Featured post

Gender in the Merovingian World

Tuesday, 18 January 2011

There is Unrest in the Forest, there is Trouble with the Trees (or 'Climate change brought down the Roman Empire, scientists *don't* find')

[Research on Tree-Rings has recently grabbed the headlines as showing that 'Climate change may be responsible for the rise and fall of the Roman empire'.  This is unlikely.  The problem lies principally in the scientists trying to explain phenomena which are not now thought to have existed in the form the scientists believe.  Furthermore, there are all sorts of logical problems in arguing that climatic features explain the multifarious, divergent features of a 300-year long period of European history!]

The chance to use the opening lines of a Prog classic as the title for some writing on late antiquity comes but rarely in a career, so when it does it should be seized with both hands. 

My texts today are taken from The Daily Telegraph and from the BBC's news website:

The abstract of the research in question can be found here:

Reduced to a couple of headline-grabbing paragraphs, what the research is claimed to show is that:
"Wet and warm summers occurred during periods of Roman and medieval prosperity. Increased climate variability from 250-600 AD coincided with the demise of the western Roman empire and the turmoil of the migration period," the team reported.
"Distinct drying in the 3rd Century paralleled a period of serious crisis in the western Roman empire marked by barbarian invasion, political turmoil and economic dislocation in several provinces of Gaul"

The first thing that needs to be said is that this approach is not new.  Back in the mists of time, when I was preparing to go to university, I read Geoffrey Parker's Europe in Crisis (recently a second edition has appeared, which I have not read [http://www.amazon.co.uk/Europe-1598-1648-Blackwell-Classic-Histories/dp/0631220275] - these comments may well not apply to the new version, though a quick skim on Amazon suggests they do).  This began with a discussion of climate history and how tree-ring research suggests that the seventeenth century was a period when bad weather might have helped to produce the turmoil of the era.  At the time this fired my imagination (if I hadn't turned out to be an early medievalist I would almost certainly have specialised in the sixteenth or seventeenth century) as a real New History.  Over time, of course, my faith in this sort of environmental determinism (as in all other sorts of determinism) waned.  I ceased to be convinced by this kind of explanation, parodied by a friend and colleague as 'there was an unusually heavy frost, so the Thirty Years War broke out.'

As with my comments on DNA, my gripe is not (it cannot be) with the science itself, with the reality of the observations.  Similarly, if the aim of the researchers is to use these findings to try and spur governments on to act further and faster on climate change, then I can cut them quite a lot of slack.  But we do need to show a considerable degree of caution about their historical conclusions, at least as far as the Roman Empire is concerned.  Here are some points which you might like to bear in mind:

As with the DNA research, the scientists have not critically probed the reality of the historical phenomena they claim their research explains.  First, recent research has poured considerable doubt on the idea that the third century was an economic crisis in the sense that used to be understood, that is as a period of economic decline.  It might have been a crisis in terms of a weakening of the economic unity that had characterised the Empire in its very early phase but in large parts of the Empire (such as Britain) there is no trace of any decline at all: quite the opposite.  And the fourth century in Britain and elsewhere was an era of prosperity.

Second, the turmoil of the migrations.  Again, one needs to think much harder about what these migrations were like.  No one sane denies that people migrated in the last century of the western Empire's existence, but people have always migrated.  We do not know whether more people migrated in the fifth and sixth centuries than did in the third or fourth, or the first or second.  There were, allegedly, tribal movements on a considerable scale in the Roman Republican period too.  Large numbers of people migrated from the heart of the Empire out to the peripheries during the centuries of Roman expansion.  It is a curious view that sees the Roman invasions as a period of 'prosperity' and the probably far less dramatic migration of barbarians in the fifth century (as stated, the continuation of a set of relationships that had lasted for centuries) as one of turmoil.  More curious still that one movement might be explained by one climatic set of conditions and the other by another. 

