A dung debate?

I never imagined I would end up being a specialist on the subject of archaeological poop, but there you go. Coprolites, animal dung, palaeofaeces you name it. So I was very intrigued by an article published in the Proceedings of the National Academy of Sciences last week, claiming evidence for the earliest use of manure by European farmers 8000 years ago.

Firstly let's consider the different methods that we can use to identify dung in the archaeological record. These can be divided into direct and indirect methods. The former include definitive evidence for the presence of dung, that is actual dung pellets, or lipid biomarkers which are found exclusively in dung (i.e. faecal sterols and bile acids from the gut). The latter include evidence which is suggestive of dung, for example weed seed assemblages. Although indirect indicators are often the only thing available in archaeology, in my opinion one cannot make definitive statements based on indirect evidence. The best we can do is suggest likely possibilities, or hypotheses that can be further tested. The PNAS article uses a new indirect indicator - nitrogen isotope values of crops - to suggest the addition of dung to the plants (specifically an increase in N 15 in manured crops - Bogaard et al. 2007, Fraser et al. 2011). I am not an isotope expert, so cannot comment too much on the reliability of this new method, though the comparison of 'inferred' values of forrage to show enriched levels in cereals seems problematic. Herbivore forrage values were only estimated for 6 of the 13 sites studied, using a very indirect method of taking herbivore bone collagen values and subtracting an estimated 4‰ average to compensate for trophic shift. I would much prefer to see real values measured on actual forrage plants.

Let's also consider archaeological evidence for the presence of agriculture - crop remains (preferably domestic forms), farming tools, crop processing tools. To say that agriculture is present at a site, we would like to see a number of these indicators. Multiple lines of evidence are better than one. Likewise, in order to say manuring was taking place, we would like to see definitive evidence for the presence of dung, and also evidence that it was being deliberately applied to crops. There is no mention in the PNAS article of any other evidence that would support the hypothesis of manuring at the sites investigated. Although it is a reasonable interpretation to make, I would prefer to see more direct evidence of manuring practise rather than relying on something as indirect as nitrogen isotopes in crops. Even in the case of sterol biomarkers, which provide direct evidence of the presence of dung, or even when actual dung pellets are observed, it seems to me a bit of a leap to say that this indicates deliberate manuring of crops. Another possibility is that animals simply wandered around in the same area as crops were being grown, adding their dung deposits without intention.

How can we say for sure that people were deliberately and consistently manuring these crops? This would require 'manuring markers' to be present in the majority of crops over time. Unfortunately it is impossible to determine this, as the istope analysis was conducted on bulk samples i.e. multiple individual grains (this number varied from 2 to 'at least 10' according to the paper) were combined together for analysis, giving an average signal for all the crops from a single stratigraphic unit. How do we know what the variation in manure presence was across these individual plants and across different periods? And are 2-10 individual grains really representative of agricultural practises? These are very low numbers of samples which always makes me a little uneasy, even if it is sometimes the case in archaeology that we just have to work with what we have rather than having statistically significant numbers of samples.

The article mentions significant diversity in N values within some of the sites, interpreted as local variation in manuring rates, comparable to modern 'traditional' farming regimes. Couldn't this just be local variation in the presence of manure? Just some random thoughts - I actually really like the article, and the hypothesis is a good one. I'm just not keen on making definitive statements when there are other possible explanations. Hopefully further testing of the hypotheses will provide other lines of evidence in support!

Comments welcome as always.