Feudal Transformations XVI: two fields or three?

Diagram of a three-field agriculture system

Diagram of a three-field agriculture system

As with so many of the best bits of learning, a while ago I came up against something in a book that I was reading, for completely different reasons, that made me think anew about the fabled old feudal transformation (and you might think I’d thought enough about that, these days). This thing was a chapter by one Helmut Hildebrandt about the spread of the three-field system in Central Europe, by which he turned out to mean substantially Germany with a glance at the Paris basin.1 Over that area he argued that over the eighth to tenth centuries the system of using three fields in rotation, one for sowing a winter crop to be harvested in spring, one for a summer crop to be harvested in the autumn and one lying fallow to get the next winter crop, became fairly widely established, whereas it had been largely missing before that. I wasn’t thinking much of this till the date 1000 crossed the text and I suddenly wondered about the effects of increased yield on the economy, since as you may recall as much of an answer as I have to the whole transformation question is that, “it’s the economy, stupid”. So, does all this add up to anything I should have thought about by now?

Well, the evidence Hildebrandt had for the phenomenon is kind of horrible: he’s very largely arguing from the ways that rents were collected by ecclesiastical landowners over the period, as revealed in sources like polyptychs and leases. That makes sense in so far as they’re kind of all we have (though in England I don’t think anyone would try and argue about this without using survey evidence too2) but it has all kinds of issues. Churches had long-term land strategies in ways that lay landowners didn’t necessarily: they could be supporting a lot more unproductive mouths than the average lay household, not just in the community but in terms of poor relief and hospitality, and of course their land was never divided by inheritance so they could plan in a longer term.3 Their rents might therefore be exactly the place we’d expect to see systematisation but we can’t really argue from that that it got any further into the community. Of course, the churches were a lot of landownership, so it’s still significant. But since Hildebrandt was very happy here to argue against deducing significant change from such evidence where it would take away from his overall picture, on the grounds that underneath a rent structure the land can be organised any way that pays it, there are still problems.4 In fairness, there he was mainly talking about common fields, and the more work I see about common fields in the earlier Middle Ages the more I think the debate is basically anachronistic in the hands of everyone but Gaspar Feliu.5 No, I am cautious about accepting this phenomenon as anything like universal, but then so was Hildebrandt, emphasising variation and alternatives and making a complex picture of a tendency towards a three-field system that in some areas with special conditions worked out differently.6 But even if we say that it’s only a trend and that the ecclesiastical landowners we see doing it may be leading that trend, it ought to make some kind of difference to how much wealth is in the system and that is kind of the motor of change either side of 1000 as far as I’m concerned. So is this where that change is coming from?

Teaching diagram of the Feudal Transformation, by me

Teaching diagram of the Feudal Transformation, by me

Well, as we’ve seen before and will doubtless see again, there’s a problem with most ’causes’ of broad social change in the Middle Ages, which is that they tend to happen together and so one can usually argue that any one is causing all the others. This is the point of my famous diagram, above, after all. The biggest problem I have with this change in agriculture being such a driver is that it was new to me, because in Catalonia the situation is a lot more varied, with three-field going back to an uncertain date but two-field arable and grassland rotation equally common and vines messing up the picture by being a cash crop. People here at the right time tended to have land in a variety of small plots good for only one thing and a system is hard to construct for it.7 Hildebrandt’s picture really only covers Germany, the Low Countries and North-Eastern France, and the problem is that only the last of these really undergoes something that is easily recognisable as part of the feudal transformation model, and even there there’s a degree of top-down collapse of authority for other reasons that might be enough all by itself.8 Meanwhile, where this change is most marked is where there’s least other change. So if it’s a motor it isn’t much of one.

