Seminar CLXXXV: checking what the genes mean

The week after the seminar just reported, I was back down at the Institute of Historical Research’s Earlier Middle Ages Seminar to hear no less a notable than Professor Patrick Geary of UCLA address us with the title “Preliminary Reflections on Genomic Evidence and Medieval Migration History”. Perhaps because I was only just out of Oxford at this point, I had all kinds of misgivings about this talk: other notable historians who’ve lately got into scientific methods have perhaps let enthusiasm outrun rigour, and the only previous time I’d seen Professor Geary talk I’d taken away what seems to have been a quite misleading impression. In fact, this talk was earnest, humble even, critical and very interesting.

Map of distribution of Y-haplogroups in modern European DNA samples

Map of distribution of Y-haplogroups in modern European DNA samples, source unclear

Professor Geary was introducing to us a seriously international collaborative project that seems to have no name, quite hard to do these days, but which is described on his web-pages as “Tracing Longobard Migration through DNA Analysis”. You can probably see immediately why my alarm bells were ringing: a lot of fashionable work has been done with DNA lately that hasn’t thought terribly hard about what the actual meaning of its data might be. Professor Geary took us through the immediate problems: if you are using the DNA of modern populations, then you are looking at the sum of their total genetic background, including not just the episode of change in which you as historian might be interested but everything before and since, including the Black Death, the Wars of Religion, Industrial Revolution, various kinds of diaspora, and ultimately every human population change since the Second Ice Age, so it too often becomes a matter of proving what we already wanted to believe rather than showing anything new.1 On the other hand, historical DNA is far harder to come by, rarely adequately preserved, usually only recoverable in either its mitochondrial components (which descend in the female line) or its Y-chromosomal components (which do so in the male line) but not both (or so he said) and still has the same back-history problem even if subsequent accretion can be eliminated. And of course, both of these approaches tell you only what the (successful) components of someone’s biological descent were, not what they thought they were or how they behaved. DNA is not, now that we’re out of the nineteenth century, much of a determinant of identity. So what on earth can we hope to show with it?

Supposedly 'Lombard' funerary goods from a burial at Hamburg-Marmstorf

Supposedly ‘Lombard’ funerary goods from a burial at Hamburg-Marmstorf. Photo by James Seakley [GFDL or CC-BY-SA-3.0], via Wikimedia Commons

Well, the answer is of course in comparison. We can, with adequate funerary archæology, hope to see change in the DNA record, and can get some idea of the size, gender and composition of the groups in movement in the so-called Age of Migrations. Checking these findings against texts, material culture and stable isotope analysis can then also give us some means of establishing whether culture or population change in any one correlates with change in any of the others, which is really quite important to do since so much of what is wrong with current work on migrations in any one of these genres of evidence is the assumption that it must do so.2 Already, at what was then an early stage of this project, Professor Geary’s colleagues sifting through data from recent excavations were getting results showing mixed populations in movement, ‘Lombard’ cemeteries that did not include just Lombards and so on, though there was not enough asked here about how the checking category was constituted, I think. It was easy to miss such things when the following example was of a cemetery in Thuringia where they had women buried with material culture kits usually held to signify Thuringian origins and ones betokening Saxon identification and found from the isotopic analysis that it was the Saxon ones who were local! Given that most of us would perhaps cynically have expected no correlations at all, or in the case of some perhaps hoped for positive ones, negative ones was not what was expected by anyone! So some synthesis of this project’s results should prove eye-opening for us all, but will hopefully also allow some actually founded generalisations about what, in fact, an early medieval migration might have been like to be in the middle of.

Artistic depiction of the Vandal army arriving in Carthage

Depiction of the Vandal army arriving in Carthage, apparently what the Deutsche Post still think that looked like

There were, as you might imagine, lots of questions, and this not least because the actual presentation was relatively short. Some of the questions seemed to be aimed to reassure the questioner that this work would not be bringing back the spectres of nineteenth-century racialism that Professor Geary has previously done so much to dispel, but he assured us that the wrong questions of that era would continue to go unanswered.3 Others were keen to be assured that this work would not, ultimately, be able to disprove migrations, those others having their book sales to think of after all.4 The most interesting questions however came from Dr Mark Thomas, responsible for some of the DNA work that Professor Geary had attacked, and who was clearly coming from a very different place to the rest of us that made me briefly sad I was no longer so close to scientists as I had once been, so as to debate it with them: he was out to argue that historians’ tendency to make models more complex ineluctably made them subjective and therefore biased, and that the only hope for such work is to generate simple models that can thus hope to escape the biases inherent in, for example, choosing ‘ethnic’ markers.

