A Jewish garrison town in Carolingian Catalonia?

Please forgive a gap in posting. On the 4th started the biggest conference in a medievalist’s calendar, and I was running sessions on the first day; 29th and 30th also had a different conference in them, and a family house-move needing my driving fell between the two events. The week before that had been the finalists’ marking deadline, so I’d got very little ready for either conference till then, and by the third day of the conference this week I felt ill and, when tested, turned out to have caught Covid-19. Since then I’ve mainly been asleep, sweating feverishly or otherwise useless in our spare room. So it’s not been full of blogging opportunities. But all this time I have been trying, now and then, to finish this for you, refreshed over many weeks now from an old draft. The title of the post is a conscious riff off Arthur Zuckerman’s infamous and, erm, let’s say ‘disputed’ book A Jewish Princedom in Feudal France, but I am actually reacting here to a reaction to it, published in 1980, by Bernard Bachrach of whom we were lately talking.1 I write here because although I can deconstruct Bachrach’s paper and see many things wrong with it as argument, I can’t actually dismiss it all out of hand without better access to the evidence than I have, and that frustrates me. So when I read it in 2019, I wrote the beginnings of this to try and work it out. Since then I made the effort to get hold of some important extra evidence that allowed me to write the closing section, and now at last I inflict it all upon you.

Spine of Arthur Zuckerman's A Jewish Princedom in Feudal France, 768-900 (New York City NY: Columbia University Press 1973)

Spine of Arthur Zuckerman’s A Jewish Princedom in Feudal France, 768-900 (New York City NY: Columbia University Press 1973), image by Dranoel26own work, licensed under CC BY-SA 4.0 via Wikimedia Commons, because this is a work which has its own Wikipedia page

So, for those unaware of it, Zuckerman’s book is a tour de force of medievalist imagination from 1965, in which some fourteenth-century references to a ‘king of the Jews’ in Narbonne are built up, with the aid of anything that could possibly be used as evidence even when it’s really not, into a concession agreed by Charlemagne and the Caliph of Baghdad to establish an autonomous Ashkenazi principality on Narbonne that was eventually shut down by various interests colluding with the pope in 900. Everything in the area gets wrapped into this theory, to the extent of the line of Saint Guilhem being figured as Jewish because of the reported size of one of their noses (says Bachrach, anyway).2 The most generous reviews of this book thought it might, just, have shown that the Jewish ‘king’ of Narbonne was a real dignity, of rather uncertain nature, in what certainly was a city with a big Jewish community in it.3 Bachrach didn’t even accept that much, but in this chapter he performs a clever move and, using the credibility gained from the fact that he is critical of Zuckerman, proposes a different understanding of Jewish presence in the Midi that has nearly as many problems, even if it’s less ambitious. The logic is quite complex, however, and needs expounding (and exploding) step by step. It goes like this:

  1. Among the considerable evidence for a Jewish presence in Narbonne, we find in the Visigothic-period Historia Wambae a reference to the Jews of that city expelling the local authorities in support of the rebellion of Duke Paul of the Tarraconensis in 673. For Bachrach, that shows they could muster armed force. Admittedly, that rebellion was unsuccessful, but though the Jews were expelled they were subsequently allowed to return.4
  2. When King Pippin III of the Franks took Narbonne in 759, the populace were induced to surrender by a guarantee that they would be allowed to retain their own law. Bachrach argues that this concession would have thus reinforced the Jews in their local position.5
  3. A letter of a Pope Stephen is recorded complaining to Archbishop Aribert of Narbonne about all the concessions Pippin and then Charlemagne made to the Jews in Narbonne and saying that Aribert needs to roll them back as soon as he can. Since there is no other record of this archbishop, or anyone of that rank in the see of Narbonne until the tenth century, this has usually been taken to be a forgery; Zuckerman, indeed, connected it to the end of his ‘princedom’ in 900. Bachrach rehearses these arguments, agrees the letter probably can’t be accepted, but somehow it remains in his argument as support for a Carolingian generosity to Narbonne’s Jews.6
  4. Since we have militarised Jews at Narbonne in 673 (at least per Bachrach) and an assurance that in 759 the Jewish importance in Narbonne would have been protected (per Bachrach), we can now introduce a third element, the service of all free men in the Carolingian army that is demanded by various Carolingian capitularies. From that we can, or at least Bachrach can, conclude that the still-militarised Jews of Narbonne would have been among the troops subsequently deployed in campaigns on the Spanish March.7
  5. In one of these campaigns, in 798, as readers of this blog will know, the old fortresses of Casserres de Berguedà, Ausona and Cardona were reactivated by a Count Borrell. Ausona is the odd one out here as it had been a city, as it would again become. However, Bachrach observes, by 900 (recte 906), the newly emplaced bishop of Osona could complain that there were no Christians in his diocese, and there is also apparently a Hebrew responsum from a rabbi in the Middle East to an Iberian-peninsula contact of his of c. 850 saying that there are ‘no gentiles’ in Ausona. The explanation is of course obvious, to Bachrach: no gentiles, no Christians, because the town had been settled by the Jews of Narbonne as a regular Carolingian garrison.8

Now, you can probably tell already that I don’t buy this. I’m not against the idea of Jewish settlement in the Spanish March, at all: it explains a few place-names, like Judaigues in Besalú where the comital family of Barcelona later had land.9 Moreover, there is fairly solid evidence of Jewish landholding in the south of France in this period, including someone Jewish whose lands had been encroached upon appealing directly to Emperor Louis the Pious and having his case upheld, as well as the various rather earlier or later evidence for a Jewish presence in Narbonne, in Barcelona and in Girona.10 My credulity runs out, however, before being able to accept a Jewish military garrison town that no source describes as such.

