Origins 7 – The Ghost of the Sea

Max Planck (1858-1947) was a trailblazer of quantum theory, that confusing branch of physics which describes the inherent unpredictability of the very small. It’s kind of appropriate, then, that researchers at his namesake foundations – two of the Max Planck Institutes in Germany – are shedding uncertainty in the tiny world of molecular archaeology.

It’s a response to what was, arguably, last year’s blockbuster story in agricultural archaeology: the identification of 8,000-year-old wheat DNA in sediments at Bouldnor Cliff, at the bottom of the Solent. Farming Unearthed covered the story here. The shocker, as I wrote then, is that this makes the wheat Mesolithic – 2,000 years older than the first evidence for farming in Neolithic Britain. It was suggested on this basis that early farmers from further east were bringing their wheat products to Britain long before cultivation reached these shores – an idea which would have paradigm-quivering implications for agricultural prehistory in north-western Europe.

It sounded so improbable, and there were sceptical responses from the start, but the argument also seemed very strong. The team seemed to have done everything possible to avoid contamination, prove the integrity of their samples, and demonstrate that they’ve really got wheat DNA.

But, it’s now being argued, they may have missed something. Bouldnor, we have a problem.

Geneticists from the afore-mentioned Max Planck Institutes (for Development Biology and Evolutionary Anthropology) have applied rigorous tests to investigate the antiquity of the sedimentary ancient DNA, or ‘sedaDNA’.  I confess that I can’t follow all the arguments and statistics of Weiß et al. 2015. But the crux of it seems to be about deterioration. Ancient DNA, even if it’s good enough to identify, still doesn’t look like new DNA. It doesn’t even look like recent historic DNA. Essentially, ancient DNA has a distinctive pattern of reasonable wear and tear, fraying at the edges: damage, whereby cytosine starts to look like thymine at the ends of the molecules, known as ‘C-to-T substitution’.

The teams conducted thorough statistical comparisons between the Bouldnor Cliff sedaDNA and other libraries of well-dated ancient and modern DNA, and conclude that the pattern of C-to-T substitution at Bouldnor doesn’t look anywhere near 8,000 years old. Apparently, it looks even more modern than material collected 85-170 years ago (Weiß et al. 2015, p.3).

Now, like I say, most of the details of this research are beyond my ken. Way over my head. But the essential clash of results between the two research projects is deeply confusing. The original research appeared to rule out all conceivable means of contamination: the sediments were collected in stainless steel boxes straight out of sealed Mesolithic contexts deep under the Solent, and analysed in pristine labs that hadn’t handled wheat before. Not only that, but the type of wheat – where identifiable – is described as ‘domesticated einkorn wheat… Near Eastern wheat, as distinct from distantly related species in northern Europe and Britain’ (Gaffney et al. 2015, pp.25-6). Einkorn, one of the earliest domesticated cereals, hasn’t been popular in Britain for thousands of years. It was old news when Boudicca was a lass. It still exists today, to be sure, but why – of all possible modern contaminants – should einkorn turn up here?


Einkorn – a golden oldie (source)

Einkorn, forsooth!

Any cultivated plant DNA in a Mesolithic context would seem implausible, but a Near Eastern strain matching einkorn is arguably the least implausible of the bunch. At least there’s a hypothesis – albeit a rather earth-shattering one – to explain how it might have got there. But if it’s actually less than 200 years old, how do we explain that?

Its presence is surprising if it’s ancient, but also pretty mind-boggling if it’s modern. Am I missing something? Could the state of being buried under the seabed somehow inhibit molecular spoilage?

One thing’s for certain: we haven’t yet heard the last of this mysterious story.



You’ll be pleased to know that the new research is in an open-access article:

Weiß, C.L., Dannemann, M., Prüfer, K. & Burbano, H.A. (2015). “Contesting the presence of wheat in the British Isles 8,000 years ago by assessing ancient DNA authenticity from low-coverage data”, eLife 4,
The original work is discussed by:
Gaffney, V., Garwood, P. & Momber, G. “How sedimentary DNA brought wheat to mesolithic Bouldnor Cliff”, British Archaeology no.142 (May/June 2015), pp.22-27
And the first, seminal article can now be accessed with free registration to Science:
Smith, O., Momber, G. et al. (2015). “Sedimentary DNA from a submerged site reveals wheat in the British Isles 8000 years ago”, Science 347, pp.998-1001.

A Grand Challenge: to blog or not to blog?

This post is a belated contribution to ‘The Grand Challenges for Archaeology’, a blogging carnival organized by archaeologist-blogger Doug Rocks-Macqueen.


