Rebel without a pension: the mystery of Aizó

On the way to a really great meeting in Oxford a few days ago, about which I’ll write separately, I took with me Jordi Camps’s Cataluña en la época carolingia and re-read a couple of articles in it by Immaculada Ollich that I’d skimmed for book purposes a while before but not, apparently, fully absorbed. Both of them heavily featured this one figure who seemed good material for a blog post, a man who threw back Carolingian rule in part of Spain for nearly sixty years, or so it is said, and about whom we know almost nothing. So I thought I’d do an exposé in the style of Carla Nayland or Judith Weingarten, complete with headings. But over the several days of on-and-off construction it’s turned into a four-thousand word monster (I am having real trouble typing that instead of `monastery’ these days you know) which closely resembles genuine scholarship and I thought perhaps it belonged behind a cut. I’d be delighted if you can find the time to read it but if not, don’t worry, there’ll be time later for other things.

The events of 826-8 on the Spanish March

From takeover to rebellion

Since 785, as I’ve described at length before and won’t recap here, the Carolingian Empire had extended over the Pyrenees into what is now Spain, and in 798 the vulnerability of this new frontier was addressed by Louis the Pious, then King of Aquitaine, who ordered the refortification of a number of centres along the River Ter, most especially at the once-Iberian hillfort of l’Esquerda, also known as Roda de Ter, and a sensible and defensible line therefrom, partly to secure what had been gained but also to try and cut off Barcelona, still Muslim-controlled, prior to the campaign to take it that eventually succeeded in 801. I’ve talked about this refortification here before as well, which was entrusted to Count Borrell I, probably of Urgell, mainly because it included l’Esquerda, which is in the book, but it also appears to have involved a garrison at old Roman Ausona, where a Roman temple still stood (still stands, indeed) and where the Visigoths had apparently had a palace of sorts and a military garrison too.¹

The Roman temple of Vic, preserved by its incorporation into the later castle, ironically now gone

After the fall of Barcelona, this area was placed under the superior control of Count Bera I (of Barcelona, 801-20). He was a Goth, or at least half a Goth, seems to have been well-known in the area, and dealt fairly independently. In the end, he dealt too independently, because in 820 he was appealed before Louis the Pious, now Holy Roman Emperor, by his fellow count and rival Bello of Carcassonne, for making unauthorised peace treaties with the Muslims. The affair nearly went to trial by battle but Louis intervened, removed Bera from office and exiled him to Francia on a pension. Count Bernard of Toulouse was the man to benefit here, along with his brother Gaucelm, who were probably Bera’s half-brothers in fact; Bernard became Count of Barcelona, though he is remembered as Bernard of Septimania (and the husband of woman of maternal letters Dhuoda), which came after the events I’m about to relate. These appointments have been seen as a decided change, from a hands-off policy that left a slightly Arabicised Hispano-Gothic élite more or less where they had been but under different masters, as happened for example at the Carolingian takeover of Narbonne in 759, to an interventionist one manned by powerful Frankish nobles.² Some truth in that, perhaps, though the kindred ties between Bera and Bernard have also been emphasised as possibly making the Toulousain count the best man for the job anyway. Barcelona seems to have stuck with him.³

A commotion on the frontiers

But certainly there was opposition. Bera I had left a son, Guillemon, who was effectively removed from power, and we find in the Royal Frankish Annals the following, in a report of a council meeting in 826:

There came there news of the flight and treachery of Aizó, to the effect that, having entered into Ausona by trickery and been received by that people, whom he had won over by cunning, he destroyed the city of Roda and reinforced the castles of that same region, those which appeared strongest, and he sent his brother as emissary to ‘Abd al-Rahman the king of the Saracens for help, which he was seeking, and accepted the orders of the same king against us.

Wait, what? Where does Aizó fit into the story? No Frankish source names such a person before this. More to the point, why has this got anything to do with Bera and Goths? Well, the Annals go on, after some other business, with an entry for 827:

The emperor sent Helisachar, priest and abbot, and along with him Counts Hildebrand and Donatus, to settle the disturbance on the Spanish March. Before their arrival, Aizó, relying on the help of the Saracens, brought many adversities upon the guardians of that frontier, and beleaguered them so much with carefully-mounted raids, that those of them who were left, who should have kept guard, withdrew from their castles. There defected to him both Guillemon son of Count Bera and very many others of that people [alii conplures gentilicia], motivated by the desire for new arrangements [novarum rerum levitatibus cupiditate]. And, joined with the Saracens and Moors, they afflicted the regions of Cerdanya and Vallès daily with rapine and fire.

