We’re going on a bere hunt

Cereal farming emerged around 11 or 12 thousand years ago, in the so-called Fertile Crescent of the Middle East. Over the next 5 or 6 thousands years, arable farming ploughed its way westwards and northwards across mainland Europe, over to the British Isles, up into Scotland, even unto Orkney, Shetland, and the somewhat less fertile crescent of the Western Isles.

Needless to say, what works in Syria doesn’t necessarily work in Stornoway. Superficially, the idea of carrying agriculture more than 2,500 miles from the Levant to the north-western fringes of Eurasia might not seem very promising. But if Jurassic Park has taught us anything, it’s that “life, er, finds a way.” And so it did, genetic variability allowing for the adaptation of cereal species to environmental conditions in all kinds of locale: a combination of natural and artificial selection creates a genetic paintbox to colour-match your crops to your crags.

Such locally attuned geno-/phenotypes are known as landraces. While much modern farming relies rather on the precision-breeding of “lab races”, a few landraces cling on, including a six-row hulled barley in the Scottish islands known as bere barley.

Bere is thought, but not proven, to have been introduced by Norse settlers around the eighth century. But how can we test this archaeologically? And, by extension, how can we distinguish and trace other landraces in the archaeological record, as a measure of genetic variability among historic and prehistoric crops? Even when written accounts survive, it can be very difficult to judge whether they’re talking about what we think they’re talking about. The bere of historic documents isn’t necessarily the bere that we know today.

In theory, you could look for genetic information in charred cereal grains from archaeological deposits. In practice, not only is this costly, but it’s also pretty difficult to get good material: DNA doesn’t respond well to charring, and it’s all too easy to contaminate samples.

So Sheffield University’s Dr Michael Wallace has taken a pioneering new approach to the study of bere, called geometric modern morphometrics (GMM). The study has recently been published, and it’s open access, so you can read all about it here:

 Wallace, M., Bonhomme, V., Russell, J. et al. (2018). “Searching for the Origins of Bere Barley: a Geometric Morphometric Approach to Cereal Landrace Recognition in Archaeology”, Journal of Archaeological Method and Theory.

So, when it comes to cereal grain, what is geometric morphometrics? Well, think phrenology – except, instead of a pseudoscientific characterisation of criminal brain-shape, this is a statistically-rigorous characterisation of grain-shape from a series of microscopic mugshots, aided by Dr Vincent Bonhomme’s acclaimed Momocs package in R. So, on second thoughts, don’t think phrenology. Sorry.

Anyway, this GMM method has certainly come up with the goods. Bere barley proves to have morphologically distinct grains. And, critically, these are distinct measurements that you couldn’t observe just by staring obsessively at the grains: it needs precise measurement and multivariate statistical rigour. Neither are they characteristics that are disastrously distorted by charring: the grains do shrivel up a bit when charred, but in a predictable way that doesn’t obscure the landrace distinctions, which is crucial for archaeobotanists intending to apply this approach:

“GMM can characterise the grain morphology of some specific groups of landraces, providing a novel archaeological method for exploring the spread and use of landraces.”

Wallace et al. 2018 p.13

So this methodology is definitely one to watch: being able to investigate past crop types, not just at species-level but even more precisely, could give us fantastic new insights into the history of arable farming. And meanwhile, Scottish archaeobotany is well equipped for a bere hunt.