It’s about half-eight in the morning, and I’ll soon have to be at work. But I’m snatching a moment to blog about blogging in archaeology.

These days I’m paid to do admin, not archaeology. But the past still has a habit of catching up with me. After all, archaeology is a vocation. One cannot simply ignore it when the money stops.

It is also addictively stimulating – cognitive caffeine.

Academic pursuits such as this come at a price, of course. The problem is that, given half a chance, the past can devour the present. Research and dissemination are intellectually demanding, time-hungry activities. Time marches on. The internet can speed the process along, but also bogs the researcher down in endless links, distractions, tweets, comments, bile, graphics, videos… and blogs.

Blogs. Now there’s a mixed blessing for writer and reader alike. The so-called blogosphere: global coffee house and repository for corridor conversations, challenging insights, hot new research, sloppy spelling, pub politics, fact and factoid, enticing hyperlink roads that all lead – at some point – to Wikipedia. The busy, burgeoning blogosphere: free of charge but time-expensive, open-source but unaccountable. Anyone with a mind to blog can jump in and do it. I took the plunge, in the first instance, after being favourably impressed by the blog of a friend from the natural sciences.

A question has dogged my typing ever since: is it worth the time and effort to stay in step with the digital zeitgeist? Here we come to the crux: a grand (or navel-gazing) challenge for the online archaeologist: is my blog a real vehicle for research and education, or merely ‘go-faster stripes’? The fact that many enthusiasts and academics alike maintain blogs does not necessarily, by itself, legitimize the medium. Popularity has some force, in the cyber-democratic milieu of the 21st century. To put it bluntly, if I stop getting hits I’ll have precious little excuse to carry on – if nobody’s reading it, I might as well just keep a private research diary.

But the existence of readers still doesn’t quite answer the question: what is it for? Why, if it all, is it useful for those readers? I started blogging largely as an experiment because it seemed like ‘the thing to do’, and – flatteringly – people still read it. Colleagues at conferences comment favourably on it, often adding ‘I’d like to blog too, if I had the time.’ Indeed. I hardly ever blog these days, not because I don’t want to do so but because, let’s face it, it takes ages to write something worth reading, and there’s no obvious, tangible return from this major time-investment.

Or is there? When I get around to doing it, composing a new post exercises my writing and thinking muscles – valuable simply on a personal level, like physical exercise. As for the end result, it’s still evolving: if the blog is to have any lasting value as a form of archaeological literature, the blogger should presumably look at which posts attract the most interest, and give the readership more of that (in my case, the unexpected answer is lazy beds, corn-dryers, and the origins of farming). But this can be balanced by innovation and experimentation – testing the waters in the inherently fluid environment of the internet. This, perhaps, is the unique strength of the blog, the best way to exploit its  incremental, transient/permanent, paradoxical character. It is, by its very nature, a constant work in progress – it can evolve, it can mature, in step with the blogger’s own career and changing interests, and the wider development of the discipline.

After three and a half years, Farming Unearthed is still finding its feet. Maybe it always will be.

Maybe it always should be.


“Get off my land!” in the Bronze Age

I shouldn’t have been so surprised.

Way back in 1784, when I started this blog, I commented on how farming seems to permeate archaeology as a discipline – from the Neolithic onwards, it’s a crucial thread in so much of human history. It’s hard to ignore.

But equally, it’s often a bit of a poor relation. Farming isn’t always a fashionable topic to research in some regions and periods. Burials, for instance, frequently attract much more attention, I suppose because of the immediate (and mysterious) human interest of a skeleton and its funerary trappings.

So on this occasion, I was surprised. I was reading the latest edition of British Archaeology, with a fascinating article on evidence for mummification in Bronze Age Britain:

Booth, T., Chamberlain, A. & Parker Pearson, M. ‘The Mummies of Bronze Age Britain’, British Archaeology 145 (Nov/Dec 2015), pp.18-23.

This is a fascinating piece of research – mummification is convincingly argued from ingenious scientific methods to have been widespread in Bronze Age Britain – and truly unexpected, certainly to someone like me with limited knowledge of the Bronze Age.

Most unexpected of all, however, were the speculative interpretations of Bronze Age mummification, and this is where agriculture comes in (p.23):

“After 1500 BC… large, planned (‘co-axial’) field systems were laid out. Mummies could have been important for legitimising access to ancestral rights and securing claims on land and property. In a world without written documents or legal title, what better way of producing tangible and incontrovertible proof of ancestry, ownership or descent!”