So, again, this science is bolted on to explain a historical 'problem' whose outlines have been entirely changed as the result of serious historical enquiry.  Sometimes they have changed so far as to make it no longer a problem at all, or at least no longer a problem that can be explained in those terms.  How to explain this by analogy?  This is the best I can do.  It is 'a bit like' a situation where 'scientists' claim that the kind of wheat grown in later ninth-century England would produce a flour that, if used in cakes, could suddenly and unpredictably burn.  'Scientists show Alfred unfairly blamed for burning cakes!' then gets plastered across every news site in the land.  But we now know that the story of Alfred burning the cakes at all is a twelfth-century invention...

That is the main flaw in these sorts of explanation.  But there are myriad other issues.  The collapse of Roman civilisation in Britain was the affair of a couple of generations around 400, after a century of prosperity.  Difficult to explain that through climatic variations over the longue durĂ©e, but socio-political explanations do perfectly well.  Migration is a constant of human existence.  Barbarians had been moving into the Empire for centuries, from Caesar's time (and indeed before) right through to the fifth century, and the factors and processes behind their movement remained demonstrably analogous throughout.  Whether any more (or fewer) moved in the fifth century than the third is impossible to know, but the migrations of the third century did not spell doom for an Empire in the thoes of all sorts of other political problems, whereas those of the fifth century did become a focus for political developments that ended up in the fragmentation of the western Empire.  Climate seems incapable of explaining this difference.

Equally, the fourth century was a period of strong imperial rule and remarkable coherence for the Roman Empire, in spite of its situation in the heart of the period where climatic features are supposed to be causing crisis.  The economic revival of the West during the seventh century could be said to fit with the findings of this research but again problems soon arise.  It seems clear (to me, at the moment, at least) that this revival began in the late sixth century and has perfectly adequate socio-political explanations.  More importantly, this does not explain the fact that, although the north-west may have revived, the Mediterranean world experienced a comparative economic decline at this time.  What about the Arab invasions of the seventh century?  Is this sort of 'turmoil' not explained by climate (people used to think it was but - as far as I am aware - this is an explanation in desuetude)?  If it is explained by climatic optimum, then the opposite climatic conditions cannot in themselves explain migration, can they?  Not without considerable modification to the thesis.  And if the Arab expansion is to be explained by politics, society or ideological/religious factors, then why can the northern barbarian migrations not be explained in those terms?

And so on, and so forth.  Let us be clear that this research is valuable and interesting and that it doubtless adds important and interesting details to the picture we have of these centuries.  But we need to think much more carefully about how and why they are important and interesting.  That means asking new and better historical questions that these data can go some way towards answering, not in bolting them on to explain problems which they can't explain and which, in some cases, are no longer problems in any case.

The article concludes with the researchers in question saying "We are very interested in understanding past civilisations and making our research more dense."  More dense.  Ah, that'll be because of the 'less favourable growing conditions' then: the funding 'drought'.  That, at least, is one conclusion that all of us in British academia will currently find it hard to disagree with!


  1. I think you may be a bit quick to condemn this as sloppy history here, Guy: the researchers don't actually say their research is an explanation, merely observe a correlation and suggest that there must have been human effects which may be part of the historical epiphenomena. (The paper's only three pages of text (plus one of references, four of graphics) so it's a fairly quick study.) At that rate, then, it might be fairer to say that they may have pointed to good evidence that prolonged reduced harvest yield, which presumably would have affected tax take and the ability to feed the army and brought about settlement desertion and the contraction of agriculture, was a factor at a time when, for example, we have the third-century crisis to explain. Equally, the later prolonged mildness and steadiness of climate would presumably put more wealth into society, with who knows what results? These are complex systems.

    I also think there's a difference between barbarians from outside the wealthy zone wanting in, perhaps in time of famine, and Romans on the inside wanting out (and back) during times of prosperity, and perhaps of demographic expansion. I don't think those two phenomena are paradoxical when you reason them out; it's a matter of where the resources and their users are thickest. To reduce it simply to 'it's the environment, stupid' is obviously going too far, but I don't think that's what these researchers are doing. I also note that their historical caution may be down to the fact that Michael McCormick is a co-author. This may or may not inspire you with renewed confidence...