View over Barcelona looking towards Montjuïc

The most obvious place-name mentioning Jews on the Spanish March is however probably the crown of Barcelona, Montjuïc! Image by Fabio Alessandro Locatiown work, CC BY-SA 3.0, Wikimedia Commons

Basic lack of positive evidence isn’t the only issue here, either. Every one of those steps above has its own problems, which I should set out.

  1. The Historia Wambae, of course, reports the suppression of autonomy at Narbonne, and says nothing about the terms on which the Jews were allowed to return; one might imagine that it was not swords in hand, although Bachrach just waves at the lack of evidence for Wamba having cared about Jews very much and assumes he’d have been cool with that.11 But also, importantly, the Historia was written by documented polemical anti-Semite Bishop Julian of Toledo, and so it’s not a given that these Jewish actions are even historical, rather than a way to blacken the name of the rebel Paul with people whom Julian would have seen as distasteful and unholy associates. Blaming the Jews for the fall of cities is a good strong tradition in this era, after all.12 So to get from that to an organised Jewish political faction in the city, with regularised military capacity, is what you might call an over-reading of this source. What Bachrach suggests is not impossible, but it’s a long way from being what the source says and there are reasons to mistrust what the source says on such matters. This will be a repeated theme in what follows…
  2. Next, whatever position the Jews held in Narbonne in 673 then needs to have been preserved eighty-six years until the Frankish conquest, and of course that period also contains the Muslim conquest of the city in 721 or so. It is likely that that materially improved the situation of the Jews in the city, but it is, I’d have thought, extremely unlikely that they would have been allowed to continue to bear arms, if that was actually something they had been doing!13 Bachrach simply doesn’t mention the Muslim conquest, which gets him round that particular problem, but doesn’t do anything to remove it.
  3. If, nonetheless, we somehow still wind up at 759 and the Carolingian capture of Narbonne with a powerful Jewish faction in the city with an old right to bear arms, the Visigothic Law that it seems reasonably safe to say that King Pippin III guaranteed at Narbonne in 759 actually pretty much denies Jews any civil rights whatsoever, in an accumulation of legislation from the final years of the Visigothic kingdom that has attracted a lot of scholarly attention.14 It may be easy enough to imagine that those laws were never enacted or had been repealed—Jews are still attested in these territories, after all, however thinly—but to guarantee or restore them and the old Jewish privileges they deny at the same time would take a level of double-think we don’t usually attribute to the Carolingians. It’s certainly not inherent in what the sources actually say, and in any case it requires an assumption of prior continuity that is hard to credit given the likely disruptions to it which Bachrach doesn’t mention.
  4. The council record of 906 in which Bishop Idalguer of Vic says there are no Christians in his diocese is clearly inaccurate; we have land-charters from people in his diocese going back to 880, and in fact we have Christian burials from the city that probably belong to this period.15 It is also, however, spurious as it stands, having been inserted into a record of 788! This becomes more comprehensible when one realises that the plea is made as part of an attempt to be rid of a levy up till then paid by the new bishopric to the metropolitan of Narbonne. What Idalguer was supposed to be saying, in other words, was, “I don’t get enough tithe to afford this.” A certain amount of exaggeration is therefore easy to understand. Less easy to understand is how he wouldn’t mention that his episcopal city was a Jewish military colony, however; I feel that also might have made a good part of such a case. Arguments from silence are always more difficult, but this is really quite a loud silence. The record does talk about the difficulties the area had faced because of ‘the infestation of pagans’, but that, pace Bachrach and Zuckerman both, seems much more likely to refer to the Muslim conquests, in the same basically fictive way that other tenth-century sources from this are wont to do when seeking to justify a land claim.16 These were educated Christian clergy to whom Jews cannot have been unfamiliar (though if they were, it wouldn’t do much for Bachrach’s argument that there was a town of them right next door). Christianity has been dealing with the Jewish religion since its birth out of it, and churchmen knew that Jews were not pagans, whereas Muslims remained in a rhetorical and intellectual space where that could still be alleged.16bis
  5. Last of all, but important, another thing that Bachrach doesn’t mention, like the Muslim conquest of Narbonne, is the 826–827 rebellion on the March under the mysterious Aizó, which took Ausona out of Carolingian control. We don’t in fact know that that control was ever regained, at least before the area was brought back under the authority of Count Guifré the Hairy of Urgell and Barcelona in the 870s; it has been suggested that the town was completely deserted and it has been suggested that it became a Muslim fortress allowing a series of raids into the Frankish interior that seem to have stopped in the 850s.17 Either of these cases might be a pretty good explanation for why a Hebrew letter of 850 might say there were no gentiles there, but Bachrach’s arguments rely on continuity, a long long continuity right the way from Narbonne 673 to Ausona 906, so unsurprisingly, as with the Muslim conquest of Narbonne, he doesn’t mention this rebellion. Frighteningly, Zuckerman’s case actually fits better here, as he saw a reimposition of Jewish rule in this area c. 852 under ‘Abbasid pressure on the Carolingians, but that would wreck Bachrach’s argument, so he ignores it and in this case, that’s probably fair enough!18

So at the end of this, we have a very long chain of over-read sources, which, if every one is accepted, can indeed be lashed together in some dreadful Heath-Robinson fashion that allows one to bridge the gap between 673 and 906, but whose lashings are rotten at every join, and which has to reach over some really quite serious discontinuities that Bachrach ignores. It’s perhaps not completely surprising that I only lately discovered this paper because I’ve only ever seen one citation of it despite working on the county that grew up around reoccupied Ausona; there really is no reason to take this theory seriously, and people mostly haven’t.