The other problem is one of the chicken and the egg. Here this is especially important. Hildebrandt did consider why this change that he saw was happening, and his belief was that the change towards common fields, at least, which is later than the change of field rotation as he saw it, is down to the increase of population requiring a greater yield from existing land and so idle land in awkward locations being brought into cultivation where before individual ownership had not been able to work it usefully.9 I think that seigneurial renders should probably also be considered as a driver there but we can easily guess where I got that from.10 Either way, the shift of systems is a consequence here of other things that have their place in the debate as causes. Even though it’s earlier than most of the big social changes embroiled in the feudal transformation model, a partial change in crop rotation seems likely to be an effect, not a cause, part of the bigger take-off run of the European economy in this era.11

So at the end I don’t think this gets me any new answers. But I am suddenly very conscious that to the best of my knowledge this kind of work has not been done for my area, and I’m not sure that sources exist from which it could, as yet. And that bothers me, because if I’m going to discount this there I’d like to do so from more than silence.

1. Helmut Hildebrandt, “Systems of Agriculture in Central Europe up to the Tenth and Eleventh Centuries” in Della Hooke (ed.), Anglo-Saxon Settlements (Oxford 1988), pp. 275-290. You can see why it was an unexpected find given that it’s about neither Anglo-Saxons nor settlement.

2. I suppose I think of Christopher Taylor, Village and Farmstead: rural settlement in medieval England (London 1983) but his “The Anglo-Saxon Countryside” in Trevor Rowley (ed.), Anglo-Saxon Settlement and Landscape. Papers Presented to a Symposium, Oxford 1973, British Archaeological Reports (British series) 6 (Oxford 1974), pp. 5-15, might be a better parallel.

3. Nowhere witnessed so thoroughly as in the regulations on bread in the Statutes of Adalhard Abbot of Corbie, translated as “Of Bread and Provisions in the Statutes of Adalhard of Corbie” in Paul Edward Dutton (transl.), Carolingian Civilization: a reader, 2nd edn. as Readings in Medieval Civilizations and Cultures Series 1 (Peterborough 2005), no. 32.

4. Hildebrandt, “Systems”, pp. 284-287.

5. Gaspar Feliu i Montfort, “La pagesia i els béns comunals” in Els grans espais baronials a l’Edat Mitjana: desenvolupament socioeconòmic. Reunió científica. I Curs d’Estiu Comtat d’Urgell (Balaguer, 10, 11 i 12 de juliol de 1996) (Lleida 2002), pp. 23-40; cf. C. T. Bekar & C. G. Reed, “Open fields, risk, and land divisibility” in Explorations in Economic History Vol. 40 (Amsterdam 2003), pp. 308-325, doi:10.1016/S0014-4983(03)00030-5, about which as you may remember I had views.

6. Hildebrandt, “Systems”, pp. 279-284 (esp. 282-283) and 287-290.

7. Working from Peter J. Reynolds, “Mediaeval Cereal Yields in Catalonia & England: An Empirical Challenge” in Acta Mediaevalia Vol. 18 (Barcelona 1997), pp. 467-507, and further work collected in Immaculada Ollich, Maria Ocañ & Montserrat Rocafiguera (edd.), Experimentació arqueològica sobre conreus medievals a l’Esquerda, 1991-1994, Monografies d’Arqueològia Medieval i Postmedieval 3 (Barcelona 1998), online at http://books.google.co.uk/books?id=sK1ptZDwfV8C as of 28 July 2011, where they really don’t have an archaeological basis for separating the early and high medieval field systems; Reynolds’s initial paper describes them using both two-field and three-field in their tests precisely for this reason.

8. Classically described in Jean-François Lemarignier, “La dislocation du « pagus » et le problème des « consuetudines », Xe-XIe siècles” in Charles-Edmond Perrin (ed.), Mélanges d’histoire du moyen âge dédiés à la mémoire de Louis Halphen (Paris 1951), pp. 401-410, repr. in Lemarignier, Structures politiques et religieuses dans la France du haut Moyen Âge, ed. Dominique Barthélemy, Publications de l’Université de Rouen 206 (Rouen 1995), pp. 245-254.

