A Theory under Siege

You may recall that some time ago I recorded that I had been reading David Bachrach’s Warfare in Tenth-Century Germany as part of my ongoing reworking of my paper about military service in tenth-century Catalonia.1 There are ways in which that was really useful and ways in which we are just talking about different things, but a thing that we both talk about but take quite different views on is the normal size of early medieval armed forces. Now, this is an old controversy, which goes back far beyond this book and involves names like Halsall, Sawyer and not least, Bernard S. Bachrach, and in some ways I would prefer not to get into it, not least as several of these people including Professor Bachrach (the younger) have been known to read this blog, but I can’t avoid it.2 I would confine myself to an argument that the situation I see in Catalonia, of guardposts and border-raiding and no serious armies except when an Andalusi one comes calling, is just very different from the full-on imperial warfare of Ottonian Germany, and in the article I expect that I will. But still, there are premises to the large army argument, as set up in this book at least, that I find hard to accept, and this post is an attempt at a critical examination of one of them. Where most openly stated, it goes like this:

“Ottonian military operations were consistent with warfare throughout the medieval millennium, which was dominated by sieges, particularly in the context of campaigns that were intended to conquer territory. Contrary to the long-established narrative that the Ottonian kings fielded small armies of a few hundred to a few thousand heavily armed mounted fighters (Ritterkrieger) led by warrior aristocrats, the siege operations that dominated warfare in the tenth century required very large armies, composed predominantly of foot-soldiers.3

One is, initially at least, left to assume the premises behind this. It’s easy enough to come up with some: walled cities are naturally quite large, and need a lot of defenders to keep a whole perimeter secure. If you’re attacking them, you must need more attackers than defenders, right? As it turns out, in fact, you allegedly need quite a lot more: once you get through to p. 226 you learn that, “in order to storm a strongly held enemy fortress an attacker required four to five men for every defender.” The justification for that, however, is farmed out to an article by Bernard Bachrach and Rutherford Aris.4 Now, if you actually get hold of that article, that doesn’t seem to me to be what it says: instead, by virtue of some extremely hypothetical probablity mathematics, it says that a charge by a Viking warband at a typical Anglo-Saxon burh defended by archers of the number implied by the Burghal Hidage (1 to every 1.3 m of wall) would probably have resulted in one to two attackers in every four being hit before they reached the wall.5 Even if that mathematics were somehow realistic, it’s quite a specific situation and one calculated on the basis that Vikings attacking a fortress would only have ladders to deploy so needed to make that approach.6 I don’t see how it can be transportable to a large-scale military operation with siege engines such as (David) Bachrach thinks the Ottonians were able to mount.7 Moreover, it doesn’t actually provide the numbers that he employs, only an implication that if you were to outnumber your opposing force once atop the walls you needed to allow for a fifty per cent casualty rate when planning your attack. That seems like awful odds that no commander would have risked to me, but it’s obviously not what an Ottonian planner would have been facing.

Interior view of the restored Romanesque city walls at Worms

Interior view of the restored Romanesque city walls at Worms, one of the fortress cities Bachrach considers and possibly not too much unlike this in the Ottonian period? CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=167751.

But let’s take one step further back. These are figures based on the idea that victory at a siege is achieved by storming and carrying the enemy’s defences, and indeed so is Bachrach’s account of Ottonian warfare against fortifications more generally.8 But this is, surely, not the only way. More conventional, if much slower, would be simply to starve the defenders out. Now, in that case, your army requirements drop radically. Whereas before you perhaps had to have men all round the fortification, now you really only need to guard the points of access. Possibly someone could resupply a city by hauling sacks up the walls in the dead of night but not, you might think, in any real quantity, especially if you as attacker are sending patrols around every twenty minutes or so, which I assume you, as a thinking tactician, probably would be. Their supply will still not be equal to their demand. This makes siege warfare a much less demanding effort in terms of numbers; one must still be able to supply the besiegers for a prolonged period of time, but that’s easier if there’s fewer of them needed.

Illustration of a (small) siege from the fourteenth-century Codex Manesse

Even more anachronistic, but far too good a picture not to use, an illustration of a (small) siege from the fourteenth-century Codex Manesse

Of course, you still need to have more men than the defenders do, otherwise they could just sortie out and squash your forces. But wait a minute. How many more men than you do they have to have to do that before it would mean leaving the walls partly undefended, giving your troops access to the city and thus losing the war for the sake of a battle? There’s more balance here than might immediately appear, because both sides need their forces at least partly dispersed to hold their positions. The various ways that concentration of force could play out here seem to me to be very hard to calculate with and I’m not at all sure it reduces to any simple arithmetic beyond the basic logic that the more troops whom you can feed you have available, the more tactical options you have. But in a siege situation, that proviso about feeding is quite important, often for both sides.

Crop-marks clearly showing a fortress, supposedly the Slavic fortress of Gana, at Hof-Stauchitz

A different (and more usual?) scale of opponent, crop-marks clearly showing a fortress, supposedly the Slavic fortress of Gana, at Hof-Stauchitz

I should say that I am not, a priori, against the idea that the Ottonians could sometimes field quite large armies, meaning in the double figures of thousands of men, although whether such were necessary for all their campaigns I rather doubt. When they were, though, they had a whole empire to draw upon and Bachrach has here a whole book full of details about how such things might have been organised, resourced, supplied and led which almost make his argument for him; one can reasonably assume that the Ottonians would have wanted to raise large armies and Bachrach shows us how they could have. But he argues it the other way around, from the necessary existence of large armies, through a reading of the sources which illustrates that, to the techniques for their provision and operation, and if these are the starting premises, they seem very shaky to me.

1. D. S. Bachrach, Warfare in Tenth-Century Germany (Woodbridge 2012, repr. 2014).

2. In reverse chronological order, more or less, Guy Halsall, Warfare and Society in the Barbarian West, 450-900 (London 2003); Bernard S. Bachrach, Merovingian Military Organization 481-751 (Minneapolis 1972), what I think of as the wellspring of the maximalist argument, and Peter H. Sawyer, The Age of the Vikings (London 1962, 2nd edn. 1971), which is the place where I first met a sustained attack on the numbers which early medieval sources used for army size. In all these cases, albeit Sawyer least and Bachrach most, references could be proliferated; a lot of Bachrach’s most relevant works are collected in his Warfare and Military Organization in Pre-Crusade Europe, Variorum Collected Studies 720 (Aldershot 2002).

3. D. S. Bachrach, Warfare in Tenth-Century Germany, p. 1, which is a good place to state your axioms after all.

4. B. S. Bachrach & R. Aris, “Military Technology and Garrison Organization: some observations on Anglo-Saxon military thinking in light of the Burghal Hidage” in Technology and Culture Vol. 31 (Baltimore 1990), pp. 1-17, on JSTOR here, repr. in B. S. Bachrach, Warfare and Military Organization, III, cit. D. Bachrach, Warfare in Tenth-Century Germany, p. 226 n. 2. A much fuller version of the argument here is however to be found in B. S. Bachrach & David Bachrach, “Early Saxon Frontier Warfare: Henry I, Otto I and Carolingian Military Institutions” in Journal of Medieval Military History Vol. 10 (Woodbridge 2012), pp. 17-60, which was presumably not available to cite in D. S. Bachrach, Warfare in Tenth-Century Germany, and would have made that a much longer book to reprise there. I engage with the version in the book here, which I assume to be self-standing.