Cathedral of Sant Pere de vic seen from the Riu Gurri

The cathedral of Sant Pere de Vic, seen from the Riu Gurri, photo by Enfo (own work), licensed under [CC-BY-SA-3.0] via Wikimedia Commons

And yet, there is apparently this rabbinic letter… The letter is the one piece of this puzzle I can’t point at and show Bachrach doing bad history with it, simply because I can’t read Hebrew. There are so many things that could be wrong with it: its date, the identification of the place-name it uses, its basic authenticity… but if it is what Bachrach says it is then I can’t ignore it. So I reluctantly picked up Zuckerman and, actually, he gave a lot more information. Firstly, we learn the name of the relevant rabbi, Natronai Gaon of Sura. He was based in Qayrawān in what is now Tunisia, and was consulted on several occasions between 853 and 868 by Jews in what Zuckerman insisted on rendering as Ispamia, and one of his letters of advice went to, “the town Ausona (Al-Osona) bordering on Barcelona County”.19 Zuckerman explained that hitherto this had been rendered as Lucena by scholars of Natronai’s letters, but preferred Osona because of the other evidence Bachrach would later repeat. One thing that Bachrach does not repeat, however, is that the letter also advises the Jews of the town not to buy cattle, fish or flour if the market day falls on a Jewish holiday, which tells us pretty clearly that the Jews were not organising the market or it presumably wouldn’t ever have done that thing.

Arthur J. Zuckerman, A Jewish Princedom in Feudal France, 768-900 (New York City NY 1965), p. 318

Arthur J. Zuckerman, A Jewish Princedom in Feudal France, 768-900 (New York City NY 1965), p. 318 and nn. 5-6…

Arthur J. Zuckerman, A Jewish Princedom in Feudal France, 768-900 (New York City NY 1965), p. 319

… and p. 319 & nn. 6-8.

Now, Zuckerman’s notes are not clear (as you see above). They are anchored to strange places in the text, too, making it less than easy to see what note is supposed to cover what assertions. But it seems that we’re dealing here with pp. 318-319 nn. 6-8, in which as you see he prints what I suppose is some of the relevant Hebrew and various references, including to one German translation. That, thankfully, is on the web, and from that I can render the German as follows (badly):20

“Non-Jews often bring in oxen and rams from outside the city on Sunday and Friday, and from time to time the market day falls on a Jewish holiday; should one then buy from the non-Jews? – We decide: One should not buy on a holiday, and not fish or flour either. – And what you have asked: ‘On what ground?’ – Answer: Because Lucena is a Jewish town and has a very great number of Israelites – may the Eternal, the God of our Fathers, multiply them!21 – There are there almost no non-Jewish inhabitants, so surely the objects for sale are brought chiefly for the use of the Israelites; if non-Jews sometimes also find themselves there on market day, surely their numbers are vanishingly small against the majority Israelite population. Were this even in Córdoba, where the seat of government is, but where the Israelites are in the majority over the Arabs, there would be fear that the Israelites would be attracted to market more than would otherwise be done; how much more in a city like Lucena!”

Now, from this lots of things arise. Firstly we see that the German translators, Winter and Wünsche, assumed that the place concerned was Lucena, but we’ve already seen how Zuckerman headed that off, and I have to say that if he was right about the Hebrew, of which of course I’m no judge, then Osona seems more likely. Let’s assume it is for now, but that doesn’t end the questions by any means. Winter and Wünsche also did not offer much help in finding this text; they reference only Warnheim’s edition, as given by Zuckerman in the notes above, and say nothing about where we have this letter, from when, what its transmission is and so on.22 Now, all of that stuff could be really quite crucial in the interpretation of this letter; did its copyist likely have an idea what it was really about, and if not, what might he or she have corrected it to? Further inspection reveals that Warnheim’s edition is actually in Hebrew, with a German subtitle, and that Zuckerman’s helpful transliteration of its main title is not what’s actually on the title page – and neither is Winter and Wünsche’s, so even finding it may be beyond me, let alone reading it. I don’t suppose anyone else is able to help here? Manchester apparently have a copy…

All the same, if the text in question is what either of these writers say it is, i. e. a letter from a near-contemporary well-informed about Andalusi matters, I have at least to consider it. But even from the German, some important things emerge which neither Bachrach cares nor Zuckerman cared to mention (though the former Zuckerman at least implied).

  1. It’s clear that wherever this town was, the Jews were a majority there, a substantial one indeed, but not the only people present. At least, the market provision makes it seem otherwise, and Winter and Wünsch translated the Hebrew that Zuckerman renders, “Al-Osona is a Jewish place without gentiles” with an all-important qualifier, “gar”. The line that both Bachrach and Zuckerman also quote, “there are no gentiles in Osona”, is not actually in this source, and Zuckerman’s notes, once gleaned, say it’s “perhaps by the same author” (p. 319 n. 6) but cites it from a different edition with no further details.23 Again, help getting at this would be lovely!
  2. Much more important, though, is the reference to Córdoba, because that shows that Rabbi Natronai Gaon believed this place ‘al-Usuna’ to be in al-Andalus, under Muslim rule. He must have done, because that was the government whose seat Córdoba was! And that changes the picture rather.

Wherefrom follows a rethink. Around 850 is actually a bad time to see Vic as having been in Carolingian hands, as already discussed; it had certainly been in pro-Muslim ones only 24 years before and is not recorded in Christian ones again till 885 (though late 870s is likely).24 And while we can ignore some of the Christian reports that Jews let enemies into Christian cities, so much more easy to bear than Christians actually having lost them, we maybe need to consider Arabic reports that sometimes local Jews were put in charge of recently conquered towns; the Egyptian historian Ibn al-Athir says that the conquering general Mūsa ibn Nusair did this in Seville, for example.25 What if Vic was such a place? That is, maybe when the Muslim army arrived in 827 they took the place over, but installed a Jewish colony there rather than settle it themselves. Then Vic would indeed be a Jewish garrison town, but for the Muslims, or, probably more likely, a Jewish town with a Muslim garrison. That might be what this source is actually reporting!

