward says:

    Another twist: The Civil War. “The Union Army had a set minimum height of 5’3″ since they believed smaller men would be unable to stand the rigors of the march, and a preferred maximum height of 6’3″ since larger men would be more easily fatigued. These guidelines were not always followed, since there are records of men who didn’t fall within them.”
    Source: link to

    It sure would be nice to know the height distribution of men killed in the Civil War.