5. B. S. Bachrach & Aris, “Military Technology”, pp. 5-10 with an appendix for the mathematics pp. 14-18.

6. Ibid., pp. 3-5, on the not unreasonable basis that the Burghal Hidage uses the same ratio for number of defenders required for places with eight-foot thick Roman stone walls (Winchester) and earthen ramparts with wooden palisades (Wareham), so the writers obviously weren’t thinking in terms of attacks actually upon the defences.

7. D. S. Bachrach, Warfare in Tenth-Century Germany, pp. 151-167, covers Ottonian siege equipment, including ladders but also mantlets, rams, ballistas, catapults and mobile towers, largely by providing detail from Vegetius and anchoring it in less-detailed reports from Ottonian sources.

8. Ibid., pp. 12-13 and repeatedly exemplified in his narrative of Henry I’s and Otto I’s campaigns provided pp. 14-69; see also Bachrach & Bachrach, “Early Saxon Frontier Warfare” for a fuller statement of the position.

32 responses to “A Theory under Siege

  1. David Bachrach

    Dear Jonathan,

    I read your post today on your blog about my book. I must say that I appreciate the publicity. However, I also should note that I do not believe that your response is well substantiated. If you would prefer, I would be happy to respond directly on the blog site and stimulate controversy and eyeballs. But there are a few points that I think are worth considering based on what you have written, which I am putting in below.

    all best,


    1. How large do you believe were the armies and fleets mobilized for the capture of the Balearic islands and Barcelona by the Carolingians in 801-802? Since you grant that the Muslims were able to mobilize large armies, what kind of force was necessary to block the relief of the siege of Barcelona? Given that we know the size of Barcelona, how large an army was required to keep the Muslim garrison from launching an attack? Similarly, how large do you supposed the Carolingian armies were over the next decade that captured substantial garrisoned fortresses held by the Muslims, also keeping in mind the need to fend off Muslim relief armies. Your suggestion that the Carolingians merely had guardposts and engaged in border raiding does not seem to account for this decade and more of military operations.

    2. In your criticism of the article by Bernie and Gus Aris, you state:

    “Even if that mathematics were somehow realistic, it’s quite a specific situation and one calculated on the basis that Vikings attacking a fortress would only have ladders to deploy so needed to make that approach”.

    The first point here is a classic fudge and is quite tendentious. Do you actually have a reason for doubting the mathematics based upon a knowledge of the subject, or are you simply obfuscating matters because the results are inconvenient for your a priori assumptions?

    As for the second issue raised here, either you have misunderstood the point made by B and A, or you are ignoring the core issue. The specific formation that they point out is the one that would make it possible for the attackers to suffer the least number of casualties in an assault. In any other type of assault, the number of casualties would be much higher.

    3. The military planners who developed the garrison numbers for fortresses in Anglo-Saxon England were very well informed about Viking tactics. It is not clear how you come to your conclusion in the note to this paragraph that the use of a consistent ratio for both earth and timber, and stone fortifications indicates that they were not interested in actual Viking assaults. If they were not interested in viking assaults, what then were their concerns?

    You also assert that the ratios used by the Anglo-Saxon military planners are not transferable to large scale military operations in Ottonian Germany. This is problematic for several reasons. First, cities such as Winchester had circuits of walls as large as those as cities in Germany, e.g. Mainz and Regensburg. Second, you offer no reason why similar technological situations would not lead to similar technological solutions. Third, many scholarly studies including the exceptionally important article by Brooks, who makes very similar arguments to Bernie, make clear that the vikings could and did mobilize very substantial armies numbering in the many thousands. Moreover, they did so on a population, wealth, and administrative basis much smaller than that available to the kings of Wessex, and miniscule in comparison to the kings of Germany.

    Finally, in this paragraph, you misleadingly state that I do not provide numbers for defenders. But this is not the case. In my discussion of the siege of Mainz, I provide numbers for both the defenders and the attackers, and these are based on a range of sources of information, including the most advanced demographic studies of Mainz.

    4. In terms of military planning, not having the option to assault a fortress placed severe limits on the strategic choices available to a military commander. As I point out in the book, there were a number of cases in which the Ottonian commanders were able to compel the surrender of a fortress or city by making clear that they had the ability to storm it, and thus avoided both bloodshed and a lengthy siege. There are, however, clear cases when Ottonian commanders did undertake costly assaults on fortifications. They were not alone in this. It is clear that assaults to carry the walls were undertaken by commanders throughout the Middle Ages when the commanders deemed it necessary. Throw away lines are cute, but you actually need to pay attention to what real military commanders did in real situations.

    5. As I point out in my book, Ottonian commanders often did undertake lengthy sieges. But again, you have a certain number of throw-away lines here that are not actually substantiated by real analysis or numbers. The siege of Regensburg in 954 required the blockading of two major gates that were separated from each other by over a kilometer. Otto had to divide his army into two parts, and blockade each gate. In addition, Otto had to be concerned about relief armies, a point that you do not discuss in your post. As it turned out, each of these blocking forces had to sustain the full weight of an attack by the defenders of Regensburg. In order to develop a model for the size of the armies that a particular military commander had to bring to bear against a city for a lengthy siege, you need to know what he knew about the size of the forces within the city and also the size of the forces that might relieve the city. At one level, your comments about the need to disperse forces are accurate, but you do not begin to scratch the surface of the information that is necessary to develop a model.

    Finally, I do not understand the basis of your criticism in the final paragraph. How is it a problem to make the case that the Ottonians engaged in large scale siege warfare, and therefore required large armies? The state of the question, or at least one version of the state of the questions pushed by Reuter and Halsall and others (perhaps you as well?), is that there was no siege warfare in this period, and that therefore there was no need for large armies. By making the case that there was substantial siege warfare, I am overturning part of the historiographical tradition. Having made that point, I make the complementary point that the Ottonians certainly had the capacity to support large armies in the field. Making the second point first and the first point second is to miss the nature of the historiographical discussion in the field. In addition, from a logical basis, why would one make the point that the Ottonians had vast wealth and the organizational chops to mobilize it for war, unless one had already made the point that they had a need to mobilize large armies?

    • Dear Professor Bachrach,

      thanks for this comprehensive set of objections! I am very happy to take this debate into blog comments, which is hopefully not just publicity for me but also for you (I do see the problematic nature of provoking someone to bring me traffic like that, but that was not my intent, and in any case I get very little benefit from such traffic and they would be coming to read you also). There are some things in what you say in your mail that make me think I must have been sufficiently unclear as to be misunderstood, however, so perhaps it is worth a first exchange by e-mail to ensure that we don’t argue at cross-purposes.