Now, I would want a lot of those vital details about source transmission and indeed identity in hand before I started seriously proposing that last thing. But both Bachrach and, before him, Zuckerman just left these details out because they didn’t fit their respective wild hypotheses. I hope I’ve shown that Bachrach’s hypothesis has to be discarded whatever the results of this enquiry should be; but there could be an almost equally surprising alternative to their ideas derived from the same sources, and more easily I’d say, which neither of them for some reason wanted to discover. It makes you wonder, doesn’t it, about the preoccupations which drive our enquiries…

1. Arthur J. Zuckerman, A Jewish Princedom in Feudal France, 768-900 (New York City NY 1965, repr. 1972); Bernard S. Bachrach, “On the Role of the Jews in the Establishment of the Spanish March (768–814)” in Josep M. Solà-Solé, S. G. Armistead & Joseph H. Silverman (edd.), Hispania Judaica: studies in the history, language and literature of the Jews in the Hispanic world, Estudios 2 (Barcelona 1980), 3 vols, I pp. 11-19, repr. in Bernard S. Bachrach, Armies and Politics in the Early Medieval West, Collected Studies 405 (Aldershot 1993), chapter XV, online here from the reprint.

2. On inspection, this is less racist and more crazy than Bachrach, “Role of the Jews”, p. 12, makes it sound; Zuckerman notes the name Naso given to Bernard of Septimania as a pseudonym by Paschasius Radbertus, in his polemical diatribe the Epitaphium Arsenii (Zuckerman, Jewish Princedom, p. 263), which Calmette explained as being a reference to a big nose, but which Zuckerman in fact sees as the ancestral title Nasi born by his alleged Jewish princes. Even in his critique of others, therefore, Bachrach doesn’t really represent what his source says accurately.

3. Nahon Gérard, “Arthur J. ZUCKERMAN, A Jewish Princedom in Feudal France, 768-900” in Annales : Économies, Sociétés, Civilisations Vol. 30 (Paris 1975), pp. 363-364; for the wider background at Narbonne see with more safety Jean Régné, Étude sur la condition des juifs de Narbonne du Ve au XIVe siècle (Narbonne 1912), one of several secondary sources that Bachrach uses rather than cite actual evidence for Jewish presence. After the discoveries of the previous note, one may justly wonder whether checking these would actually back up his points at all or if these citations would also turn out to be misread.

4. Bachrach, “Role of the Jews”, p. 14, citing two chapters of Régné and himself, “A Reassessment of Visigothic Jewish Policy 589-711” in American Historical Review Vol. 78 (Washington DC 1973), pp. 11-34, repr. in idem, Armies and Politics, chapter XI, pp. 26-27, rather than the actual source, Hist. Wamb. c. 5 (he says there). This is now available as Joaquim Martínez Pizarro, The Story of Wamba: Julian of Toledo’s Historia Wambae regis (Washington DC 2005), on JSTOR here.

5. Bachrach, “Role of the Jews”, pp. 13-14; the source is the Annals of Aniane, which are printed in Claude Devic and Jean Vaissete, Histoire Générale de Languedoc avec les Notes et les Pièces Justificatives, ed. by Edouard Dulaurier, édition accompagnée de dissertations et actes nouvelles, contenant le recueil des inscriptions de la province antiques et du moyen âge, des planches, cartes géographiques et des vues des monuments (Toulouse 1875), 16 vols, vol. II, online here, col. 7.

6. Jacques-Paul Migne (ed.), Anastasii Abbatis, sanctæ Romanæ ecclesiæ Presbyteri et Bibliothecarii, opera omnia: editio præ aliis omnibus insignis, ad fidem manuscriptorum codicum et juxta probatissimas editiones expressa, Blancsini nempe Romano-Vaticanam, quod Librum Pontificalem, Mabillonii, Cardinalis Maii, etc., etc. Accedunt Stephani V, Formosi, Stephani VI, Romani, Pontificum Romanorum; Erchemberti Cassinensis monachi, Angilberti Corbeiensis abbatis, S. Tutilonis Sangallensis monachi, Grimlaici presbyteri, Wolfardi presbyteri Hasenrietani, Anamodi Ratisbonensis subdiaconi, Scripta vel scriptorum fragmenta quæ exstant. Tomum claudit Appendix ad Sæculum IX, Patrologia cursus completus series latina CXXIX (Paris 1879), 3 vols, vol. I, online here, col. 857; Bachrach discusses this and its problems over “Role of the Jews”, pp. 12-13 n. 6, in which he both accepts and rejects the arguments for a tenth-century date before using it as straightforward evidence for Pippin’s granting of land to Jews p. 14 n. 9.

7. Bachrach, “Role of the Jews”, p. 16, “The count of Narbonne, whose military contingent surely had a substantial proportion of Jewish allodial landholders among its members…”, with n. 16 there providing cites only for Jewish military service three centuries before or five centuries after, both in other countries.

8. For the refortification see ‘Astronomer’, “Vita Hludowici imperatoris”, ed. & transl. Ernst Tremp in Tremp (ed./transl.), Thegan: Die Taten Kaiser Ludwigs. Astronomus: Das Leben Kaiser Ludwigs, Monumenta Germaniae Historica (Scriptores rerum germanicum in usum scholarum separatim editi LXIV (Hannover 1995), pp. 278-558, online here, cap. 8; for the council of complaint, see Ramon Ordeig i Mata (ed.), Catalunya Carolíngia IV: els comtats d’Osona i Manresa, Memoòries de la Secció històrico-arqueològica 53 (Barcelona 1999), 3 vols, vol. I no. 75; for the letter, see below.

9. Judaigues occurs in Santiago Sobrequés i Vidal, Sebastià Riera i Viader and Manuel Rovira i Solà (edd.), Catalunya Carolíngia V: els comtats de Girona, Besalú, Empúries i Peralada, rev. by Ramon Ordeig i Mata, Memòries de la Secció històrico-arqueològica 61 (Barcelona 2009), 2 vols, doc. nos 312 & 523 at least, and I think at least one more, but can’t check right now given my situation.