Fig. 3 from Thomas et al., "Evidence for an apartheid-like social structure in early Anglo-Saxon England"

I would feel better about his criticisms of historians did Thomas’s own work not include things like this graph, fig. 3 from his and others’ “Evidence for an apartheid-like social structure” cited below

This boiled down into the clearest two-cultures confrontation I’ve ever observed, Professor Geary wishing to abjure Occam’s Razor to describe a process we can see must have been incredibly varied and Dr Thomas obdurate that if we historians tried to find that variation in the scientific data all we would wind up doing was repeating the process of choosing the patterns we like best, because the results would be too diffuse to clearly prove anything. Either of them could have convinced me separately: put together, it made me suddenly wonder whether we weren’t here exactly at the point where what we can genuinely know starts to dissolve, and amid the general hope that we would know a lot more for Professor Geary and his colleagues’ project I thus took away a nagging reminder that in real terms we actually ‘know’ almost nothing about the past, which implies not least that convincing people with racist agendas that they’re wrong is going to be tricky. Still: since even hard science is finding out these days that data does not convince, I’m not sure that leaves us so much further apart after all…4

1. One example of such work used by Professor Geary was Mark G. Thomas, Michael P. H. Stumpf & Heinrich Härke, “Evidence for an apartheid-like social structure in Anglo-Saxon England” in Proceedings of the Royal Society B Vol. 273 no. 1601 (London 2006), pp. 2651-2657, DOI: 10.1098/rspb.2006.3627, a paper also criticised by Magistra from a mathematician’s point of view here, and this one was indubitably the most sharp critique just because, unknown to Professor Geary of course, its lead author was in the audience. Also singled out, however, were Cristian Capelli, Nicola Redhead, Julia K. Abernethy, Fiona Gratrix, James F. Wilson, Torolf Moen, Tor Hervig, Martin Richards, Michael P. H. Stumpf, Peter A. Underhill, Paul Bradshaw, Alom Shaha, Mark G. Thomas, Neal Bradman & David B. Goldstein, “A Y Chromosome Census of the British Isles” in Current Biology Vol. 13 (Amsterdam 2003), pp. 979-984, DOI: 10.1016/S0960-9822(03)00373-7 and most of all Peter Ralph & Graham Coop, “The Geography of Recent Genetic Ancestry across Europe” in Public Library of Science: Biology Vol. 11 (San Francisco 2013), e1001555, DOI: 10.1371/journal.pbio.1001555, this last for spotting Hunnic migration using modern state boundaries, uncontrolled dataset and strictly out-of-date scholarship with which to interpret, according at least to Professor Geary: the historical cites actually seem fairly up-to-date to me, but this is not a paper one can skim for solidity of interpretation, so I haven’t…

2. Isotopic work on other hand takes a particular joy in exploding everyone else’s ideas, e. g. Janet Montgomery, Jane A. Evans, Dominic Powlesland & Charlotte A. Roberts, “Continuity or colonization in Anglo-Saxon England? Isotope evidence for mobility, subsistence practice, and status at West Heslerton” in American Journal of Physical Anthropology Vol. 126 (Hoboken 2005), pp. 123-138, DOI: 10.1002/ajpa.20111, and it’s hard not to like it for that but it is still using one form of evidence to show a bigger condition, even if that condition is undefined free flow of movement.

3. Patrick J. Geary, The Myth of Nations: The Medieval Origins of Europe (Princeton 2003).

4. If this all leaves you wanting more, you may be pleased to know that Professor Geary has written what seems to be a short version of this paper as “Using Genetic Data to Revolutionalize Understanding of Migration History” in The Institute Letter Spring 2013 (Princeton 2013), web version here.

5 responses to “Seminar CLXXXV: checking what the genes mean

  1. Surprising if Geary said Y-chromosomes and mitochondrial DNA are mutually exclusive, since every male has both.

    • The exclusion he was asserting was in the usual possibility of extracting both types. I don’t understand myself why that should be an issue, but that would be a measure of how far we are beyond my training!

  2. You’re both right.
    They’re stored in different compartments of the cell. the Y-chromosome is found in cell nuclei (nuclear DNA, nDNA), but the mictochondrial DNA (mtDNA) are in mitochondria, i.e. compartments within the cell responsible for energy generation. DNA extraction of whole cells will free both nuclear and mitochondrial DNA from the cell interior because the initial lysis step breaks down membranes and releases DNA from interior cellular compartments, but then you have to pass it though further techniques to isolate and amplify your signal of interest. Different, specific techniques are required to isolate nDNA or mtDNA (although a recent publication claims to have been able to do both at once). Greater isolation of your target DNA gives better results because the signal is tiny and needs amplification, which is more effective if the signal is fairly ‘pure’ to begin with – otherwise you merely amplify the impurities as well.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s