      To your no. 1, I should make it clear that I am not maintaining that the Carolingian war effort during Louis the Pious’s campaigns in the Iberian Peninsula was based on the tiny inaccessible guardposts which survive, or their concomitant forces; my paper is going to argue that this is a sub-Carolingian situation into which the area descends post-830. You are right that the siege of Barcelona must necessarily have involved larger forces, though the model of Narbonne and Girona, where the local garrisons were easily expelled by the ‘Goths’, suggests that the real danger to the besiegers in this country was probably from outside. I don’t recall any other cities captured after Barcelona apart from Tarragona, however, which I don’t believe there is any indication was considered defensible by the Muslims (and the Carolingians seem to have thought similarly, given how quickly the gave it up again!).

      To your no. 2 I suspect we have a philosophical disagreement. I’m sure the arithmetic is accurate; every one of its premises is an unknown, however, and if each one could somehow be accounted for as an error bar in the final calculation, given how many such variables have to be fixed down to calculate with, it would be meaningless. You’ll notice that I don’t actually dispute the figure in the post, only its applicability to Ottonian forces, but it is by its nature unverifiable, and worse, the result of multiplying numerous unverifiable components. I hold that such data can never be usefully `right’.

      As for the argument about the ideal formation for such an attack, the point I am making in the post is that Ottonian forces would not be restricted to unsupported open-field attacks with only ladders, so the formation in which they would not do that is irrelevant, but in any case I still don’t see how the casualty figures in any version of the article’s hypothesis lead to the figures for which you cited it.

      As to your point 3, I don’t at any point in my post suggest that the Burghal Hidage is not about how to defeat Viking attacks. I think that perfectly plausible and more or less say so in my n. 6. I’m not sure how I have been misread here but I’m not saying what you defend against here.

      Nextly, I am not saying that the Burghal Hidage ratios would be inappropriate for defenders of an Ottonian-period forces. I am saying that the situations differed, but that is again a difference on the attacking side; your Ottonian armies were equipped for siegecraft with artillery, mines, towers and all the tricks of Vegetius. It is the attacking forces, and your suggestion that they needed to be four to five times the size of the defending forces, with which I am contending there. The Burghal Hidage gives us plausible figures for the defending forces (and as the other article I cite suggests, those could obviously be much larger), but I can’t elicit from either article or your book where your multiplier has come from if not these hypothetical casualty figures for a much less well-equipped Viking force attacking an equally well-defended fortification. These situations seem very unlikely to make a good parallel to me as far as attacking forces go.

      The rest, we can argue about profitably without modification I think: your numbers for Mainz are hypothesized and while your count for the defenders is plausible, if the city’s full population had stayed for the siege (or even if only the militarily-capable ones had) the count for the attackers is based on your multiplier which again you base on the 1990 article by Professor Bachrach and Aris, without closer specification. I agree entirely that obviously sometimes taking by storm was done, and perhaps on those occasions larger forces were raised (or, indeed, the option existed because such forces were available). I don’t think a workable model for necessary size of army to protect against a sally by defenders is derivable–too many specifics in each case. I don’t hold that there are no sieges in this era as per Halsall and more or less accept both your point and its difference from the historical tradition in that respect, although I would differ from you in the reading of Widukind you deploy in proving it–a future post I fear. As for the logic, I merely suggest that showing the Ottonians’ interest in and ability to field large armies to argue that they did so avoids the problem of taking the large armies as a given (or as proven from insufficient evidence) and then interpreting the sources in the light of that conviction.

      • David Bachrach

        First, I am glad that we agree that the Carolingians deployed large armies for the conquest of Catalonia. About matters in the 830s in this region, I am both agnostic and more importantly ignorant.

        Regarding point two, however, I do believe that we have a philosophical disagreement, but I also think that you are misstating the nature of the information and conclusions used and drawn in the Aris/ Bachrach article.

        With respect to philosophical disagreements, my view is that if we are given concrete numbers by our sources, a great rarity for the early period, it is our task to use these as best we can to develop models. We can argue about the models, and try to fine-tune the models, but to reject model building itself seems to me to be contrary to the goal we have as historians. But it sounds as if you have rejected the possibility of modeling with respect to siege warfare.

        With respect to Aris’ modeling of the rate of arrow fall, and the probability of individual arrows hitting individual men under a specific set of circumstances, it is not clear to me what variables you believe are unknown, unknowable, or not reasonably modeled.

        We know very well, based on contemporary finds of bows about the power of the weapons that were available to both Anglo-Saxons and Vikings, and for that matter, the Ottonians. The finds of bows at Haithabu, for example, are useful in this regard.

        Physics has not changed in the last 1100 years, and so given our knowledge of the power of bows, we can know very well the range of these bows, the velocity of the arrows, and consequently, the probability of a certain number of arrows falling within a specified space within a specified period. In terms of the rate of fire of the archers, Aris and Bachrach used a conservative estimate so as to provide a best case scenario for the attackers and a worst case scenario for the defenders and thus NOT to exaggerate the number of attackers who were necessary to carry a fortress by storm.

        We also know how fast adult male human beings encumbered with equipment, on average, can move over specific distances. This information has been developed in a range of experimental contexts.

        In addition, we know the minimum number of men who were supposed to be in a garrison of any particular fortress if we know the length of that fortress’s walls. This is a minimum figure because all of the able-bodied men, and likely some able-bodied women within the fortress can be expected to have helped in its defense, and some number of these will have been equipped with bows as well.

        As a consequence, it is not clear to me to which unknowns you are referring when you state:

        “I’m sure the arithmetic is accurate; every one of its premises is an unknown, however, and if each one could somehow be accounted for as an error bar in the final calculation, given how many such variables have to be fixed down to calculate with, it would be meaningless.”

        I will return to the question of modeling casualty rates below.

        In terms of the postulated open formation of the attackers used by Aris and Bachrach, as I indicated in my previous note, this is the formation in which the attacker would sustain the least number of casualties when attacking over a killing ground. Aris and Bachrach used this model again to offer a best-case scenario for the attackers and a worst case scenario for the defenders so as to avoid exaggerating the number of attackers who would be required to carry a fortress by storm.

        Consequently, your comment that

        “the point I am making in the post is that Ottonian forces would not be restricted to unsupported open-field attacks with only ladders,”

        misses the point entirely of their article, and also does not take into account the thinking of Ottonian military commanders, whom I deal with in my book.

        As to whether the Ottonians were limited to attacking in open formation with ladders, you are correct that they had available to them a considerable amount of siege equipment, including various types of artillery and rams, as well as ladders. But, significantly, they do not appear to have siege towers available to them, at least during the reign of Otto I. The earliest reference to a siege tower is in Richer c. 1000, and refers to the siege of Verdun by Lothar IV during the reign of Otto II.

        Just as significantly, it is also the case that the defenders of fortresses in the Ottonian period possessed artillery. This was true of both German and Slavic opponents. The walls of cities and large fortresses were equipped with stone throwing and spear casting engines that had much greater range and power than the self-bows postulated by Arris and Bachrach in their discussion of the Burghal Hidage. Consequently, the casualties that the defenders could inflict were greater than those that forces equipped with self-bows alone could inflict. These greater defenses required larger not smaller numbers of attackers.