An intriguing theory indeed. What clearer way of articulating the time-honoured cry of “Gerroff my land!” than with the corpse of a desiccated patriarch?


Barnstorming in Roman Wiltshire

So there they were, poised to knock up a retail distribution centre outside Chippenham, when up pops a Roman barn. Turns out there’s a nationally important Romano-British rural settlement on the site, and Historic England (formerly English Heritage) has scheduled it accordingly – so the development has stopped.

This news nugget caught my eye in the latest edition of British Archaeology magazine, but it also made BBC local news, given the contest between developer and state over what should become of the site (not to mention local petitions).

What’s particularly interesting for Farming Unearthed is the discovery of an unexpectedly large barn, in a part of the country where Roman agrarian developments aren’t that well understood. British Archaeology reports: “Excavation of 12… postholes found second century AD pottery, suggesting that what might otherwise have been interpreted as a large medieval barn was in fact Roman.” As if that weren’t enough, Historic England itself has been excavating at another Romano-British farmstead in Wiltshire, of a rather different character, but nonetheless with “the corner of a large barn.” Are we on the verge of a scholarly overhaul of Wiltshire’s Roman farms?

Broadening the picture, this seems an appropriate moment to flag up a new (and growing) online resource courtesy of the Roman Rural Settlement project. It’s a mammoth collaborative effort from the University of Reading and Cotswold Archaeology, funded by the Leverhulme Trust and Historic England. The project team has been busy harvesting vast swathes of excavation data, much of it previously unpublished, regarding rural settlements in Roman Britain. There’s an important agricultural strand led by Dr Martyn Allen, expert in zooarchaeology but also interested in things like corn-dryers (a hobby-horse of mine, you may recall). Anyway, the team is magnanimously uploading their data into a massive barn – I mean, database – on the Archaeological Data Service. Find it here, and have a browse. It’s a fantastic resource and a remarkable achievement.


Sorting the sheep from… the other sheep

Continuing this blog’s impromptu ‘DNA season’, I’d like to flag up another recent article. Now this is ingenious stuff. It made me chuckle with admiration, in an I-wish-I’d-thought-of-that sort of way.

First, the problem: historic livestock. How can we know what breeds of livestock were kept in the past? How far back can we trace so-called ‘heritage breeds’? For more recent periods of history, there might be realistic paintings or detailed descriptions that allow us some insight, but such sources will get much less reliable the further back we go. By the medieval period at least, we’d be skating on very thin ice. What about the evidence of animal bones? True, that’s extremely useful. Skeletal remains can allow a specialist to calculate the original size of animals (withers heights, etc.). This can allow the identification of selective breeding –  but only if the farmers were selecting for size. Potentially more information could be acquired by extracting aDNA (ancient DNA) from the animal bones, but these are usually far too degraded to allow any such thing. In most cases, after all, they’ve been butchered, gnawed, and thrown away.

That’s where parchments come in. Documents have been written on animal-hide parchment for centuries and, although they’re not guaranteed to survive forever, it’s safe to say that they won’t usually be gnawed or butchered. Indeed, because they often have legal value, a great number have been carefully curated over the centuries. In Britain alone, it’s estimated that more than a million parchments may survive from the last millennium. And now it’s been demonstrated that they can be used as a good source of livestock aDNA. Researchers from Dublin and York have done a pilot project, successfully getting verifiable aDNA from tiny samples of sheepskin parchment from the 17th and 18th centuries.

The Past Horizons website has a good summary.

Not only that, but the original research is published in an open access article, free for anyone to read and download:

Teasdale, M.D., van Doorn, N.L., Fiddyment, S., Webb, C.C., O’Connor, T., Hofreiter, M., Collins, M.J., & Bradley, D.G. (2015). ‘Paging through history: parchment as a reservoir of ancient DNA for next generation sequencing’, Philosophical Transactions of the Royal Society B, volume 370.

From the two parchments that they analysed, they suggest a speculative interpretation: that one represents an ‘unimproved’ sheep breed (i.e. closer to its natural progenitor) and the other a selectively ‘improved’ breed: potentially, we have here a major new tool to explore how livestock changed during the era of agricultural improvement in England – and of course other periods and places too, if the material survives.

Oh, and if curators don’t mind samples being trimmed from their parchments.


No spelt, please, we’re Saxon

My guest post from the “Not Just Dormice” blog

Not Just Dormice - Food for Thought

Guest blogger Mark McKerracher considers the fate of foodstuffs after Roman rule…


The boats safely beached, four Germanic feet touched the sands of old Britannia. The heavily moustached faces of Hengest and Horsa looked out over these strange new shores, littered with imperial detritus. A limp, decaying sack lay at Horsa’s feet.

