Murder of a factoid about Mallorca

My backlogged blog chronology is getting a bit out of step here, as I find this in my drafts folder from November 2018 referring to an unusual luxury I’d been able to permit myself that summer, which was a trip to Cambridge University Library. I was privileged enough to do most of my undergraduate and doctoral work out of that library (even though my doctorate’s from London) and there are still times when it’s invaluable to get there, for the simple reason that unlike most big research libraries a good proportion of its stock is on open shelves. The speed factor this adds to checking references is hard to exaggerate; when ordinarily you might have to wait an hour or two for your books to arrive, being able to go straight to not just the things you already knew you needed, but also then the things which they reveal you also need to check, which otherwise might normally mean a second trip at some future point, is invaluable. I was at that point up against a tight deadline to finish my article “Nests of Pirates” and, among other things, I was able to check the thing I want to tell you about now, and because of the open shelves track it to its root rather than just the next layer down. It’s about the Islamic conquest of Mallorca, I think by the definition offered some time ago by frequent commentator dearieme it counts as a ‘factoid’, and I think I killed it.¹

The Balearic Islands had a rough Late Antiquity.² Taking part fairly fully, as far as patchy archaeological evidence and a few textual anecdata can so far reveal, in both the third-century crisis (during which Mallorca’s then-capital, Pollentia, burned down) and then the general shrinkage of economy and settlement suffered by the western half of the Roman Empire over the fifth century, they fell in the course of that century into the maritime empire of the Vandal kings of Carthage, where they remained until returned to imperial control during the Byzantine conquest of that kingdom. They then remained under at least some kind of Byzantine obedience into the eighth century, to judge by seals found at the hilltop fort of Santueri in Mallorca, and perhaps even later, but that’s where the trouble begins, because the terminus post quem non is of course Islamic conquest, and we don’t really know when that happened.³

Wall of the castle site at Santueri, Mallorca

Now, first of all some important preconditions. There are five major islands in the Balearic archipelago, as you see above, in descending order of size Mallorca, Minorca, Ibiza, Formentera and Cabrera, of which Ibiza and Formentera sit apart as part of a separate group called the Pityuses, all over an area of about 150 square miles. There is much more evidence about what happened to Mallorca than any of the others; indeed, archaeologically and documentarily, we don’t actually have any proof so far identified that Cabrera, Formentara, Ibiza or even Menorca were actually occupied between the seventh (or for Menorca, eighth) and tenth centuries (mid-ninth for Menorca, as we’ll see), though it’s probably more likely that they were than they weren’t.⁴ Nonetheless, people who write about this area always seem to do so as if what happened in one island can be generalised to all the others, apparently believing that because they are governed as a unit now, and have been since, well, mumble mumble mumble, they must always have behaved as one. But it ain’t necessarily so.

Ruins of the late antique Christian basilica in Illa del Rei, its own separate islet off Menorca, by Pytxy – own work, licensed under CC BY-SA 4.0, Wikimedia Commons

So what we’re actually able to discuss is the Islamic conquest of Mallorca, not the Balearics, and we just have to live with that. Now, currently the archaeology can really only tell us that at some point people started using Islamic-style ceramics and burying like Muslims, not when or why.⁵ For when or why the answers must for now come from texts. The terminus ante quem this time is 933 CE, by which time there was an Andalusī (i e. from al-Andalus, Muslim Iberia) fleet using Mallorca as a base for raiding Christian Francia, according at least to the chronicler Ibn Ḥayyān, who had access to some of the caliphal archives of Córdoba somehow and thus had reason to know.⁶ So the conquest was before that, but when? Excitingly, there are four different dates recorded, all in different sources, none of which seem to know the others’ stories. For dramatic reasons it’s most fun for me to go through them from latest date to earliest, which means that the first entrant is fourteenth-century CE polymath, bureaucrat, lawyer and underrated sociologist, but questionable historian, Abū Zayd ‘Abd ar-Raḥmān ibn Muḥammad ibn Khaldūn.