And since Abbot Helisachar and the others with him sent by the emperor were administering considerable, and appropriate, remedy, with the counsel of their companions, to the Goths and Hispani of that region who required settling and soothing, and Bernard Count of Barcelona was also most pertinaciously resisting with shrewdness and fraudulent machinations the schemes of Aizó and those who had defected to him, and was making all their temeritous attempts useless, an army had been sent, brought to the aid of Aizó by the king of the Saracens Abd al-Rahman, over which Abu Marwan, kinsman of the king, had been placed, and was announced to have arrived from Saragossa, so that from the persuasions of Aizó a by no means uncertain victory was in promise. Against which, the emperor ordered his son Pippin King of Aquitaine with immense Frankish forces to guard the boundaries of his empire. And this might have been done thus, but for the idleness of the leaders who were in charge of the army of the Franks; the army came more slowly to the March than the necessity of matters dictated, because of those who were leading it. This tardiness was so harmful that Abu Marwan, having wasted the fields and villages of the people of Barcelona and Girona with fire, and having defeated those who had massed outside the cities, brought himself back to Saragossa with his army in columns before he could be sighted by our army.

And the Annals go on to say that this defeat was presaged by terrible phenomena in the sky and that Louis fined those in charge according to their rank at the next assembly.⁴ His eldest son and co-Emperor, Lothar, was sent to the March with Counts Hugh and Matfrid, the latter Lothar’s father-in-law, to order things further, but it was gathered that a Saracen attack was imminent so Pippin was got to come too with extra forces, and then the attack never materialised so they had to bring the troops home. This was the last time a Carolingian king ever went south of the Pyrenees. Soon afterwards Louis also removed Matfrid and Hugh from their offices and made Bernard chamberlain, which prompted schemes against Bernard and by 830 the empire was riven by a coup against the emperor, so really, Aizó’s stab at the Empire and the responses to it arguably destabilised Louis’s court enough that one could see his rebellion as the equivalent of the assassination of Archduke Ferdinand, the local event that tipped off a Europe-wide political convulsion. Be that as it may, I want to get at the local context, rather than the coups against Louis that have been so well studied elsewhere.⁵

The politics on the Spanish March of Aizó

It seems clear from the conciliatory measures that Helisachar and the counts he brought with him had to adopt that there was a good deal of discontent with Carolingian rule among some of the peoples of the March. One might assume that the discontents felt that the terms under which the Carolingians had taken power, which are obscure but probably included an agreement that the locals might retain their law and judges as had happened at Narbonne, were not being respected; Carolingian missi dominici were active in the area and there is a whisper of evidence that Frankish law was being applied in courts, albeit for Franks; that might have been enough, especially when prominent locals were excluded from power after the fall of Bera.⁶ There is also some evidence that the new settlers who were being given immunities in return for guard duty, the Hispani referred to, may have been resented at local levels; we have documents from them complaining of being encroached on by both local peasants and Frankish counts, so they probably felt especially betrayed and the people upset by them probably saw them as interlopers themselves.⁷ It’s also possible that the locals weren’t ready to bear the stress of Carolingian campaigning, which continued for as long as Louis was King of Aquitaine (till 814) and was presumably renewed on a local level after Bera’s removal from office. This may have made people feel endangered, though the court point of view seems to have been that a secure military zone was needed to protect the lands that they had actually taken control of.

Approximate domains of Count Borrell I in their context prior to Aizo's revolt

Anyway, Aizó seems to have had no trouble raising supporters, albeit from the very edge of the occupation where it must have been sparsest and most military, but contrariwise from the Annals it seems that Helisachar could in fact offer enough reassurances that the problems would be addressed to calm people down. Aizó was apparently weakened enough that he realised he needed help, quite possibly because his troops were being stabbed in the night by Bernard’s men, his transport sabotaged, springs poisoned, his scouts subverted or his guides bribed, whatever the fraudulent machinations that the RFA grudgingly admit Bernard’s success with actually were. Bernard’s subsequent career in exile in this area makes it clear that he was especially ready to make whatever promise might serve and then abandon it when advantageous, and since this is the sort of behaviour the Frankish sources regularly decry barbarians for, it’s easy to see why the RFA might be uncomfortable with it working to the Frankish advantage.⁸ So it might be Franks versus Goths at its simplest, but Aizó raised troops that include ‘Saracens’ and ‘Moors’ while Goths and Hispani seem instead to have felt threatened generally by events rather than joining him outright. The exceptions here are the disinherited nobility and the alii conplures gentilicia, which I feel should mean Bera’s kindred (which was possibly quite large and rich, as they went on for a long time yet) and their followers rather than a gens proper, but it’s hard to say.⁹