‘Spelt flakes,’ it read, ‘naturally rich in Romanitas.’

‘Pah,’ muttered Horsa. ‘Foreign muck.’


Behind this stirring vignette of the birth of England lies a real archaeological conundrum: why didn’t the Anglo-Saxons eat more spelt? The facts are simply stated. When charred crop deposits are excavated from Romano-British settlements, the wheat component is practically always dominated by one type: spelt wheat. Yet from the 5th century AD onwards, in deposits from Anglo-Saxon settlements, bread wheat takes its place, dominating the wheats to the near-total exclusion of spelt – as it has done, pretty much, ever since…

View original post 1,005 more words

Origins 6 – Fashionably early?

After a long pause, the blog is back – and it’s a blog of surprises. These particular surprises reached my attention through both British Archaeology magazine and the online news pages of Science. For those of you lucky enough to have full-blown access to Science, the reference is this:

Smith, O., Momber, G. et al. (2015). “Sedimentary DNA from a submerged site reveals wheat in the British Isles 8000 years ago”, Science 347, pp.998-1001.

Admirable clarity in that title, snappily stating the two things that had me so surprised – or “gobsmacked”, as they said in the Mesolithic.

Surprise 1 – sedimentary DNA

Bouldnor Cliff is a Mesolithic hunter-gatherer site with a twist: it’s underwater, 11m deep in the Solent, off the coast of the Isle of Wight. Although under threat of disturbance from modern shipping and fishing activity, it’s still remarkable for its surviving organic remains, including Britain’s oldest known piece of string (over 8,000 years old). The inundation of the site, when the Solent grew to separate the Isle of Wight from the mainland, has helped to preserve remarkable remains such as this, which have been investigated by maritime archaeologists since 1999: see what the folk at the Maritime Archaeology Trust have to say. But in this latest study, it’s not artefacts that are causing a stir: it’s aDNA in the sediment. Yes, they’re sequencing aDNA (= “ancient DNA”) left in the sediment by plants that have long since decayed. I had no idea that this was possible. Received wisdom (among non-geneticists, at least) seems to be that extracting and sequencing aDNA from any biological remains is fiendish enough, let alone from traces left in the sediment. But they’ve done it, and convincingly too, with unexpected results.

Surprise 2 – wheat?!

Yes, wheat. Besides the trees and wild grasses that you’d expect in a Mesolithic landscape, the team also identified domesticated wheat DNA in the sediment.The deposits were sealed on the seabed, and modern contamination has been ruled out: wheat seems, quite genuinely, to have been present at this hunter-gatherer site, circa 6,000 BC. And, to be blunt, it shouldn’t be there.

That is to say, it jars with the accepted narrative. Cast your mind – or your browser – back to a previous post in this series: “Britain goes nutty”.  This reviewed an article by archaeobotanists Stevens & Fuller, who looked at the dated evidence for crop remains in prehistoric Britain. In accord with the conventional narrative, they found that domesticated crop remains appear in the archaeological record pretty soon after 4,000 BC. So the Bouldnor wheat DNA is 2,000 years older than we might expect. It predates the introduction of agriculture to the British Isles in the Neolithic period. So what’s going on here?

Well, Gaffney and co. stress that this new evidence doesn’t mean that we should redate the British Neolithic. Just as one swallow doesn’t make a summer, one site with wheat DNA in the sediment doesn’t make an agricultural dawn. Things would be different if we had securely-dated pollen – a stronger indicator of local cultivation – but the DNA could easily represent something imported from faraway lands. That in itself would be of huge significance, implying far-reaching networks linking British Mesolithic hunter-gatherers with new-fangled Neolithic farmers further east, long before farming communities became established in northwestern Europe.

With evidence like this, the cultural picture of this period could become a lot more complex, along with our understanding of the dynamics of early agriculture and its spread across Eurasia. Heady stuff, but for now I’ll leave you with the image of those Bouldnor hunter-gatherers, peering curiously at an object newly arrived on their shores: “Congratulations! You are the lucky winner of a Luxury Wheat Gift Hamper from the Fertile Crescent Company. Try before you sow! Call now to reserve your free Neolithic Package while stocks last…”



Gaffney, V., Garwood, P. & Momber, G. ‘How sedimentary DNA brought wheat to mesolithic Bouldnor Cliff’, British Archaeology no.142 (May/June 2015), pp.22-27.