10. Jewish landholders appealing to Emperor Louis the Pious in Devic & Vaissete, Histoire Générale de Languedoc, II, Preuves: chartes et diplômes, no. 97; for wider context see David Romano, “Els jueus de Barcelona i Girona fins a la mort de Ramon Borrell (1018)” in Federico Udina i Martorell (ed.), Symposium Internacional sobre els Orígens de Catalunya (segles VIII-XI) (Barcelona 1991-1992), 2 vols, vol. II , pp. 123–30, online here.

11. Bachrach, “Visigothic Policy”, p. 27, with only secondary references.

12. On Julian see Abdón Moreno García and Raúl Pozas Garza, “Una controversía judeo-cristiana del s. VII: Julián de Toledo” in Helmantica Vol. 53 nos 161–162 (Seville 2002), pp. 249–69, online here, and on Visigothic anti-Judaism more widely Rachel L. Stocking, “Early Medieval Christian Identity and Anti-Judaism: The Case of the Visigothic Kingdom” in Religion Compass Vol. 2 (Oxford 2008), pp. 642–658, DOI: 10.1111/j.1749-8171.2008.00087.x; for the trope of Jews causing the fall of Christian cities to invaders, see among many other instances Janet L. Nelson (transl.), The Annals of St-Bertin, Ninth-Century Histories 1 (Manchester 1991), s. a. 852.

13. Norman Roth, “Dhimma: Jews and Muslims in the Early Medieval Period” in Ian Richard Netton (ed.), Studies in Honour of Clifford Edmund Bosworth (Leiden: Brill, 2000), 2 vols, vol. I, pp. 238–266, on Academia.edu here.

14. S. P. Scott (transl.), The Visigothic Code (Forum Judicum), translated from the Original Latin, and Edited (Boston MA 1910), online here, XII.ii.3-18 & iii.1 & 3-28; for discussion, as well as the works in n. 12 above see Bat-Sheva Albert, “Les communautés juives vues à travers la législation royale et ecclésiastique visigothique et franque” in John Victor Tolan, Nicholas De Lange, Laurence Foschia & Capucine Nemo-Pekelman (edd.), Jews in Early Christian Law: Byzantium and the Latin West, 6th‒11th centuries, Religion and Law in Medieval Christian and Muslim Societies 2 (Turnhout 2014), pp. 179–193, online here.

15. Christians in Osona before 906 in any of Ordeig, Catalunya Carolíngia 4, doc. nos 1-74, really; for the burials see Immaculada Ollich i Castanyer, “Dos exemples d’arqueologia medieval al nucli urbà de Vic: la casa de la Plaça de Dom Miquel i la necròpolis del Cloquer” in Ausa Vol. 10 nos 102–104 (Vic 1982), pp. 375–385, online here, and Joan Casas Blasi, Anna Gómez Bach, Raquel Masó Giralt, Imma Mestres Santacreu & Montserrat de Rocafiguera Espona, “Ciutat de Vic: darreres intervencions i línies de recerca” in I Jornades d’Arqueologia de la Catalunya Central: Actes. Homenatge a Miquel Cura, Publicacions d’Arqueologia i Paleontologia 14 (Barcelona 2012), pp. 220–224, online here.

16. The Church history background is set out in Élie Griffe, Histoire religieuse des anciens pays de l’Aude (Paris 1933), 1 vol completed, online here, pp. 246-250; even now this is a tremendously perceptive and thorough book and I wish he’d finished the rest. Bachrach cites it, “Role of the Jews”, p. 17 n. 22, because it establishes a 906 date for the council text, but otherwise ignores what Griffe says was going on. For the trope of Muslims as pagans here, see Jonathan Jarrett, “Caliph, King, or Grandfather: Strategies of Legitimization on the Spanish March in the Reign of Lothar III” in The Mediaeval Journal Vol. 1 no. 2 (Turnhout 2011), pp. 1–22, DOI: 10.1484/J.TMJ.1.102535, at pp. 15-16; cf. Bachrach, “Role of the Jews”, p. 17 and Zuckerman, Jewish Princedom, p. 319.

16bis. On Christianity and Judaism, an interesting range of perspectives is to be found in Adam H. Becker and Annette Yoshiko Reed (edd.), The Ways That Never Parted: Jews and Christians in Late Antiquity and the Early Middle Ages, 2nd ed. (Minneapolis MN 2007) , though goodness knows there are others. For Muslims as pagans, see John Victor Tolan, Saracens: Islam in the Medieval European Imagination (New York City NY 2002), pp. 105-134.

17. Imma Ollich i Castanyer, “Roda: l’Esquerda. La ciudad carolingia” in Jordi Camps (ed.), Cataluña en la época carolingia: arte y cultura antes del románico (siglos IX y X): 16 diciembre 1999 – 27 febrero 2000, Museu Nacional d’Art de Catalunya, Palau Nacional-Parc de Montjuíc (Barcelona 1999), pp. 84–88, transl. as “Roda: l’Esquerda. The Carolingian Town” ibid. pp. 461-463; cf. Ramon Ordeig i Mata, Els orígens històrics de Vic (segles VIII-X), Osona a la butxaca 1 (Vic 1981), online here, pp. 22-26.

18. Zuckerman, Jewish Princedom, pp. 316-319.

19. Ibid., ‘Ispamia’ p. 317 and thereafter, quote p. 318.

20. Jakob Winter and August Wünsche (edd.), Die jüdische Litteratur seit Abschluss des Kanons: eine prosaische und poetische Anthologie mit biographischen und litterargeschichtlichen Einleitungen (Trier 1894-1896), 3 vols, vol II, online here, pp. 23-24.