9. Hildebrandt, “Systems”, pp. 286-287.

10. Chris Wickham, “Problems of Comparing Rural Societies in Early Medieval Western Europe” in Transactions of the Royal Historical Society 6th Series Vol. 2 (Cambridge 1992), pp. 221-246, rev. in idem, Land and Power: studies in Italian and European social history, 400-1200 (London 1994), pp. 201-226, and his “Sul mutamento sociale e economico di lungo periodo in Occidente (400-800)” in Storica Vol. 23 (Roma 2002), pp. 7-28, repr. as “Per uno studio del mutamento di lungo termine in Occidente durante i secoli V-VIII” in Quaderni del Dipartimento di Paleografia e Medievistica Vol. 1 (Bologna 2003), pp. 3-22, transl. Igor Santos Salazar & rev. Iñaki Martín Visó as “Sobre la mutación socioeconómica de larga duración en Occidente durante los siglos V-VIII: on the long-term socio-economic change in the West from fifth to eighth centuries” in Studia Historica: historia medieval Vol 22 (Salamanca 2004), pp. 17-32, the last of which is where I read it.

11. On which see La croissance agricole du haut Moyen Âge : chronologie, modalités, géographie. Dixième Journées Internationales d’Histoire, 9, 10, 11, Septembre 1988, Flaran Vol. 10 (Auch 1990).

25 responses to “Feudal Transformations XVI: two fields or three?

  1. Once again by some mistake of settings I have left a scheduled post comments-off, sorry. That and a number of typos are now fixed.

  2. Excellent post Jonathan and I really should read up on some of the sources more before commenting but I found your next-to-last paragraph very interesting. I think it’s a misconception among some of “the masses” (I’m about halfway one of them myself) which can be easily inferred from some of the very general references in things like Wikipedia and online encyclopedia entries that this kind of change was some sort of organic natural evolution in response to increasing populations competing for limited land resources.

    I have a feeling that if I ever get a chance to really look at the sources/evidence for this I’ll find it fascinating and puzzling. How and why would landowners have realized that the three-field system would have been more efficient and told their peasants to adopt? (I can’t come up with this change on individual holdings being done without the landowners so directing) Maybe realized is the wrong term and it should be, “suspected but didn’t adopt until they were forced to by some sort of crisis.” I find myself wondering if the Benedictines or other orders might have been first adopters and with time thinking on this by secular landowners would have changed from one of, “look at what those crazy monks are doing” to, “gee, my peasants are making too many babies for me to feed and with increasing opportunities to market excess production I need to find a better system which the monasteries seem to be using.” One of a bunch of options – for all I know (haven’t read it) Varro’s writings on agriculture may have been known and discussed this.

    Anyway, I love posts which make me think and yours did and thank you. Sorry about all the parentheses, I’m just emerging from a turkey-coma and my thoughts are less clear than I’d like.

    • I thought you might like this, Curt, but I don’t really have an answer to the question. I did, on the other hand, see a paper by John Blair relatively recent arguing that the obvious transmission context for such technical knowledge was, in fact, monastic, and we should expect that every now and then someone running a monastery’s estates to try and put this stuff into practice after reading it. I do, however, think that the main pressure on aristocrats themselves is either tax, in the states that can raise it (and that mainly means England in this period) or else simple competition. As I’ve said before, if there’s more wealth generally in the system and wealth is agricultural, richness increases at the bottom first and everybody gets some. That makes keeping up with the medieval Joneses rather tougher; suddenly a lot of little lords are contending for the next step up, and showing you’re clearly ahead of the pack means spending that bit extra. So, where’s it coming from? And the answer is as you describe, I suspect!

  3. Here’s a different view, from Joachim Henning (forthcoming 2012, says the academic.edu page), in case you hadn’t come across it.

    • I had downloaded it but not read it yet. Now I have. He’s looking at a much earlier change than I am, but I don’t think I have a problem; I just want it to be earlier than the great Transformation. That said, with the three-field system specifically, the evidence is very thin and the spread of the system isn’t really demonstrated, which is fair enough, he’s only looking at a region. But how typical is the region? Why doesn’t it happen elsewhere? And why don’t the supposedly resource-effective areas suffer the Transformation so much? Is it really all about competition, as I suggest above?