        But you have argued that the very premise of applying models of attackers in the Anglo-Saxon context to a German context is invalid because the Ottonians possessed a significant advanced poliorcetic capability.

        However, the Vikings who posed a threat to the fortresses listed in the Burghal Hidage also had advanced siege equipment available to them. As you will recall from Abbo’s account the siege of Paris, the Vikings in 885 had mantlets, rams, as well as stone-throwing and spear casting engines. This is at least a decade and probably two or three decades before the actual composition of the burghal hidage.

        Your distinction between the capabilities of attackers in siege warfare in England and in Germany on a technological basis is not valid. On a material basis, of course, the Ottonians had vastly more resources than the vikings and Anglo-Saxons combined.

        If there is no technological difference between attacking vikings and attacking Germans, what then is the basis for discounting the viability of the Aris-Bachrach model for Ottonian fortifications?

        It seems very likely that valuable military information such as the appropriate size of garrisons was not discussed by important figures in Wessex and Germany? Otto I, himself, was married to an Anglo-Saxon princess. Did this relationship bring no information in its train about common enemies? Moreover, this relationship was hardly new. Alfred the Great imported a Saxon from the duchy of Otto I’s grandfather to serve as a military adviser.

        However, even if there was no transfer of military science from Wessex to Germany (and vice versa), why should Anglo-Saxon military planners be the only ones who understood the minimum necessary number of garrison troops to hold a fortress of a particular size in light of current military technology? The Ottonians possessed a vast technical expertise in the construction of fortresses. It seems unlikely that they did not have a commensurate skill in determining how best to defend them. In fact, Thietmar of Merseburg provides considerable information on this very point, making clear that the Ottonians understood quite well how to manage the size of garrisons to offer an effective defense of strongholds. His comments on the siege of Lebusa are particularly instructive in this regard.

        In turning to the next issue, I am glad that you agree that Ottonians did storm fortresses, and also that they sometimes captured fortresses simply by the credible threat of storming them.

        But the next point that must be made when commenting on the various tactical options available to commanders, is that you need to think about matters the way that military commanders had to, i.e., you need to work through the options and obstacles that they faced in real time.

        With regard to the question of assaulting with an open formation, it is certainly the case that Ottonian commanders deployed the artillery and other equipment, but in the end taking a fortress by storm, as Henry I did at Gana or Otto I tried to do at Rosstal, meant that men had to advance up to the walls under enemy fire and go over the top with ladders. When they did so, they could be deployed in a number of formations, but the one that would limit casualties the most was the open formation described by Aris and Bachrach.

        If Ottonian commanders massed their troops, other than the ladder carriers who could not be covered effectively with shields, and advanced under the cover of their shields in a testudo (as Arnulf the Bastard’s men did at Rome in 895) or under the cover of mantlets, as Conrad the Red’s men did at Rheims in 947, then they had the possibility of protecting themselves from some arrows. However, they also made themselves much better targets for stone-throwing engines and spear-casters, and also for greater massed arrow fire, because defenders could direct all of their fire against the dense formation. In fact, Conrad the Red’s men were driven back by the hail of enemy missiles at Rheims. There is no easy way to storm a fortress, and all ways give options to the defender that are bad for the attacker.

        Finally, we come to the size of the armies that the Ottonians took on campaigns that were specifically directed toward the capture of one or more fortresses.

        Of primary importance in this providing a ballpark figure for these armies is establishing the size of the forces that the attackers likely had to face when undertaking a major siege operation.

        This brings us back to the question of the number of defenders of particular fortresses. The study of medieval urban demography is a very advanced field that uses a wide range of analytical tools as well as models. If you want to say that the particular demographic studies of places like Mainz are incorrect, it seems to me that the burden is on you to demonstrate the failures in the models used by the demographers. Otherwise, you must either reject modeling as lacking even heuristic value, or you put yourself in the position of saying that you simply do not like their conclusions.

        Unlike modern historians, however, Otto I had to make real time plans on the basis of the information that he had available to him. In the case of the siege of Mainz, Otto I and his military advisers had a good sense of the size of the population of Mainz and its concomitant militia forces, and he also had a very good sense of the size of the professional military households available to his various opponents.

        With these data in mind, Otto had to take a worst case scenario into account when deciding how large an army to mobilize. He could not start his campaign by assuming his enemies were lazy, stupid, or incompetent. He had to assume that they were well organized, would follow the best possible modus operandi, be hell bent on defeating him, and would mobilize the most effective available forces to do so.

        As any competent military commander today will tell you, hoping for the best but planning for the worst is the only way to survive a war.

        In the case of Mainz, Otto’s assessment had to include not only the enemy forces ensconced within the walls, but also the enemy forces lurking within striking distance who might be in a position to relieve the siege. Thus, in order to pose a credible threat to storm a fortress, he need to have sufficient forces to cover both eventualities at the same time.

        Whether this force was 2x, 3x, 4x or more of the number of men in defending force within the walls can be debated. The model used by Aris and Bachrach is the only one available, and uses conservative estimates about the number of attackers being killed. They do this, as I noted, both by offering a low estimate for the number of arrows each archer could launch in a given timeframe, and also by offering the best possible scenario for the attacker to avoid casualties in the advance. Under these circumstances, they conclude that 4-5 attackers per defender is necessary. My colleagues writing military history in the later medieval period argue for similar ratios of attackers to defenders based on their reading of contemporary narrative sources and documents.

        It is possible, of course, to play with the model designed by Aris and Bachrach by assuming, for example, that only half of garrison had bows, or that all of the members of the garrison had bows as well as a substantial portion of the population of the fortress. Support for a variety of models can be found in narrative sources. One might also add to the model figures for the casualties caused by stone casting and spear casting artillery. One might also model the ways in which well trained forces advancing in a testudo formation might be able to protect themselves from arrows, but also taking into account that the very density of the formation made it a better target for massed arrow fire as well as artillery fire. As I said, working with a competent specialist in physics and mathematics can generate a number of interesting possible outcomes. These are not unknowable and certainly are not meaningless.

        If you have another model for estimating the sizes of military forces engaged in large-scale siege operations, one that accounts for likely casualty rates for attackers in light of known variables about the available weapons and the likely minimum number of defenders, then that is one you should share. But if you are rejecting models of this type outright, as you seem to do in your note, then this seems to me to be a methodologically indefensible position.

        For the sake of clarity and simplicity, I chose the model that was available, and applicable, for the reasons I noted above, to the situation in tenth-century Germany. The armies may well have been bigger than those that I suggested, and they may have been smaller if they faced weaker and less well equipped opponents than seems to me to be case based on my reading of the sources, but they could not have been small, i.e. in the range postulated by Reuter or Halsall.

