Human longevity in prehistory

A new article in PNAS, authored by Oskar Burger, Annette Baudisch and James Vaupel, is gathering some attention to the issue of the evolutionary history of human demography and longevity. The article itself covers a lot of old ground (excuse the pun) in terms of patterns of human demography, relying heavily on data gathered by Michael Gurven and Hillard Kaplan for a 2007 article on the topic (Gurven & Kaplan, 2007). The new study, as reported by Science News, finds that only recently have humans begun to regularly live a long time:

“It’s amazing what clean water and a bit of extra food gets you,” says Berger, of the Max Planck Institute for Demographic Research in Rostock, Germany.

A 30-year-old hunter-gatherer has the same probability of death as a Japanese person today who is 72 years old, the study found. At 15, a hunter-gatherer has a 1.3 percent probability of dying in the next year; Swedes hit those odds at age 69.

As I said, many of the findings in the current study can be seen in the longer 2007 piece by Gurven and Kaplan. In that article, the researchers attempted to assemble the most complete record available of high-quality demographic studies of contemporary hunter-gatherer populations, making distinctions between the degree of acculturation or horticulturation present within those groups.

The discussion reminds me of an important classic article by the archaeologist Martin Wobst on “the tyranny of the ethnographic record” (Wobst, 1978). There is a tremendous appeal to bringing empirical data to the discussion of a complex theoretical issue like human demography. And yet, simply because the data are empirical does not mean they are appropriate. I have discussed this before, but one of the great challenges in paleoanthropology is the identification of appropriate comparisons across time and space. The contemporary demography of populations occupying different regions in different circumstances is tremendously value to our understanding of these issues on both a theoretical and practical level. However, simply because they are valuable does not mean they represent an appropriate analog for thinking about events across evolutionary time.

A few things that I think can be said about evolutionary demographic changes in humans:

1) There have been genetic changes in recent humans that play a role in reduced mortality, increased longevity

The expansion in human population size over the last 15,000 year and expansion into and modification of a host of different environments has created a strong selective environment on humans. Surely, some of this selection has found its way to genetic characters associated with fertility, mortality and longevity. Some of these novel genetic elements are probably shared throughout our species, some are likely present at differential frequencies within and between populations.

2) Whatever genetic changes have occurred, most of the variation in contemporary demographic parameters results from cultural/technological differences, not genetic differences

Yes, there have been genetic changes in humans of demographic significance. But the reality is that the biggest drivers of the key demographic parameters are cultural. How populations relate to their environment, how they relate to each other, and how they conceptualize reproduction and survival–while certainly not abiological–are dominated by culturally produced, encoded, and inherited constructs. This is evident even in the PNAS paper:

The variation between the highest and lowest mortality populations is remarkably large (Fig. 1B). The lowest age-specific death rates are enjoyed by the current populations of countries such as Japan and Sweden. The worst-case mortality for humans is approximated by 19th century slaves on Trinidad, who suffered death rates at all ages that were higher than those for hunter- gatherers.

This large variation in mortality is the product, primarily, of cultural forces, not biological differences. I would point out that Figure 9 from the Gurven & Kaplan paper (pasted below) also points to this issue:

3) The implication of the above two statements for the evolutionary history of humans is not entirely obvious

If such tremendous variation exists across contemporary populations, it is easy to accept that such variation may have existed in the past. Not that a large number of prehistoric populations would have had massively reduced adult mortality or extended longevity, but that the occupation of varied environments, the differential application of technology to ecological issues pertaining to fertility/mortality, and different cultural practices would likely have supported considerable variation in prehistoric demographic patterns. The general U-shaped mortality pattern associated with humans I think is likely valid throughout prehistory, but the location of modal longevity expectations, the distribution of mortality risk both early in life and later in life, and varying fertility curves could have created a lot of viable and differing demographic profiles.

Wanting to stay true to my pledge to share more math, consider the following age-specific survivorship function l(x) using a Siler mortality hazard model:

This functions gives you the survivorship likelihood at age x given an initial infant mortality rate (a1), age-independent mortality (a2), initial adult mortality rate (a3), rate of mortality decline (b1), and the rate of mortality increase (b3). The nature of these mathematical relationships means that even small shifts in the underlying parameters can lead very quickly to large changes in the observed demographic patterns. We know that living populations do have different parameter inputs into this model, which much of the difference coming from differing cultural/environmental components.

Going back to Wobst’s complaints about the tyranny of the ethnographic record, I think it is wonderful to have the kind of compiled data put together by Gurven and Kaplan to empirically address and model demographic issues. I doubt, however, that contemporary hunter-gatherer groups adequately sample the range of demographic variation present in our evolutionary past. As just one example, even for the more un-acculturated hunter-gatherer groups identified by Gurven & Kaplan, they still exist in a world of several billion people, which by itself changes the global evolutionary playing field for infectious diseases and parasites. A minimally acculturated hunter-gatherer group is not a Late Pleistocene population of hunter-gatherers. As such, the challenge of extrapolating deep evolutionary patterns based on that assumption, which I think is at least part of what the PNAS paper’s authors are doing, is problematic. Mind you, it is an endeavor that I like and would like to see more critical discussion of, as I think the demographic realities and variation in our evolutionary past are a critical component of the human evolutionary story.

*****

1. Burger, O., A. Baudisch, et al. (2012). “Human mortality improvement in evolutionary context.” Proceedings of the National Academy of Sciences. doi: 10.1073/pnas.1215627109

2. Gurven, M. and H. Kaplan (2007). “Longevity among Hunter-Gatherers: A Cross-Cultural Examination.” Population and Development Review 33(2): 321-365. E-ISSN: 17284457

3. Wobst, H. M. (1978). “The Archaeo-Ethnology of Hunter-Gatherers or the Tyranny of the Ethnographic Record in Archaeology.” American Antiquity 43(2): 303-309. ISSN: 00027316

About Adam Van Arsdale

I am biological anthropologist with a specialization in paleoanthropology. My research focuses on the pattern of evolutionary change in humans over the past two million years, with an emphasis on the early evolution and dispersal of our genus, Homo. My work spans a number of areas including comparative anatomy, genetics and demography.
This entry was posted in Demography, Ethnography, Evolution and tagged , , , . Bookmark the permalink.