The identity of Aizó

So who was this rebel, so successful at Ausona, so connected to Córdoba, so dangerous or significant as to have been penned up in the Frankish court but escape, and apparently familiar enough to writers based there that they didn’t think he needed explaining? Suggestions have of course been made, that he was a disadvantaged Goth, a local noble of some description, but it’s all rather vague.¹⁰ The first thing to do is pay close attention to what the sources say.

Just the facts

First, it’s worth stressing that as far as we know Aizó is not a Gothic name, or come to that a Frankish one. It’s also notable that from the beginning the Frankish sources associate him with Andalusi help, without which it’s implied he couldn’t have operated. Furthermore he apparently had connections in Córdoba that commanded some respect. Yet I feel sure that if he could have been identified as a Saracen or a Moor, the ARF would have happily done so. Saracen is a tricky term but if they can use it of the Cordoban armies, there is presumably some way in which Aizó was not part of that group, even if it was only the familiarity of him having been round court for ages.¹¹ That also needs thinking about. He seems to have been there on a pledge, because his flight was somehow ‘treachery’ for the Frankish chronicler; he was no mere prisoner, therefore, but probably a hostage.¹² That in turn implies, as do the other indications that people took him seriously, that somewhere he was important and presumably (as the brother emissary also implies) important to someone else in his family who was more powerful.

A cunning suggestion and its problems

It wasn’t until 1967 that a better possibility was offered for Aizó’s identity. Fernando de la Granja, who was doing an epic work of translating Arabic sources on Spain into Spanish, that year published the Tarsi al-ajbar wa-tanwi al-atar wa-l-bustan of al-Udri (1003-1085), a geographical-historical compendium about the post-Caliphal kingdom of Zaragoza.¹³ This entailed quite a lot of local history of dubious provenance and among them were stories about the sons of the governor of Barcelona who famously decided not to hand over his city to Charlemagne in 778 after all, thus inadvertently causing the Chanson de Roland. His name was Sulaiman ibn Yaqhdan al-Arabi, and one of his sons was called ‘Aysun. Furthermore, the story of that son is that he was held captive by the Franks and escaped, by a means that could only really exist in such a story: he pretended to an eye condition that meant he had to adopt a full-face hood, and after this had become accepted, one day got the servant who brought his food to swap clothes, including the hood, and got out in the guise of the servant, who exposed the trick once his master was well on the way to Córdoba. Al-Udri is more interested in the servant, who went on to do great things as a general, and by the Franks’ debate over whether he should be punished or not, but he does say that ‘Aysun was breaking his word to his captors by trying to escape, so the name, the Cordoban connection and the imprisonment and escape all gel quite nicely with the Frankish sources.

Obverses and reverses of two C8th Arabic copper fulus found in digs under Sant Miquel de Barcelona (Museu d'Història de la Ciutat de Barcelona 17160 & 17161)

However, there are problems. The fact that al-Udri doesn’t mention ‘Aysun raising the standard of revolt on the frontier is one, but perhaps not significant since the actual big history was of less interest to him than the morals and character of his subjects. On the other hand there is the objection that we first see ‘Aysun in the Arabic sources leading forces at the time of his father’s broken deal with Charlemagne, meaning that he was at least a young adult in 778. By 826, therefore, he’d have been around seventy.¹⁴ Is this really an age to run as a fugitive halfway across Europe and then lead a rebellion, you have to ask? Whatever we think about the energy of ninth-century septuagenarians, there is the further problem of Aizó’s brother; ‘Aysun certainly had a brother, Matruh, who was indeed known at Córdoba in his time, but he had been assassinated in 791/2, so it seems unlikely that it can be him the RFA mean.¹⁵ Lastly, note that the family surname was al-Arabi. This implies a descent from actual Arabs from the East, in distinction to local converts, and makes the Annals‘ failure to call Aizó ‘Saracen’ even stranger. Of course, the Arabic origin would have got further away with each generation of marriage, but one (or this one) still wonders if al-Udri hadn’t somehow got hold of the Carolingian story and put one and two together looking like four, rather than actually having any information on the subject. (Knowing who ‘Aysun’s mother was might resolve some of these problems, but, all of them?)