        • Thanks again for a generous and detailed reply! This is making things a lot clearer about where we really disagree, which is certainly not at all points. I agree, for example, that the Carolingian or Ottonian empires at full power could have raised armies in the tens of thousands, and therefore probably did, although we are probably still at odds about how high those tens were and how often that was done. The Carolingian legislation on the general levy, after all, has a whole range of differing lesser possibilities before the immediate and major threat to the patria that meant mobilising all their forces. I am also happy to agree that Wessex and Germany probably, collectively and royally, knew as much about weaponry and tactics as each other, though I don’t think the way you use Asser in the book to make that point is valid—another future post. But that doesn’t matter, because Æthelweard’s contact with the German royal family serves the point just as well if not better. I’m also happy to agree that the figures for the population of Mainz around 1000 that you use in your book as as good as we can get, though I personally would not then go on to publish calculations using them.

          You are right that in general, as that previous sentence implies, I am philosophically against using mathematically-informed modelling. Our starting figures simply aren’t complete or accurate enough, and their errors can almost never be estimated, so that we begin with built-in but unknown error factors which are then multiplied. This old post of mine illustrates my position on this better than I can reprise in a comment here.

          Thus, while as I say I am more or less happy to concede that 30,000 people might be a good guess at the population of Mainz in the year 1000, what we can’t guess is whether that figure was higher or lower in July 953, and by how much, and then the relation of that figure to the actual fighting strength of its defence through to September. You say (p. 232) that the population figure was probably much higher (though it’s not clear from your text why you think so) and that it doesn’t include the people living in the city’s suburbia. It’s certainly most rigorous to start from the figure we have without changing it, but this is already too many variables! You say yourself that population was expanding rapidly throughout the tenth century; that implies that nearly fifty years before 1000 we ought to expect Mainz’s population to have been rather smaller. How much smaller? We have no idea. You reason from the city’s population as reconstructed to a population of able-bodied males of 9,000; but the article on whose basis you do this was using data from C19th Liberia. Is this comparable, given different pathogens, climates and medical technology and practices? We have no idea. In any case, we don’t know how many of these able-bodied men were there or, as you say, the sizes of the professional contingents that were initially reinforcing them, though the presence of those contingents of course means that the 9,000 figure cannot be wholly right to start with. Would the people of the suburbia have joined the defence, or run away, or even supplied the besiegers? We have no idea. What was the casualty rate, and how did the effectiveness of the defending forces drop over the course of the siege, as morale and nutrition diminished? Did they even have enough arrows? We have no idea.

          The figure you construct is plausible, then, perhaps, but it is constructed, and with different assumptions (as above) might have been much lower, or indeed much higher, and it is of course itself derived from an estimation with its own error margins. And then you multiply these accumulated errors to obtain the size of the necessary attacking force, using a multiplier which we are also in debate about and which you concede above could easily be much lower in some circumstances. I’m with my source, Ted Buttrey, on this:

          When we enter on these kinds of calculation, we can be confident of two things. First, the answer will be wrong. Whatever it is, it will be wrong, since it cannot be right—once you are guessing, the number of possible permutations is gigantic.

          So I don’t have an alternative model hidden up my sleeve, because as you must see by now, I doubt their validity to the extent that I would not dare offer one without a very firm and immediately applicable basis in contemporary sources.

          I have, in any case, to wonder why you invoke the instance of Mainz at all, since in the end as you describe (pp. 52, 54) it was not taken by storm anyway, but surrendered almost a year later! Moreover, again as you describe, part of its defending force had managed to escape through the Ottonian cordon only three months in, and after that Otto gave up because his forces couldn’t or wouldn’t maintain the pressure. I suppose, since the city only surrendered much later, that he then reverted to a simple blockade such as I suggest in my post. So whatever guesses were made by planners about necessary force sizes on either side here were evidently wrong or unrealisable! I think Mainz actually supports the argument I make in the post rather better than it proves any of the things we seem to be arguing about.

          So, what that seems to leave where we might progress is: the sourcing of your multiplier for deriving necessary attacking numbers from numbers of defenders in the Bachrach and Aris article; the basis of the mathematical modelling in that article; and the transportability of its conclusion to Ottonian scenarios.

          My stake in the Bachrach and Aris article is, I want to be clear, not their findings about the Burghal Hidage or Anglo-Saxon warfare; it’s simply that both in your book and in “Early Saxon Frontier Warfare” you cite it as the source for your multiplier of 4x or 5x necessary attackers to defenders. Its point about formations, though interesting, is therefore irrelevant to me here except in so far as it gives you basis for that figure. Now, unless I am reading very badly, that figure is not reached in the article. So in some sense, its relevance stops there. The mathematical models involved need separate critique, if they need it at all for my point to be made. But we should still argue about the transportability of the figures, because your position seems to imply that I haven’t understood the article and I think I have.

          In the first place, let’s get my position straight that, as with the example of Mainz above, there are too many unknowns here about the comparability of situations. As you say above,

          “As I said, working with a competent specialist in physics and mathematics can generate a number of interesting possible outcomes. These are not unknowable and certainly are not meaningless.”

          And you may be right but since we cannot know which possible outcome was correct in any given siege of a Burghal Hidage fortress, or which of the outcomes would be most applicable to a generalised Ottonian scenario, if such a thing be possible, surely the exercise must stop there.

          If, as you suggest, we use the model that is available even so, its applicability needs to be demonstrable. You say that you’ve done this but I don’t think that you have. You admit that in the Ottonian situation both sides might have artillery and protective works, and despite your contention above that siege towers are unlikely so early you argue for likely continuity in their use from Roman times to the late tenth-century references that we have in your book (pp. 164-166, esp. p. 164). But the Bachrach and Aris article expressly assumes the contrary, on the basis of the differing quality of the Burghal Hidage forts:

          … it may be suggested that the men who conceived the burghal system were confident the enemy was as unlikely to sap or breach the walls of the least formidable stronghold as of the most formidable.” (p.4)


          “The inability of the Vikings, against whose attacks the burghal system was conceived and developed, effectively and regularly to mount a lengthy siege, tunnel beneath the walls, or breach the ramparts basically left them with the option of storming the defenses and scaling the walls with ladders.” (p. 5)

          So if you’re right above that the Vikings in fact could and did deploy machinery and engines against fortifications, that does not make the figures generated in this article, which assumed that they did not, any more applicable to the Ottonian situation, does it?

          Now, we don’t know how the addition of these factors to the model should be accounted for. Should both sides be allowed an equivalent extra ‘kill rate’ for the effect of their siege engines? Was the Ottonian train especially effective or numerous in this respect? You seem to argue that it was (p. 157). But we don’t, of course, know how much more so in either area than any single opponent, much less an average of all opponents which would, in any case, be wrong in any single instance, because of being an average.