  4. I have not read any of the articles so i should really not comment at all. But it us nearly irrisistible in view of another book I have read, the very recent one by Bachrach on ottonian warfare, where he argues very convincingly about the enormous size of the ottonian armies. One point here – which he does not make – is the economy in letting food walk its own way… It us so much neater to have cattle walk behind an army than carting grain along. And one of the points about the three-field system is the increase in fodder. And the numbrr of animals you don’t have to cull in november. But just to make myself clear: even in the early modern period where the threefield system was the model par excellence, there existed all sorts of permutations… E.g.combinations with outlying fields etc. And anyway this does not answer the question about what came first ..siege warfare or ???

    • There is an English book on pastoralism with the very perceptive title The Walking Larder, but of course those beasts still need feeding till you want to eat them! As for siege warfare, I don’t think it’s a good idea for unestablished scholars to try and address questions in which Bernard Bachrach is interested, really… I’ll leave it to Guy Halsall!

      • Jon, this is David Bachrach, Bernie’s son, but much the same applies. I should say that I, as a part time Ottonianist, think that David Bachrach’s work is very stimulating, but remain unconvinced that kings could regularly raise armies much larger than the largest urban centres in Western Europe at the time. Good discussion of the issues behind this can be found in some of Tim Reuter’s articles, as well as Guy Halsall’s book.

        • Thanks for the correction, and also the comparison of scales, which is quite powerful!

        • David Bachrach

          Dear Levi,
          I feel a bit as if I have eavesdropped on a conversation, but I was looking for an early review of my book, and came across your observations here, and thought I would ask a question or three. I hope this isn’t impolite. As an impartial observer, I am curious what you find compelling about the arguments of Reuter and Halsall regarding the size of armies. As I understand their work, both of them rule out sieges as an important factor in early medieval warfare, which would seem to contravene both the written and material sources. In addition, both of them assert that large armies are impossible for logistical reasons, but neither one of them ever published a study, as far as I know, demonstrating the logistical realities that they are claiming. Finally, I wonder about the connection between the size of cities and the size of armies. I have never understood this.

          Again, I apologize if this is off base.

          best wishes,


          • It’s not off base as far as I’m concerned, but there are probably more effective places to solicit opinion from either of Professor Halsall or Levi than comments on an old blog post of mine. My limited sense is that Guy’s position is peculiar in as much as it’s attacked from both directions, for imagining armies that some find implausibly large, in the low thousands, usually in an Insular context, and that others in a Continental context, apparently yourself included Dr Bachrach, find implausibly small. Thus, while Guy and Levi might make good proxies, the argument I’d really like to see is between, say, Nicholas Brooks or Peter Sawyer and the elder Professor Bachrach. In fact, it occurs to me that the latter version of this has probably already happened in print somewhere… But I hope that we can get some kind of conversation going here anyway!

            • David Bachrach

              Dear Jonathan,
              I am very new to the blog business, so I am not sure where the original post went, or where to reply, but I will try to do a credible bit here. The elder Bachrach, I usually just call him Bernie, certainly would be a good person to engage in the big army debate in person (and contacting him with an invitation at bachr001@umn.edu might well do the trick). But I will do my best as a stand-in.

              Both Sawyer and Brooks were mostly concerned with Anglo-Saxon and Viking armies, and Bernie doesn’t deal much with either, but the methodology utilized by Brooks is certainly one with which Bernie is comfortable, that is the application of Sachkritik to the problem of numbers. It is in this area that an understanding of the scale of siege warfare, which did dominate all early medieval warfare pace Halsall, is crucial not only with regard to understanding the size of attacking armies, but also the administrative and economic wherewithal that were necessary to construct, maintain, and man vast numbers of large strongholds over a very large territory.