The Jarrett conclusion

My personal feeling is that the identification with ‘Aysun fits with the letter of the sources but not the spirit. The al-Arabi family were Walis of Barcelona, and although they did turn up elsewhere it was rather further afield, Lleida, Saragossa, and so on. The fact that Aizó could apparently get into provincial Ausona and be accepted, but be fought off at Barcelona and Girona, in both of which places there would be people who remembered ‘Aysun and Sulaiman, doesn’t quite fit for me. So, who was there in Ausona to welcome him? It doesn’t seem that there was an Arabic presence there, pace Ramon Martí who asserts that there were Arabic garrisons everywhere, and certainly no destruction levels have so far been found, which suggests that the city made terms with the Muslims when they arrived.¹⁶ So the locals would at that rate probably be whatever they were before the Muslims arrived, albeit with a new Frankish, Gothic or Hispanic military presence since 798. And there would have been someone in charge. The old bishopric seems to have collapsed, but there are three cemeteries that were probably still open in the city or outside its walls at this point, and if there’s one thing my studies continue to teach me of this area’s archæology it’s that where there’s burial there is power. Usually this is more obvious because of being on a hilltop with a castle, but an ancient city with a temple and an old cathedral in it will also do as a power centre I think!

Plan of ninth- or tenth-century necropolis excavated in the precinct of the Cathedral of Vic

Either way, in this city with its Iberian name (named for the Ausetanes whom the Romans met in the area) and palæochristian history, I suspect some long Christian and urban continuities were in operation. The people who met here probably didn’t identify as Goths, which may by now have been more of a military professional’s identity than an ethnic one, in fact they may not have identified as much beyond Ausetanes still, however Romanised (and then `vulgarised’, or whatever you want to call the development of society in the successor kingdoms).¹⁷ So if Aizó could walk in there and make it his new capital, by whatever promises and schemes, I think that this was an identification he could share. There had probably been a governor under the Muslims, or at least someone answerable to Córdoba for the pact; I suspect that this person was the important one for whose conduct Aizó was serving as hostage. And if Ausona was suffering similar discontent under Frankish occupation as the other areas, especially since it was probably only seeing soldiers, and possibly having to billet and feed them, they might well have listened to the old governor’s boy come back, even if he’d had to gather Andalusi troops and take the Umayyad shilling dinar to make much of an attempt. So I think less Arab noble and more local hero, myself, but there’s obviously not much way to be sure.

The unknown end of Aizó

One further mystery about Aizó however is that we don’t know what happened to him.¹⁸ The Annales Regni Francorum don’t mention the problem again after Lothar’s failed expedition, but that a king was to be sent suggests that at that point in 829, things had not got better. Nonetheless, there is no more action about it recorded, and fair enough: Louis had problems of his own and Barcelona became a Carolingian no-go area once Bernard flew there for safety in 832. But at that rate what happened to Aizó and his rebellion?

Immaculada Ollich has made the most controversial suggestion here, pointing out that the Carolingian fortress line had been centered on Roda de Ter and that it is extremely strange that this key station was not reoccupied. Come to that, why did it take another fifty years to retake control of Ausona? Granted, Bernard had his hands full and was no longer really on the Carolingian project team, but it was a big bite into home territory for the other counts on the March and Bernard didn’t last forever. Ollich suggests that the reason these things could not be done was that Aizó was still in control there, and in fact holding both cities as Muslim redoubts against the Carolingians and their lackeys. She also points out that a surprisingly large number of Muslim raids come through the area in the following years and suggests that this is exactly because the previously secure corridor was now breached and the Muslims had a safe route almost up to Barcelona.¹⁹

Ruins of the settlement and fortress of Roda de Ter from the air

This is all alarmingly plausible, as long as Aizó was not ‘Aysun, because if he had been the Arabi septuagenarian he surely couldn’t have lived that much longer. If Aizó was son of the judge of the Ausonans or similar, however, then this works quite well along with other local figures of similar dignities based at local castles, about whom you can read in a certain forthcoming book.²⁰ But even then, we’ve no idea how long he lasted in power, to whom he might have passed such power on, how long the Muslims might have been able to keep a pacted representative in station, and how long it took before the place went back to the locals. Some mysteries we can’t solve. But the man and his legacy have been made pretty huge, and they deserve looking at at least this closely.