          So: we have a model for a different (and, I should say, very small-scale—Bachrach and Aris deploy no force figure higher than 400, based of course on the Burghal Hidage figures, which you yourself above argue are as comparable to the scale of Ottonian warfare as apples to dump trucks) military situation in which much less weaponry was available to both sides, a half-century earlier, and that model does not give a ratio for attackers to defenders, only plausible casualty rates in the first part—not the whole—of an attack by storm using only ladders on a fortress manned by “farmer-soldiers”. From this, you somehow get a multiplier for attackers compared to defenders when armies one or even two orders of magnitude larger, and sometimes considerably better trained, are in play, using all the weapons and tactics you describe so well but whose effects we can’t factor into the model and whose distribution between the sides we can’t substantiate, a half-century later. I guess that by now you’ll see why I think this can’t stand. I’m willing to go through the Bachrach and Aris article identifying unknowns if it will add anything, but the problem I have with your deductions really starts after their article’s mathematics is finished.

          • One or two secondary sources (that I inconveniently don’t have to hand) when combined, assert that Count Eudon’s contribution to William the Conqueror’s army of seaborne invasion was over a third of the total, comprising 5000 men (1000 levied spearmen, and 4000 professinal soldiers, comprising light and heavy cavalry, short and cross bow men, and axemen) borne on 100 boats. Admittedly, “Ship List” has no information about this, and indeed Duke William’s chaplain William of Poitiers (who remained in Normandy) mentions no Bretons by name. However, the ratio of 50 soldiers on each ship is consistent with other accounts from Brittany.

            Could Eudon have raised 5000 troops? If we estimate a hide as 120 acres and apply the English hidage “rules” to the region under Eudon’s control, which was one third of Brittany, one obtains a capacity of 5000-6000 soldiers. If we credit William of Poitiers comment that there were alarmingly large numbers of Bretons under arms in daily life, then Eudon’s actual military capacity may have been higher, which would have been just as well due to his ongoing conflict with his deceased brother Duke Alan III’s descendants, especially his bellicose nephew Conan II who controlled the other 2/3 of the duchy.

            On the basis of the hidage calculations, Brittany, when united, could have raised some 15000-18000 soldiers for an expeditionary force. Historic Brittany today has very roughly 10% of the population of France, so if we suppose that Otto’s empire in the 900s was of similar size and the same efficiencies applied, then he could have raised 150,000-180,000 soldiers for the purposes of a siege or other act of aggression.

            On at least one occasion when Charles the Bald, as King of West Francia, attempted to invade Brittany, he added to his own troops Saxons and some men from the army of East Francia. A figure of 4000 has been guessed as the size of his invasion force, but even if the populations of the Carolingian empire in the mid 800s was significantly lower than in France and Germany in the 1000s, that number seems too low: how could he have expected to conquer Brittany with so few? The hit and run tactics by the Bretons as Charles’s army proceeded through Maine suggest that they were relying on mobility, not numbers, to win the day. When attacked, the dusciplined Frankish troops formed a defensive shield wall, against which the Bretons launched pilae from horseback, Parthian style. In the final battle, at Jengland, even though they took few casualties while inflicting continual damage, it took the Bretons until the third day to eliminate the Frankish and Saxon forces. All things considered, Charles’s army must have been large.

            • There’s a lot of ‘if we assume’ in there too, but not so much as to affect your overall orders of magnitude in calculation, I think. The problem is of course that we don’t know the basis on which Breton leaders could demand military service. Really, the only polity of the era for which we have even claims about doing this (other than Byzantium, where they are old) is the Carolingians and not only is it often hard to show capitulary legislation actually being carried out but there is a good deal of other capitulary material complaining about the abuses of those who were supposed to enact it! The high figures may be an ideal number, but would presumably have been eaten away at by non-attendance, exemptions official or less so, and of course the relevant leaders’ strategic choices about the cost in men, agricultural labour and supplies, as well as long-term obedience, of calling out the maximum possible force. Unfortunately, those are the figures we can’t reconstruct.

              • I suspect there was less coercion and more enticement: Brittany was comparatively wealthy through its mercantile ventures, which its leaders encouraged. The extant records indicate that Breton dukes and counts paid men good money to fight for them. Conan II’s army of conquest melted away when he died outside Chateau Gontier on 11 December 1066. Alan Rufus personally paid for the splendid attire, equipment, victuals and entertainment of the 200 men of William I’s household guard when they were left to fend for themselves at Camp Beugy near Sainte-Suzanne.

              • Another question occurs to me here, which is what what you’re calling ‘an army of conquest’ was supposed to do in Brittany. Surely the goal of the Carolingians here, as with so many other medieval campaigners, was not actually to occupy the whole of Brittany, garrisoning every town and guarding or demolishing every fortress and so on, but rather to do sufficient damage that it became politically necessary for the enemy to make terms and resume their tributary status. Pitched battle might achieve the same effect but would be much riskier, as Jengland proved. That changes the numbers you need for your campaign rather, if you don’t actually need or expect to defeat the enemy’s collected forces…

  2. I one helped Rutherford Aris get off a tram at the right stop. However good his Latin and Greek were, the arcane German word for university seemed to fox him.

    And on a more serious note: when I was a boy I liked the drawbridges at ancient castles, and the notion that they would allow a sally by the besieged. I had a notion that the besiegers might just dig a huge hole so that the drawbridge would decant the outwards charging knights to their doom. In fact, castles seemed to me to be rather vulnerable to besiegers building their own earthworks.

    I suppose the point was that a castle could let a few men hold out until a relief force could arrive, whereat the besiegers’ earthworks might not be much use unless they, in turn, were on a large scale and two-sided.

    • It is very noticeable, especially in the Anglo-Saxon context in which the Burghal Hidage was written, that the standard move of pursuing a campaign against a Viking fortress in the Anglo-Saxon Chronicle seems to be to raise a counter-fort, and this usually does the work of convincing the enemy to make terms. I suppose that the effect is to provide an unshakeable base for cutting off the besieged fortress’s supply lines. And of course in later eras there’s work to undermine the defenders’ walls. In general, military work in this era involved a lot of digging, I think!

      • Digging to undermine the walls, as at the Siege of Exeter in 1068. (One kind of trench warfare.)

        Alan would have needed dynamite, and lots of it, to undermine Sainte-Suzanne. I have no idea what William’s plan was there, unless he meant to return but keot getting interrupted.

        One of the Chroniclers did write that after the execution in 1076 of Waltheof, whom the English considered innocent and subsequently a saint, all William’s ventures went awry, the implication being that he was now cursed.

  3. Since I picked ancient rather than medieval (let alone early medieval) history for my graduate training, I am even more reluctant to comment. I have only read a few things by the Bachrachs, including the wonderful stirrup article and some nice document-based articles on crossbows. But I would certainly agree that there are plenty of ways to conduct a siege other than a sudden charge or building a full circumvallation with ramps and rams and towers, and that sieges where the attackers lack the men to circumvallate the place are the kind where both sides are usually desperate for cavalry or chariots. (This is implicit in the Amarna texts, where the governors of cities under pressure ask for five or ten or even two chariots, but I would think that the most detailed and accessible case studies are medieval, unless Philon of Byzantium gives a rule of thumb).