              To give just one example, it is clear from both written and material evidence that Otto I maintained during the 950s and 960s a minimum 100 substantial fortresses along the eastern frontier, and in the region between the Elbe and the Oder. This number could well have reached 250 depending on how we evaluate the archaeological reports from this region, and the date we assign to fortifications that appear for the first time in the written record in the period 973-1024. To construct just one medium sized earth and timber fortress with a mortared stone-front (a standard type) of some 600 meters in circumference required in the neighborhood of 3 million man hours, or some 300,000 man days of labor, assuming a ten hour working day without any breaks for meals, accidents, or other events. By way of contemporary analogy, the planners who established the Burghal Hidage conceptualized a fortress of 600 meters requiring approximately 450 men to defend the walls against a determined attack. Once we start plugging the amount of labor to construct fortresses, and the number of men required to defend these fortresses into the number of fortresses that we know Otto I possessed, the small-scale, primitive, archaic model begins to evaporate quite quickly.

              all best,


              • Well, Levi is a reasonably avid reader and may well see and join this conversation, my point there is merely that nothing will draw it to his attention if he doesn’t notice your name in the comments (unless he has selected the ticky-box that promises hm e-mail when the thread gets comments). Thus, he may never notice this exchange. But we can hope!

                For my part, I haven’t done the reading or the source work to critique either position very much, but I suspect that one of the things I would expect to come out of a Sawyer-Bachrach debate is that the expectations of scale that both parties have for the polities they know best are quite different. Anglo-Saxon England just wasn’t as large as Germany, and the resources required to defend it were smaller. So, for that reason alone, I am reluctant to go too far from the Burghal Hidage manpower figures toward Germany. The German fortress numbers are going to be an order of magnitude higher than the English ones and it seems to me likely that their operation could also have been of a different order. But I hope I’m open to all views with the same degree of scepticism!

  5. Verhulst, Carolingian Economy (CUP, 2002) (pp. 61-64) thinks there was three-course rotation, but not the three-field system in the Carolingian period, and summarizes the literature (including Hildebrandt, whom he disagrees with). For the very latest views, it might be worth talking to Debby Banham and Rosamond Faith, who are just finishing a book on Anglo-Saxon farming.

    • I’m working through Jean-Pierre Devroey’s Économie rurale et société dans l’Europe franque (VIe-IXe siècle) at the moment and will be interested to see how that bears on it. I am inclined to mistrust Verhulst’s book ever since I discovered (and here I think I mean, ‘you told me‘) it still used Duby’s crop yields figures, but it’s always nice to know what it has to say and I must make time to read that also, because so many others will have done.

      • I think that was the discussion in which I was being dismissive about research on crop yields, so apologies for that. But I’m still not convinced that agricultural change is going to be an important factor in the Feudal Transformation if you’re going for the FT as being a relatively short period of rapid change (50 years or so). Agricultural change (and technological change more generally) seems to me not to work on those kind of time scales. It’s also particularly unlikely to be affected by political borders, while the Feudal Transformation undoubtedly is affected.

        I think everyone would agree that the economy and population grew between 800-1000. But in order to claim that there was a sudden growth spurt at the end, you’ve either got to claim the Carolingian economy is purely subsistence (as Duby does), say that the Vikings/Hungarians/Saracens are wrecking the tenth -century economy far more than is plausible, or say that there are suddenly many more opportunities for entreprenurial peasants towards 1000 (which may work on the Spanish frontier, but I think is hard to show elsewhere).

        Or you have to have an external factor that could cause rapid change in agricultural yields, which is climate change. Have you come across Ronnie Ellenblum The Collapse of the Eastern Mediterranean: Climate Change and the Decline of the East, 950-1072 (CUP, 2012)? This seems to be suggesting that a relatively localised climate change could have major effects, but from a quick preview of the index, focuses heavily on narrative/documentary references to drought, famine etc, rather than archaeological/climatological evidence. I’m not sure that you could find a reliable enough source base to look at Western Europe systematically, especially since good weather is less likely to get recorded by chroniclers. And again, climate change really doesn’t respect political boundaries, which is problematic for the FT.