---

1. Immaculada Ollich i Castanyer, “Vic: la ciudad en la época carolingia” in J. Camps, Cataluña en la Época Carolingia: arte y cultura antes del románico (siglos IX y X), pp. 89-94, transl. as “Vic: the town in the Carolingian age” ibid., pp. 464-466; also her “Roda: l’Esquerda. La ciudad carolingia”, ibid. pp. 84-88, transl. as “Roda: l’Esquerda. The Carolingian Town” ibid., pp. 461-463.

2. The dispossession of Bera is recounted in the Annales Regni Francorum, ed. Friedrich Kurze as Annales regni Francorum inde ab a. 741 usque ad a. 829, qui dicuntur Annales Laurissenses maiores et Einhardi, Monumenta Germaniae Historica (Scriptores Rerum Germanicum) VI (Hannover 1895; repr. 1950), transl. B. Scholz & B. Rogers in eidem, Carolingian Chronicles: Royal Frankish Annals and Nithard’s Histories (Michigan 1972), pp. 35-125, with commentary pp. 2-21, s. a. 820. For (Catalan-specific) analysis see Josep María Salrach i Marés, El Procés de formació nacional de Catalunya (segles VIII-IX). 1: El Domini Carolingi, Llibres a l’Abast 136 (Barcelona 1978), pp. 39-49.

3. Bernard’s relationships to the other ruling families of the March now best explored by Martin Aurell, “Pouvoir et parenté des comtes de la Marche Hispanique (801-911)” in R. le Jan (ed.), La Royauté et les Élites dans l’Europe Carolingienne (début IX^e siècle aux environs de 900) (Villeneuve de l’Ascq 1998), pp. 467-480.

4. Annales Regni Francorum, s. aa. 826 & 827.

5. For those, see Janet L. Nelson, Charles the Bald (London 1992), pp. 75-104 and David Ganz, “The Epitaphium arsenii and Opposition to Louis the Pious” in Peter Godman & Roger Collins (edd.), Charlemagne’s Heir: new perspectives on the reign of Louis the Pious (814-840) (Oxford 1990), pp. 537-550, as starters.

6. A few early court cases featuring missi are preserved in the Cartoral de Carlemany in Girona, edited by Josep María Marqués i Planguma as Cartoral, dit de Carlemany, del bisbe de Girona: s. IX-XIV, Diplomataris 1-2 (Barcelona 1993), and there are also of course Carolingian capitularies issued to the March, best edited by Ramon d’Abadal i de Vinyals in Catalunya Carolíngia 2: els diplomes carolingis, Memòries de la Secció històrico-arqueolòica 2-3 (Barcelona 1926-52), II app. I-X.

7. On this, you can in fact safely see Cullen Chandler, “Between court and counts: Carolingian Catalonia and the aprisio grant, 778-897” in Early Medieval Europe Vol. 11 (Oxford 2002), pp. 19-44, though you should as soon as you can follow it with Jonathan Jarrett, “Settling the Kings’ Lands: aprisio in Catalonia in perspective”, ibid. (forthcoming).

8. It is a mystery to me why dedicated attention is paid to Bernard of Septimania only by crazy nationalists. Surely his personality and apparent ability is one of the most crucial things in the collapse of Louis’s reign, yet he is hardly ever allowed any agency at all, he’s just this useless malevolence that Louis can’t shake off, whereas this episode and his long subsequent survival show that he was anything but ineffective, though perhaps overconfident, and must have been a genuine force in politics. Almost the only exception, even in French, known to me is Lina Malbos, “La capture de Bernard de Septimanie” in Le Moyen Âge Vol. 76 (Bruxelles 1970), pp. 5-13, though I believe he also gets a decent outing in Léonce Auzias’s L’Aquitaine carolingienne (Paris 1938, repr. Pau 2003 for lack of any kind of replacement). I for my part am following Salrach, Procés de formació 1, pp. 73-120. As to the agendas of the RFA, there see Rosamond McKitterick, “Constructing the past in the early Middle Ages: the case of the Royal Frankish Annals” in Transactions of the Royal Historical Society 6th Series Vol. 7 (Cambridge 1997), pp. 101-129.

9. Here, Salrach, Procés de formació… 2: la establiment de la dinastia nacional, Llibres a l’abast 137, pp. 6-11 for Bera’s kindred and their importance.