  2. Regardless of what the researchers intended, unfortunately the media are somewhat less balanced - they like simplicity as well as immediacy. What's new and fresh? Who needs a myriad of explanations when the quirkiest one will do?

    Still I'm impressed that this isn't a new suggestion, and that the idea has been floated before.

  3. Thanks for mounting such a spirited defence of the proposal Jon, but when you say a correlation, a correlation between what and what do you mean, exactly?

    The '3rd-century crisis' which you say we have to explain has undergone such rethinking (Witschel in JRA is probably the best overview of this) that the sort of 'crisis' that this sort of science might explain (that is, one of economic decline) is no longer believed in by most specialists. More to the point, the crisis that did take place in the third century can be more than adequately explained with reference to changes in the relationships between imperial core and periphery developing since the first century and reaching a critical point in the C3rd. It's about political and ideological coherence. Not something that would much benefit from a climatic explanationc - all of which I actually mentioned in my post.

    Settlement abandonment. Takes place at different times (and in different forms) in different places. Around 400 and in the earlier C5th in Gaul, Britain and the N. Sea coast of Germany (but not Denmark...). Not until the very end of this period of 'turmoil' in Spain and Italy. Even later in southern Gaul.

    The C3rd 'economic dislocation' of the provinces of Gaul is either a quite egregious misreading or a straight misunderstanding, or (at best) a very old-fashioned reading, of data now understood in much more sophisticated ways.

    The "turmoil of the migration period". Why is there held to be more turmoil in this period than in the earlier Roman period (Late Republican and Early Imperial), with its civil wars etc., as well as invasions?

    And what of the Marcomannic wars - in the period of climatic prosperity?

    The fourth century is a period of prosperity and growing social stratification and complexity in Germanic-speaking barbaricum, as a brief familiarity with the archaeology of the region will more than amply demonstrate. There is not a shred of evidence of factors like famine provoking migration, which can invariably be explained by reference to the exact same factors as produced migration in earlier centuries.

    Similarly, propserity in many of the provinces of the West. The East's economy boomed in the 5th century.

    The Empire was unusually coherent and stable in the 4th century - different from the 2nd-century Empire but no less stable I would say. No evidence that tax-yields were any lower. Indeed they were almost certainly considerably higher. And that begs the question of what you think the primary function of taxation is within the Roman political economy.

    Your view of barbarian migration seems to be based upon the old Voelkerwanderung idea of migration which has been seriously questioned by pretty much everyone except Peter Heather and Bryan Ward-Perkins. (OK, I know you're at Oxford at the moment, but still...)

    What's more, the crisis in the North Sea Coast area of Germany does not seem to be of long duration.

    Economic collapse around the North Sea is a very short-term process around 400 which - again as I said in my post - is badly served by a climatic explanation. And (yet again, as I already said) perfectly well explained in terms of the socio-political structures and the effects of them of high political crisis - the advent of minority rule.

  4. Well, fair enough. I know I don't know enough about the third century at least—thankyou for the JRA reference—, and as to the rest, actually today your Barbarian Migrations made it to the top of my reading pile precisely because the local orthodoxy round my way is something I'd like to see the students argue with more; there's really a lot of textbooks on their lists that defend large numbers of people moving, so we need to get you in there.

    On the other hand, though I too think it threatens to remove the individual from history as it's usually peddled, I don't find it easy to let go of the idea that apparent climate changes must have some kind of effect on an agricultural society at a fairly basic level. Of course, this society is really many societies, and they're all sufficiently complex that the *individual* responses to changing circumstances will be very diverse, all kinds of things playing out together, and *that's* where the analysis should be, how it fits together with all the other factors. But to say, oh well it just makes no difference, that seems equally problematic as to say it makes all the difference.


Note: only a member of this blog may post a comment.