  4. Allan McKinley

    One other thing all these arguments tend to miss out is the issue of standard/experience of soldiers. Assaulting a stronghold defended by a smaller number of trained warriors might be a harder ask than taking one manned by larger numbers of peasant levies. There is even a snide comment from either Asser or the Anglo-Saxon chronicler (sorry – on a train so can’t check) about the loss of a West Saxon fort to Vikings that could be read this way, as the defenders were unready for the attack. In terms of actually attacking a fortification the problems of untrained troops would be amplified, especially if they were vulnerable to counter attack without support. The size of armies is one matter but their effectiveness is another in this regard, and it is easy to envisage a case for predominantly small mobile trained forces being the preferred option (and for Ottonians, Al-Andulas or Carolingians having the ability to put together much larger forces of this nature to be fair).

    • Allan: Indeed, a student of Vegetius would have a hard time denying that “in omni proelio non tam multitudo et virtus indocta quam ars et exercitium solent praestare victoriam.” The Roman professional tradition of thinking about warfare, as available to the Greekless in the middle ages, did not see numbers as an especially important quality. How Vegetian early medieval warfare was is of course another question …

    • I’ve always taken that Asser comment to refer to the state of the fortifications, rather than the training of the would-have-been defenders, but I know the one you mean. It’s not Asser at his nicest, really, since he more or less goes on to say, “and that’s what those disobedient to the king get from God”…

      In the Ottonian context one of the problems of the issue you point at here is that there are very variable estimations of Slav forces in terms of training, ability and equipment. One of Bachrach’s more interesting points here is that the Slavs could themselves raise large forces and mount sieges, but he never explores how they can do that on more or less equal terms with the Ottonians despite not having the post-Carolingian logistical inheritance which he sees as vital to the Ottonians’ efforts…

  5. Empires present their own issues: if the Ottonians wished to field 10,000 troops for an assault on, say, a Slavic fortress, would they thereby be compromising their defences in other regions of their realm?

    Going back a century, how many men could Charles the Bald afford to lose to repeated failed invasions of militant border states such as Brittany before these losses compromised his position elsewhere?

    Siege calculations are very interesting: how many troops did Conan II deploy to overrun the northern fortifications of Anjou in a matter of weeks? (These fortified towns had resisted the Normans for two centuries.)

    An 11th century example of a siege that failed due to lack of sufficient manpower is that of Sainte-Suzanne, when William the Conqueror gave his own bodyguard (his best knights and pethaps 200 men) orders to continue the siege in his absence. Orderic Vitalis says they loyally did this for three years (c1083-c1085). The elevated hill castle is formidable enough in itself, but it had 300 defenders, many approaches by which it could be restocked, and was reinforced in the duration by knights from Aquitaine, Burgundy and points in between. The commander and provisioner for the first year of the siege was Alan Rufus, but after a year he handed over command to his deputy, Anvrai le Breton. The besiegers continually lost men during foraging expeditions. William de Warenne was wounded while attempting to storm the castle. Richer L’Aigle and other famed Norman knights were killed, as was Anvrai, before the King agreed to a diplomatic settlement. The wonder is that any of the besiegers survived for so long.

    • Thankyou, that is a useful example both of insufficient forces and of small armies… But one might have thought that William would have known it wasn’t going to be enough, wouldn’t one?

      • Yes, it’s a puzzle why William left his household knights to fend for themselves against such odds for 3 years. He must have been inordinately distracted. What we know:
        (1) Maine had rebelled with support from Anjou, so one major task was to deal with the Angevins.
        (2) His army had reconquered every castle in Maine, except Sainte-Suzanne, which is in Mayenne, so towards Anjou.
        (3) They set up camp at Beugy, 800m from the castle and built their own fortifications there: France nicknames this “the English camp”. In the midst of this, William suddenly decamped with most of his army, except for Alan & co. Historians estimate this was sometime in 1083.
        (4) Orderic Vitalis asserts that Odo of Bayeux, who had been left in charge of England, decided on his own initiative to take troops from many of the garrisons and sail to Rome. This was during the Investiture Controversy, and the rumour was that he intended to obtain the Papacy. But William intercepted Odo at the Isle of Wight in time to prevent the fleet from sailing, arrested Odo and complained that this affair had interrupted his campaign against Anjou and Maine. Orderic seems to imply this was in 1082, but my reading is that it just might have been late in the year or even in early 1083. In the latter case, William may have been referring to his hasty departure from Beugy.
        (5) Odo was imprisoned, if I recall correctly in Rouen, and was not released until William’s deathbed amnesty, and even then most reluctantly, due to pleadings from Odo’s brother Count Robert of Mortain, after William “Rufus” had been sent to England (probably with Alan who had reason to argue against Odo’s release).

        Why didn’t William return after imprisoning Odo? A lot depends on the exact dates, which we don’t know. In 1083 his difficult son Robert “Curthose” allied with the French king against William. Queen Matilda fell ill in summer 1083 and died in November, being interred in Caen. Cnut “the Holy” of Denmark prepared to invade England, so defences had to be kept on high alert through 1085.

        1084 was the year when Alan Rufus’s cousin (on both sides) Alan IV “Fergant” (aged 21 by my reckoning) became Duke of Brittany. William’s attempted invasion was swiftly repelled, his army fleeing in such haste that a large treasure was left behind. In 1086, William sealed peace with Fergant by giving him the hand of his daughter Constance.

        Why Alan was recalled after only a year or so is another puzzle: did William need him elsewhere (perhaps because of the Brittany invasion debacle), did he attend Matilda’s funeral, did he request leave for his own purposes, or was he so attractive a target that his presence was drawing those ambitious French knights who caused so much havoc among the besiegers?

        In 1084 Alan took the opportunity to visit family (perhaps at the family HQ at Lamballe) in Brittany where he and his brother Count Brian (retired) witnessed two charters of their eldest brother Count Geoffrey, one being for the abbey of St Florent at Saumur in Anjou, a frequent recipient of Breton donations. (This shows that wartime hostilities were no bar to expressions of piety.)

        • It sounds to me almost as if the siege had become a matter of honour rather than of strategic necessity. But in that case, that is almost too valuable a force to commit to it…

          • Allan McKinley

            See my earlier comments about the standard of troops? William’s household knights were likely to be very good soldiers, so an assumption that as a body they could pursue a siege was presumably realistic. If as small a force in proportion to the defenders as here envisaged then it might be significant that they still apparently engaged in blockade and assault, albeit with limited success.

    • David Bachrach

      The issue of the scale of resources available to the Ottonians is an important one when considering how large an army Otto I or others could mobilize, and whether other forces were available for local defense or for campaigns in different regions. The comparison of Conan of Brittany or even William the Conqueror to Otto I is a bit like comparing apples to dump trucks. To take just one metric, the number of dependents living on royal fiscal estates during the reign of Otto I was equivalent to the entire population of Wessex during the reign of Alfred the Great. To give just one other example, Henry I, Otto I’s father, constructed a systems of fortifications along the eastern frontier in the space of a decade that was three times larger than the system outlined in the burghal hidage, with concomitantly large garrisoning requirements. Otto I had resources of an order of magnitude comparable to Charlemagne, who routinely deployed substantial armies in two or even three different theaters at the same time.

