The latest publication from the Petraglia & Korisettar palaeolithic research team, working in South India, was published last week in PNAS—Population increase and environmental deterioration correspond with microlithic innovations in South Asia ca. 35,000 years ago. This study dovetails nicely with recent simulation modelling effort of some UCL colleagues (Adam Powell, Mark Thomas, and Stephen Shennan, director of the Institute of Archaeology), published in Science last month. (The appearance of these two papers close together is entirely coincidental). This simulations suggest that behavioural innovations, such as modern human behaviour, should be expected to emerge when circumstances are right (especially due to demographic factors of higher interactoing populations), and to be maintained only if demographic circumstances permit (and therefore disappear if populations decline).
These two studies together represent importance counters of an orthodoxy that sees ‘modern’ behaviour as emerging once, and therefore being a great invention when hard cognitive architecture came into place, perhaps even driven by a key genetic mutation for intelligence. Such is the orthodoxy implied by classic textbooks on human evolution, such as by Richard Klein (at least as was used when I was a student) or the recent reviews by Paul Mellars (e.g. his Science paper of 2006). In this view modern humans, heir cognitive abilities and the behavioural application of those abilities emerged once (in Africa) and spread out of Africa (once) to bring intelligent modern everywhere else (perhaps at sometime between 60,000 and 40,000—depending on whether one prefers to emphasize the earliest possible dates for Australia or the Upper Palaeolithic transition in Europe). The evidence from South Asia shows clearly that toolkits were middle Palaeolithic (and in this sense not ‘modern’) from >75,000 to 35,000 years ago, and yet genetics suggests that these would have been anatomically modern humans (and they must have moved through South Asia earlier than their arrival in Australia sometime between 60,000 and 50,000). The bottom line is there is a good evidential case to be made, congruent with the modelling of Powell et al., that the cognitive architecture for modern behaviour was around but the innovations that we regard as ‘modern’ emerged when social and environmental circumstances demanded.
There is a parallel here to where thinking on agricultural origins is moving. There has long been an orthodoxy that agriculture was a great and rare invention, and that agriculture came to most regions by the migration of farmers from a few centres of the influence of a good idea. In the more extreme cases, only 3 centres of origin (Mexico, Near East and China) are accepted. But the evidence of archaeobotany—where it is available—combined with the biography of wild progenitors, and where avail able the genetics of crops/livestock, suggest that there was many more centres in which societies converged on agriculture—the behavioural changes towards manipulation of the environment in favour of the reproduction of a few food species (domestication, the genetic sense, was an unintended consequence of these behavioural shifts, when the genetics of the species allowed). In my most recent attempt to sift through the combined data, I concluded that there is strong case for 24 separate instances of agricultural origins globally—although as many of these are poorly documented and geographically close together one might reduce this to 13, as per the map below from Purugganan & Fuller:
The point is that agriculture, like modern human behaviour, was not a one time great invention, but the product of social and environmental circumstances to which human groups with the same cognitive potential responded in parallel ways. The question in both cases is: what were the common denominators of those circumstances?
6 comments:
Hello Dorian,
Great blog! As a geneticist, I am somewhat surprised that the issue of parallel inventions of agriculture is still an issue. The concept of centers of origin/domestication has been around for a century, thanks to Vavilov, Harlan, et al. Furthermore, biochemical and molecular data also show distinct, and likely, independent domestication in different geographical areas, not among only among different crops, but also within a crop gene pool.
For example, I demonstrated some time ago (1986) that common bean has two independent domestications in Mesoamerica and the Andes. Similar observations have been made in the America for cotton, amaranth, and Capsicum peppers. Even within the Mesoamerican center, there are different domestication areas for maize (Balsas), beans (Lerma-Santiago), and cotton (Yucatan). For the latter, however, one could question the notion of independence of some of the domestications. I must be missing something here, because for some time agriculture has been considered an example of multiple, independent inventions.
My thoughts exactly. I can't think of anyone in the field who accepts only 3 centres of origin (though that depends a bit on what you mean by "centre").
Dear Paul & Luigi,
Thanks for your comments. My sense is that most of the genetics community has shifted towards seeing multiple areas of independent origin, but within archaeology there is still a penchant for reducing historical complexities to as few origins as possible-- often focusing on where more archaeological research has taken place rather than considering other forms of evidence (biogeography, genetics) that should encourage us to take up research in the less-explored or unexplored areas. The problem has been a self-fulfilling research circle where we look for agricultural origins where we know they were (and have found evidence for them previously-- especially the Near East or the presumed middle Yangtze), and we find evidence for them. The incentive is reduced for looking into domestications elsewhere, and theorizing of process and cause in general comes to be based on single cases. I do think work in the New World (eastern North America, tropical South America) is clear and exemplary in indicating the non-centered or multi-centered nature of this process, but this has not perhaps been picked up by many working on Old World origins. Most of us archaeologists tend to be rather regionally-focused and parochial...
I wonder if New World genetic work really demonstrates independent domestication between different crops.
What degree of certainty do we have, for instance, about maize (Balsas) vs. Mesoamerican beans (Lerma-Santiago) as unrelated domestications from an archaeological perspective? Both origins have insufficient certainty to exclude the possibility of spatial (and temporal) overlap, I think.
The genetic analysis methods have known limitations (phylogenetic trees, for instance). "Single point" origins are interesting heuristically, but most likely domestication didn't occur in a single valley. Pre-domestication cultivation is possible, too.
On the other hand, the archaeological debate hasn't been concluded yet (e.g. highland vs. lowland origins for maize) and has not even started regarding beans.
As I alluded to in my initial comment, one can have doubts about the independence of domestications within a geographic area such as Mesoamerica.
However, to come back to the original point, and if you take the Americas as a whole as an example, it is more difficult to argue for lack of independence among eastern North America (sunflower, marsh elder, ...), Mesoamerica (maize, beans, Capsicum, ...), and South America (potato, cassava,...). One can repeat this exercise for other parts of the world.
I thought the "latter" referred to cotton in Yucatan... We don't disagree, I guess.
Post a Comment