21. Identified by Winter and Wünsche as quotation of 5 Moses 1, 11.

22. Winter’s and Wünsche’s background information covers the author (vol. II pp. 22-23) without references, but of the actual text offered they say only, “Aus „Kebuzat Chachamim‟, Wien 1861, S. 110”, which from Zuckerman and Google it’s possible to decode as W. Warnheim (ed.), קבוצת חכמים: כולל דברי מדע פרי עשתנות חכמים שונים: Wissenschaftliche Aufsätze in hebräischtalmudischer Sprache (Wien 1861), p. 110, but as I say, that doesn’t get me personally much further.

23. As you can see above, Zuckerman, Jewish Princedom, p. 319 n. 6, gives this source as “J. Müller, Teshubhot geoné mizrah uma`arabe, no. 26, p. 9a”, but websearch for that string or variants produces nothing, so I guess that the actual title is again in Hebrew, and he transliterated it into Roman, an operation I cannot reverse.

24. The first possible evidence that the city’s church was up and running again comes with Archpriest Godmar, soon to become the first bishop, who turns up in Ordeig, Catalunya Carolíngia 4, doc. no. 2, but we don’t know for sure that he was Archpriest of Vic; that association only becomes clear when he occurs as bishop ibid. doc. 7, in 885. Eduard Junyent, who first edited these documents, thought that the handwriting of the main scribe who worked with Godmar when he was archpriest, Athanagild, suggests that both had been brought in from Narbonne, which is a priori likely and would help make sense of the see’s later special subjection to the metropolitan one.

25. Accessible to me as Ibn el-Athir, Annales du Maghreb et de l’Espagne, transl. Edmond Fagnan (Alger 1901), which is no longer online whence I got it, sadly, but where the relevant bit is on p. 47.

28 responses to “A Jewish garrison town in Carolingian Catalonia?

  1. And people say the tales of King Arthur are far fetched!

    • Well, some more far-fetched than others. Back in the ninth century he was a perfectly plausible sub-Roman warrior over-king who just happened to win where the historical ones, mainly Urien of Rheged, had eventually lost. But then the twelfth century, a newly unified Britain not necessarily universally happy about that generates Geoffrey of Monmouth and his work meets Chrétien de Troyes, perhaps thanks to Walter Map, three different kinds of genius thus interact and then bang, there we are… But of these, perhaps Geoffrey is the closest comparator to the method displayed in this particular paper!

      • Joseph Brown

        Let’s not forget Wace comes into the equation too – without him, no knights of the Round Table.

        • Is that right? You’d think I’d know, but I’d thought that bit came with Chrétien. Am I wrong?

          • Joseph Brown

            Yeah, Wace came up with the Round Table. He explained its shape on the grounds that it would prevent Arthur’s knights from quarrelling over rank and precedence, a problem which a polygonal table can easily create. The idea is that all knights are equal by default, with honour being awarded according to ones deeds alone (as Geoffroi de Charny would say 200 years later, “he who does best is most worthy.”) He also adds in familiar characters like Sir Gawain, Sir Kay and Sir Bedivere for the first time, and gives a lot more attention to gallant knights, beautiful ladies and courtly stuff than Geoffrey of Monmouth did. Chretien de Troyes then adds in more knights, develops their backstories and everything. Its basically between Wace and Chretien that King Arthur stops being a sub-Roman British superhero and instead becomes a source of historical legitimacy for this new pan-European aristocratic ideology.

            • Well, I don’t honestly mind as long as I can still imagine Walter Map being the person who told Chrétien the Matter of Britain story one long weekend at someone’s court in Normandy. I can’t remember what piece of scholarship gave me that delightful idea, but there was one…

  2. Joseph Brown

    Once again, Bachrach never ceases to amaze me with the things he can come up with. I’ve entertained some pretty wild ideas when I very briefly gave a stab at historical fiction – a Gallo-Roman minor aristocratic family who can trace their ancestry back to the heroes of the early Republic still living in their perfectly preserved ancestral villa with latifundia in the Maconnais at the end of the tenth century, until a group of rogue knights destroy it because they want to get hold of some land and avoid sinking back into the peasantry. Yet I’m starting to wonder if that’s more plausible than what Bachrach and Zuckerman are suggesting was going on in Marca Hispania.

    • From my reading of Christian Lauranson-Rosaz, that would be a perfectly plausible plot if it were only set in the Auvergne! The Mâconnais, though, so infra dig

      • Joseph Brown

        I kind of reflexively chose a Burgundian setting because I’d read Bois (but hadn’t read Lauranson-Rosaz) at the time, and I’d also read a story in Raoul Glaber where he describes a hail of stones firstly appearing outside the country villa of a nobleman called Arlebaud whose family then get attacked by “knights from the castle of Auxerre.” But it sounds like Auvergne would have worked better, and thanks so much for directing me to Lauranson-Rosaz’s work. It sounds really interesting.