      • While there are probably things on which I can be brought to agreement with you, Professor Bachrach, this:

        “To take just one metric, the number of dependents living on royal fiscal estates during the reign of Otto I was equivalent to the entire population of Wessex during the reign of Alfred the Great.”

        shows why it will never be total. This comparison might, indeed, be about right, but since we have neither a full list of royal dependents on those fiscal estates of Otto I nor a full census of the population of Wessex under Alfred (or even a clear idea of what he would have counted as Wessex), but must obtain the numbers by multiplying up from the fragmentary information we have without any sound ways to check our assumptions about the multipliers, the potential error must be enough to vitiate the comparison. Different assumptions about the multiplication could easily make either population two-thirds of the size of the other and be equally well defended, I imagine!

  6. Now that Professor Bachrach has made an appearance, it’s time to reveal the conversation that’s been going on behind the scenes. I received mail from Professor Bachrach almost as soon as the post went up, endeavouring to correct me on quite a number of issues. I, as you will imagine, replied in kind, and both of us have agreed that whatever we think, it would be more useful for both of us to debate where people can see and gauge our points for themselves. So, the those first e-mails are now dropped in as the first thread of comments above, with suitably faked From: lines to try and make it clear who was writing, and I have attempted to come back for another response!

  7. I am an amateur when it comes to history but not when it comes to mathematical modelling. I shall now reveal to you Dearieme’s First Law of Mathematical Modelling.

    The greatest weakness of mathematical modelling stems from modellers’ propensity to fall in love with their models.

    This is not a jibe aimed at those of you arguing here, nor is it a defence of jeering at the very attempt to model events, it is a lesson forged in reading many papers and, particularly, attending many seminars.

  8. David Bachrach

    Dear Jonathan,

    I am certainly appreciative of the fact that you agree with me that the Ottonians could and occasionally did mobilize armies in the 30,000 man range. In fact, it is this point about army sizes (and also fortress building) that drives my larger argument generally about the scale and sophistication of the Ottonian royal administration, and which animates my non-military investigations of the period. If you find yourself being accused of being a maximalist, I apologize in advance.

    However, I also have the feeling that I ought to give Laocoon’s warning.

    The hard point of our disagreement clearly rests on our different views of the value of mathematical modeling. But in rejecting mathematical modeling, it seems to me that you have no basis, other than my brilliant rhetorical skills, for accepting that the Ottonians could or had any reason to mobilize such large armies. In fact, without modeling, it seems to me that we are the position of making any kind of assertions that we wish based on a priori assumptions about the nature of the early medieval world. To be clear, in my view models need to depend on a sound foundation that is based in the material reality of the period, and also whatever sound numbers that we can garner from the sources. But with this solid grounding, we can, in my view, come much closer to the reality of societies for which we lack systematic data, by modeling than we can if we choose to reject this tool. Obviously, I am much more optimistic than you are about the soundness of these starting positions and consequently less concerned than you are about compounding errors.

    On a different subject, I noted your discussion of the Slavic polities. It is an unfortunately reality that we lack virtually any discussion of their military organization from their own point of view in texts. For written discussions, we must rely on the Frankish and Ottonian accounts, as well as the remarkable report by Abraham ben Jacob from the 960s or 970s. However, this relative paucity of sources has been greatly augmented by the enormous volume of information produced through excavations over the past 30 years. From these, and particularly the new dating of fortresses in Poland through advanced dendrochronological techniques, a new picture has emerged of very sophisticated polities, in which the rulers had considerable control over surplus production and were able to mobilize these resources for military purposes. It is now argued by the archaeologists working in this area that in Poland, Miesco likely constructed dozens of fortresses, each of which were supported by a network of villages, in a very brief period between the late 950s and the mid 960s. The pattern of construction and the building techniques mirror those of the German system constructed by Otto I in Polabia after the battle of the Recknitz in 955 up through the formal peace agreement between Otto I and Miesco in the late 960s. Unfortunately, I did not become aware of much of this newest research until after completing the book. A similar pattern is discussed by Eric Goldberg with regard to the Moravians in his biography of Louis the German, but again we lack written evidence from the Moravians, themselves.

    All best,

    • Indeed, I’ve been dimly aware of the Polish fortress archaeology thanks to conference presentations; it’s quite impressive. There are some theories about them I find implausible, but the new data is certainly potentially game-changing.

      As to your brilliant rhetorical skills, well, I think this is why I proposed reversing your argument in my initial sally. I find your position about the large armies plausible, with suitable room for argument about how large of course, because of what you can show in your book about the supporting cultural and fiscal apparatus that looks as if it’s to do with such formations. You have better evidence for people acting as if there were widespread military obligations upon the nobility and populace than, without the mathematical models which I distrust, you do for actual army numbers. There are also aspects of what the narrative sources have to say, less about the numbers than about the wide spread of areas from which the Ottonians could sometimes draw forces, that also makes it plausible to me that large-scale levies were possible at times in this context. I’m not at all sure that we can safely arrive at how large, but I’m happy enough to be in the same order of magnitude as you prefer for at least the Ottonians and Carolingians. I didn’t agree to the 30,000 figure, though, not for an army! That I only mentioned as being the root figure for your estimates about forces needed to take Mainz in 953, and I agreed it was plausible only as a population figure for the city in 1000. I’m not saying it’s impossible for an army, I just don’t want to be on record as having stated it. David Metcalf gave that figure as an upper limit for coins producible by a die once, and it’s been used in calculations by incautious people ever since…

  9. Pingback: What if Widukind was wrong about warfare? | A Corner of Tenth-Century Europe

  10. An army of intimidation that could not hold its own in the open? The Bretons were known for their cavalry and their archers, so it wasn’t likely that an invader could avoid a sticky confrontation for long. We have the benefit of hindsight, but Charles the Bald cannot have augmented his West Frankish army with troops from Germany just for a suicide mission. He must have seriously expected to win in the field. In your scenario he would have hoped at least to fend off any attacks long enough to reach his targets and conduct the intended punitive actions.

    At their peak, the Franks occupied Vannes and built a fort even deeper into Lower Brittany, before their ascendancy there began to wane. By the reign of Charles the Bald, they had lost all their forts in the Breton March. The effect of the Battle of Jengland was that they conceded Mayenne and western Anjou, surely not the intended outcome. Later efforts were even less effective, with the loss of the Cotentin and other territories. In King Solomon’s reign, the Franks paid what were effectively fines for their audacity and even gave hostages.

    In the background of these international affairs, both sides were hiring Vikings to supplement their raiding parties. (For example, Solomon hired the famous Hastein.) In the early decades of the 900s this rebounded badly on both sides when the Vikings decided they wanted not just gold and silver but also dominion over the land.

  11. Pingback: A somewhat unexpected interpretation of Asser | A Corner of Tenth-Century Europe

  12. Pingback: A Theory under Siege — A Corner of Tenth-Century Europe | Tome and Tomb

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.