10. So, for example, Ramon d’Abadal i de Vinyals, La Plana de Vic en els Segles VIII i IX (Barcelona 1948), repr. as “La reconquesta d’una regió interior de Catalunya: la plana de Vic (717-886)” in idem, Dels Visigots als Catalans I: la Hispania visigòtica i la Catalunya carolíngia, ed. J. Sobreques i Callicó, Estudis i Documents XIII-XIV, (Barcelona 1969, 1974), I pp. 309-321 at pp. 311-315 of the reprint for the Gothic noble hypothesis; cf. Aurell, “Pouvoir et parenté”, pp. 470-471 for presentation as some unknown cadet of the Toulouse comital family. Many of Bera’s kindred might have been in a position to make either position plausible as more useful of course.

11. For medieval Hispanic usage of the word Saracen, see Dolores Oliver Pérez, “Sarraceno: su etimología e historia” in al-Qantara: revista de estudios árabes Vol. 15 (Madrid 1994), pp. 99-130.

12. On the political rôle of hostages in Carolingian politics see Adam J. Kosto, “Hostages in the Carolingian World (714-840)” in Early Medieval Europe Vol. 11 (Oxford 2002), pp. 123-147.

13. Fernando de la Granja (transl.), “La Marca Superior en la obra de al-^cUdrí” in En la España Medieval Vol. 1 (Madrid 1967), pp. 447-546, cited, quoted at length and discussed by Salrach, Procés de Formació 1, pp. 77-90.

14. Eduardo Manzano Moreno, Las Fronteras de al-Andalus en época de los Omeyas, Biblioteca de Historia 9 (Madrid 1991), pp. 220-222.

15. Ibid. pp. 77-78; it should be noted that al-Udrī, whom Manzano says (p. 220) was basing his account on the tenth-century court chronicle of ‘Isa al-Razī, says that ‘Amrus took service with Matruh after the death of ‘Aysun, which must therefore have been before 791/2 when Matruh was killed (if there’s any basis to suppose al-Udrī cared about such trivialities).

16. Ramon Martí, “Palaus o almúnies fiscals a Catalunya i al-Andalus” in Hélène Debax (ed.), Les sociétés méridionales à l’âge féodal : Hommage à Pierre Bonnassie (Toulouse 1999), pp. 63-70, cited in idem, “Del fundus a la parrochia. Transformaciones del pobliamento rural en Cataluña durante la transición medieval” in Philippe Sénac (ed.), De la Tarraconnaise à la Marche Supérieure d’al-Andalus : les habitats ruraux (IV^e-XI^e siècle). Desde la Tarraconense a la Marca Superior de al-Andalus: los asentamientos rurales (siglos IV-XI), Méridiennes : Études Médiévales Ibériques 2 (Toulouse 2006) pp. 145-166; cf. Ollich, “Vic”, p. 91.

17. On Goth as a military identity, see Jesús Lalinde Abadia, “Godos, hispanos y hostolenses en la órbita del rey de los francos” in Federico Udina i Martorell (ed.), Symposium internacional sobre els orígens de Catalunya (segles VIII-XI), 2 vols, II pp. 35-74. On long ethnic continuities in the area, see Manuel Riu, “Hipòtesi entorn dels orígens del feudalisme a Catalunya” in Quaderns d’Estudis Medievals Vol. 2 no. 4 (Barcelona 1981), pp. 195-208.

18. I find that Salrach says that ‘Aysun, whom he follows de la Granja in associating with Aizó, fled to Córdoba after his rebellion was defeated and was there put to death on suspicion of complicity in a plot to murder the Emir (Procés de formació 1, p. 89 & p. 87 n. 37), but on inspection his source for this is Francisco Codera y Zaidín, “El Godo o moro Aizón” in idem, Estudios Críticos de Historia Árabe Española, Coleccioón de Estudios Árabes 7 (Zaragoza 1903), pp. 201-224. Codera, Salrach says, was basing himself on the thirteenth-century Arabic history of Ibn al-Athīr and the Ajbār Maŷmu^ca, a probably-originally-ninth-century collection of oral traditions, and as far as I can see it is not clear that they are telling the same story as al-Udrī (and al-Razī), not that that necessarily means it’s not preferable. Unless the Arab army was supposed to carry him into Barcelona and Girona, however, it’s not clear where the failure that would have set him fleeing is and, though Codera was an absolute pioneer in bringing the Arabic and Christian sources together like this, his claim to be doing ‘critical studies’ often looks sadly misplaced by our standards.

19. Ollich, “Roda”, pp. 86-87.

20. Jonathan Jarrett, Rulers and Ruled in Frontier Catalonia 880-1010: pathways of power, Studies in History (London forthcoming).