        • Last first, I hadn’t met the Ellenblum reference, but that’s very interesting as I think all his stuff is, albeit often contestable. Thankyou for that. In fact the scientific base for evaluating such matters in Western Europe is getting slowly better, but in Catalonia at least there are also some fairly suggestive implications in the documentation.

          That all said, I’m not massively invested one way or another in the c. 1000 date for whatever set of changes we’re seeing in each area, so much as the causal relations or lack thereof between them. I’d be as happy with a slow change as a fast one as long as the nature of that change can be properly and regionally assessed. And the open nature of the Spanish frontier should be just as much of a draw around 850 as it is around 1050; one therefore needs the population surplus or the political expansion to justify the peasant one (and I think the latter, at least, is a lot easier to document. I mean, in general, my official position for the Feudal Transformation in Catalonia is that it has an easy set of political explanations—expansionist state on open frontier suffers sudden power vacuum leading to local military lordship acquiring temporarily unstoppable dominance, so much so Bonnassie—that doesn’t apply elsewhere, so, why do other places sometimes look similar eh? etc.).

          Lastly for the crop yields, yes, basically. If as it seems Carolingian agriculture is not basically disastrous, and the agricultural take-off run is socially significant in its actions, then we ought to see the effects sooner. One might argue that the imperial collapse retards that trend but then one might also argue that the resultant independence of lords and the lack of restrictions upon their exploitation of their lands unleashes that the full economic potential of the situation. I don’t know what I’d say yet but what I will say is that I’ve read the relevant section of Devroey now, and that it raises a complex enough set of purely agricultural causes and effects that I want to write about it separately, so, look for that in a few months…

          • David Bachrach

            I am not sure whether this has been covered in the discussion on crop yields up to this point, but there is a factor that is quite important in the German literature on agricultural production, and this is the rapid dissemination of the bi-partite estate in the Frankish heartlands during the course of the later 7th and 8th centuries, and the importation of this system into the eastern lands during the course of the 9th and 10th centuries. Whether or not there was an improving climate beginning during the second half of the 8th century (and I believe that it is becoming clear that there was), the self interest of the peasants who now farmed land on their own rather than subsisting as gang-slaves on Mediterranean-style latifundia would seem to be an important factor in understanding increasing productivity of individual farmsteads, and therefore the gross increase in “gnp”. It is noteworthy, I believe, that slaves appear very infrequently in charters during the ninth and particularly the tenth century in the eastern lands, and we see instead various states of semi-freedom with all of the implications that this has for the stability of the family, and the desire to pass on a better life for the children. This same argument was made quite compellingly with regard to the Byzantine empire concerning the 7th-9th centuries by a number of scholars. Werner Rösener is a major figure in this discussion for the Carolingian and Ottonian period in the eastern lands, along with Devroe and Verhulst. One significant critic of the effectiveness of the bi-partite estate in spurring agricultural productive is Joachim Henning.
            all best,


  6. Pingback: Carnivalesque #91 « Dr Kate Ash

  7. Dear Jonathan,

    I have just discovered that you have added Henry the Young King’s blog to your blogroll. I do not know how to express my (and Henry’s) gratitude.
    I think there is only one possible way of solving my dilemma. Actually I have already solved it by adding your website/blog to the list of the blogs I follow. Your work is impressive indeed and I am going to pay frequent visits to you and your “Corner”.

    Thank you,

    Kasia Ogrodnik

    • A pleasure! I read, I enjoyed, it has references and was updated in the last quarter, I think the readership here may be interested, a link seemed a good idea. Thankyou for the reciprocation!

  8. Forgive me for sounding pig-ignorant, but when *is* the first mention of cultivo al tercio south of the Pyrenees?

  9. Pingback: Feudal Transformations XVI: who wants that third field? | A Corner of Tenth-Century Europe

  10. Pingback: Two fields, three fields, four fields, five… | A Corner of Tenth-Century Europe

Leave a Reply

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

WordPress.com Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.