        • Well, I don’t know if I’d go that far, and if you now eagerly head off in search of it you may not thank me either. What I would say instead is, his essential gimmick is to claim, on the basis of monumental culture and formulae in charters that no-one but him ever saw because he never finished his edition of the Clermont archive, that the Auvergne was the deepest reservoir of Gallo-Roman aristocratic identity continuity in the whole of Gaul, proceeding more or less undisturbed by the wider world until c. 1000, when suddenly crisis, Peace of God, and rapid and violent feudalisation. But although it all ends up Dubious enough, the argumentation of the survival of Romanitas there verges on pleading as special as that discussed in the actual post here. So I’d say, try either of Christian Lauranson-Rosaz, “La romanité du midi de l’an mil (le point sur les sociétés méridionales)” in Robert Delort (ed.), La France de l’An Mil, Points-Histoires H130 (Paris 1990), pp. 49-74, or “La romanité du midi de l’an mil : le point sur les sociétés méridionales” in Xavier Barral i Altet, Dominique Iogna-Prat, Anscarí M. Mundó, Josep Marí Salrach & Michel Zimmermann (edd.), Catalunya i França meridional a l’entorn de l’any mil: la Catalogne et la France méridionale autour de l’an mil. Colloque International D. N. R. S. [sic] /Generalitat de Catalunya «Hugues Capet 987-1987. La France de l’An Mil», Barcelona 2 – 5 juliol 1987, Actes de Congresos 2 (Barcelona 1991), pp. 45-58 (the same paper for all basic purposes), and if you like Lauranson-Rosaz, “Peace from the Mountains: the Auvergnat origins of the Peace of God”, transl. Richard Landes in Thomas Head & Richard Landes (edd.), The Peace of God: social violence and religious responses in France around the year 1000 (Ithaca 1992), pp. 104-134, and see how the trick is played. There’s really only that one trick. Then you can decide for yourself how much you need the full-length exposition!

  3. Yacov Guggenheim

    The identification of אליסאנה/אליוסאנו with Ausona is due to, how should I say, the vivid imagination of Zuckerman. The author is believed to have visited France, however, by miraculous leap (which has been denounced already by Hay Gaon (fl. around the year 1000) as a fake). More on the Jews in Lucena you’ll find in the first volume of ‘The Jews in Moslem Spain’, 1992, by my late teacher and friend Eliyahu Ashtor alias Strauss, s.v. So, don’t worry about your inexistent Hebrew (in this case, at least …) BTW Natronai bar Hilais Responsa have been edited in a critical edition by Robert Brody: Teshuvot Rav Natronai Bar Hilai Gaon : edited from manuscripts and early editions with introductions, indices, explanatory notes and comments (hebr.), Jerusalem 2nd ed. 2011.
    I hope you are well again, Yacov Guggenheim

    • This is tremendously helpful, thankyou. I really need to spend some time reading Ashtor, not least because I have a student who is slowly putting together an article on Hasdai ibn Shaprut, on whom Ashtor had so much to say; and I should have thought of him for this case as well. Thankyou for the reminder. You are also one of two people to point me to Brody’s edition, which is also very useful to know about. There is an article of mine which has been held up behind this difficulty (though not just this difficulty), so I shall be putting this help to very direct use; it is much appreciated. And meanwhile, yes, I am pretty much recovered and should hopefully be able to stop isolating tomorrow!

  4. I’ve been trying to disprove that Zucherman’s jewish Osona theory for some years now, but I’ve been unable to find any reference to a jewish Lucena before Ibn-Hafsun…

    • Well, maybe this thread of comments will prove as useful to you as it seems to be to me. I didn’t know about that poem, I’m ashamed to say. When I go on holiday next week I must really take at least one of your books with me and stop myself from having to confess ignorance this often!

      But to the actual problem, it seems that we have no good way to place the town of this letter, preserved in two different versions of the name in what, I am now told, are thirteenth-century and fifteenth-century manuscripts respectively – not very encouraging – except what the text itself tells us. Without it, we seem to have no reason to suppose Jews in either Ausona or Lucena this early. So, this letter tells us, I think, that the town was considered to be in al-Andalus. That is clearly true for Lucena and not very clear for Ausona. Then there is the form of the name. An initial ‘a’ to begin a word now starting with ‘L’ isn’t too hard to suppose; but a city name never normally needs a preliminary definite article, does it, and so I find al-Osona quite difficult to believe in. And if Lucena did have a unknown Jewish community only a few decades later, those do seem like enough reasons to me to decide to place this letter’s recipients there not in Ausona. Of course, I’d rather like to be sure. But I can’t help also feeling that if there were so substantial a Jewish population in old Ausona as the letter implies (despite its still-recent troubles), we might by now have found some of their burials…

      • Allan McKinley

        We can perhaps make a more positive (if technically not certain) identification of the town. The requirement for any etymological discussion is that you try and find the earliest forms of a name. In this case Wikipedia’s article on Lucena has been potentially useful, although I’d prefer some references; the relevant text is:

        The city was originally known as Eliossana, etymologically reinterpreted as deriving from the Hebrew אלי הושענא‎ Elí hoshanna, “May God save us”. The name in Arabic is اليسانة Al-Yussana.

        My first observation is that without reading Hebrew that may not be the same name as Zuckerman cites; frankly the Wikipedia-provided Hebrew name looks like an antiquarian homophonic etymology to me, but I have no experience in dealing with medieval Hebrew naming practices, so this reservation will have to stand. My second observation though is that assuming this information is accurate then the apparent required initial-vowel sound is explained by the Latin (I assume – it could be from an earlier language) name Eliossana.

        So on the (unsatisfactory) evidence presented here the identification of the mainly-Jewish town of the letter with Lucena might be challenged because the Hebrew names cited don’t look the same, but as this is relying on a historian of dubious integrity and a form that looks horribly like a non-etymological translation, I would not regard this objection as a serious problem, and certainly we can disregard any concerns that Lucena’s name is phonologically unsuitable in this case (note Zuckerman, who presumably could read Hebrew, does not actually tell us why the name he cites is not Lucena…).

        Conversely, Zuckerman seems to think there was a town called *Al-Osana, but the attested names for Vic I can pick up seem to suggest it was called Auso in the Roman period and (at least nominally) Vicus Ausonensis in the Carolingian period. To explain *Al-Osana as a name would require the Jewish population of the town to use an (eighth or ninth-century) Arab name which somehow applied the direct article not to Auso/Vic, but rather to the regional name Ausona, which may not be impossible but requires a not-provided explanation. If someone is suggesting a pre-existing Jewish community in the area, it might also be asked why they used what would have to be an eighth-century Arabic name in favour of whatever name was used before that? Now in some circumstances an ingenious suggestion like *Al-Osana could be a clever answer to an etymological puzzle (that might be a different thing to a correct solution though!), but the simple fact remains that in this case a settlement called Eliossana existed and without good reason to doubt this relevance of this name it’s hard to see a linguistic argument that an alternative location with a similar if much-less-likely name is required: the onus was on Zuckerman to prove this, and he just waved a vague assertion around instead. When we add in the fact that Lucena was a majority-Jewish community later, a situation of unknown origins, then it seems perverse to suggest a reference to a settlement with a name very similar to Lucena’s Roman name and an otherwise-unique-in-Iberia population dynamic that matches an established dynamic found relatively-soon after the period in question at Lucena somehow does not apply to Lucena.

        I think that this is a clear case of an extraordinary claim requiring extraordinary evidence. With the provisos that I can’t comment on the Hebrew names, and also that my knowledge of linguistic developments in Iberia is extremely limited, I can’t see there being a case for hypothesing the existence of a name *Al-Osana when it’s clear the source in question could apply to Lucena. When the originator of the name *Al-Osana failed to present a case for this name being required and rather relies on his own extremely-controversial interpretation of limited evidence to provide a context for the name, I think you can happily dismiss this as ahistorical nonsense.

        • Hi Allan, and thanks for the exposition. Your conclusion seems to be the conclusion of most other commentators too, and it’s all been very helpful. You’re right that the leap from Auso or Ausona to al-Osana is the biggest stumbling block; but the apparent relevance of Córdoba is another, and in both these cases, the city near Córdoba with a known large Jewish population, i. e. Lucena, seems like the better fit whatever the Hebrew. I understand from someone who has mailed me that we have the name as al-Usuna and al-Osano (roughly) in thirteenth- and fifteenth-century manuscript copies respectively, so the Hebrew isn’t actually a very close constraint – enabling Zuckerman to do what he did, of course.

      • Absolutely, maybe Ashtor has some new evidence!
        About burials, it’s difficult. A ‘jewish only’ Osona probably it could only exists after the muslin conquest, so we are trying to detect the remains of at most 5/6 generations, and we have no clue about what distinctive traits do we have to look for. The best I could find was some tombs with support for libations in the Esquerda (and yet maybe not a jewish practice at all…!) Bonilla i Sitja, Elisabet : 2017 : “Percebre i relacionar-se en els comtats d’Osona i Manresa durant la primera meitat del segle X” p.164. Surely Ollich knows better…

        • La Profesora Ollich has indeed shown me those tombs! She regards them as a hangover of pagan practice, but probably still tenth-century because of Roda also being destroyed during Aizó’s rebellion, and while I’m not sure there really should be remaining pagan practices so late there, it doesn’t seem that Jewish (or Muslim!) burial explains it any better!

  5. Joseph Brown

    Also couldn’t Bachrach have figured that the nickname Naso could have had nothing to do with his physical appearance. After all, Naso was the personal nickname for Publius Ovidius, and like Ovid, Bernard of Septimania was very good at attracting scandal to himself – Carolingian court nicknaming culture, after all, revolves around which classical or Biblical personality you best watch up to (I.e. Angilbert being Homer, Charlemagne being David etc). A missed opportunity to point out Carolingian classicism, says I!

  6. Perhaps a nit-pick: “an autonomous Ashkenazi principality on Narbonne that was eventually shut down by various interests colluding with the pope in 900.”

    Did the Ashkenazi Jews exist as early as that, and in that part of the world? I know that geneticists argue that they were a 50:50 mixture of Levantine males and southern European females, but I have the impression that they thought the AJ became a distinctive in-breeding population in the Rhineland around the year 1000. Open to correction.

  7. I wonder exactly what he meant by it? No matter. Maybe exactness isn’t available.

    • yeah, its a weird concept. Because Ashkenazi is associated with Central European Jewry, whereas these Jews will have been Spanish/ Septimanian and thus proto-Sephardic more like.

    • Well, for better or for worse I do own the book – it floated in front of me at a conference book sale a while ago and I couldn’t pass up the possible cause to read it – so I thought it’d be pretty lazy of me not to check at this point. And it turns out that this may be another of those places where Bachrach misreports his sources, since I can’t see, on a very quick page-flip and index check, that Zuckerman used the terms ‘Ashkenaz’ or ‘Ashkenazi’ at all. So I guess I got it from the Bachrach piece, but I don’t have access to that any longer to check. It may be that I just made it up myself, though I have a vague memory of using the phrase as an echo of something I’d seen. If Bachrach did use it, he may have been reacting to Zuckerman’s claims that the diaspora Jews in Europe, including Iberia, arrived there in the first centuries AD direct from Babylonia with their institutions intact, rather than from Africa later. But Zuckerman was happy to use the word ‘Sefarad’ for Iberia, albeit only for Muslim Iberia; he insists, on the basis of older work by Joseph Calmette – the man who had the idea that ‘Naso’ referred to big noses in the family of Guillaume d’Orange… – that the Christian bit, i. e. the Carolingian Marca hispanica, was called Ispamia (p. 261). So wherever ‘Ashkenaz’ entered this conversation from, it wasn’t Zuckerman, sorry.

  8. Hello. I believe I am a descendent of William of Gellone. I am hoping to find out more information on our family lineage to write curriculum for our children (most U.S. history curriculum does not reach further back than the 16th century). Thank you for your dedication to these studies.

    • Well, once upon a time I would have referred you to Nat Taylor, friend of this blog, specialist in the right area of the world and sometime professional genealogist; but it turns out he no longer does such work on commission. I suppose he might still be able to recommend you to someone who could, but getting hold of a recent issue of the journal American Genealogist, which he edits, and seeing who is advertising, might get you to the same place faster! That’s about as much help as I can be; the only references I have to the genealogy of that family start at the early end and are in French or Catalan…

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 )

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.