Bust of Ibn Khaldun at Casbah de Bejaia, Algeria, image by Reda Kerbush – own work, licensed under CC BY-SA 4.0, Wikimedia Commons

Ibn Khaldūn, who had excellent sources for the most part, including Ibn Ḥayyān, but notoriously scrambled them in his own work, offers the latest date for the conquest, placing it in 902/903 and explaining it as the result of an Andalusī pilgrim to the East having been stranded in the islands for a while and having thus learned their weak spots, and then offering to lead a conquest of them for the Emir of Córdoba, which succeeded.⁷ Opinions vary among the people who know this report as to whether the islands could possibly still have been Byzantine at so late a stage, or had been assimilated into the Carolingian sphere after an appeal to Charlemagne for help against pirates in 798, which was answered.⁸ That appeal is documented in the Carolingian court chronicle, the Royal Frankish Annals, but only in its early version, not its revision of 829, as if by then it was no longer a working claim. So if the Balearics did swing Carolingian, it may not have lasted long.⁹ Nonetheless, after 903, says Ibn Khaldūn, they were Islamic territory.

Approach to Puig d’Alaró, Mallorca

But! The twelfth-century Granadan geographer Muḥammad ibn Abi Bakr al-Zuhri has a different story. According to him, the conquest took place in the reign of “Muḥammad, son of the fifth Emir of al-Andalus”, and although it was mostly successful, the “Rūm” held out in one particular fort, Ḥisn Alarūn, almost certainly modern Alaró seen above, for a further eight years and five months before finally running out of supplies and surrendering.¹⁰ So for him Mallorca was definitely still Byzantine territory, but when? Muḥammad I ruled 852-886 CE, but the trouble is that he was the fifth Emir, and his son was called al-Mundhir (r. 886-888 CE). Professor Juan Signes Codoñer has suggested that the peculiar way in which the ruler is identified might be explained if the name of Muhammad’s father ‘Abd al-Raḥmān II had dropped out, supposing a correct text saying “Muḥammad son of [‘Abd al-Raḥmān], the fifth Emir…” and that looks like a good solution to me, but it’s still a 30-year window.¹¹ Also, al-Zuhri said that he had heard this story told, not that it was a matter of record, and while he was nearer in time to the events than was Ibn Khaldūn, that distance was still four hundred years, so it’s not the best evidence.

But! Maybe we need neither of these, because the somewhat later Marrakech historian Abū al-ʽAbbās Aḥmad ibn Muḥammad ibn ʽIḏārī has a different report again. He, frustratingly, doesn’t tell us when the conquest actually was, but notes that in 848/849, none other than ‘Abd al-Raḥmān II sent a punitive naval expedition against Mallorca because its inhabitants had “broken their pact” and were harassing Islamic shipping in their waters. Some kind of blockade seems to have been imposed and next year Mallorca and Menorca both sent envoys begging for the renewal of the pact, though Ibn ʽIḏārī doesn’t say that they got it.¹² Nonetheless, as far as he was concerned, subjection of the islands to Islam, even if not conquest by it, had happened by then.

Modern depiction of Ibn al-Qutiya

Now it’s possible to reconcile that with either, but not both, of the previous two, and maybe even the Royal Frankish Annals, by saying that the Balearics, or at least Mallorca, had maybe been under a pact to the Muslim rulers in the Peninsula for some time but not actually conquered by them – this was also the case with Basque Pamplona, so it wouldn’t be unprecedented – and had perhaps flirted with a Carolingian alternative before being brought back into line by ‘Abd al-Raḥmān II.¹³ But! Now we reach the meat. Our earliest source for the conquest also gives the earliest date, or at least so it seems. If you look in the work of Juan Signes which I already mentioned, or that of Josep Amengual i Batle on which it often rests, you will find a report there that the tenth-century Sevillano lawyer and historian Muḥammad Ibn ʿUmar Ibn al-Qūṭiyya dated the conquest of the islands to 707/708 CE, when ‘Abd al-Malik, son of the then-governor of Muslim North Africa, Mūsā bin Nuṣayr, mounted a naval raid on Mallorca and Menorca and captured the “kings” (mulūk) who ruled there and sent them off to Damascus.¹⁴ I grant you that’s not quite the same as actual conquest, and it might even have led to the kind of pact subsequently reported by Ibn ʽIḏārī, but it is still quite surprising, not least because it’s a full four years before the actual conquest of the Iberian Peninsula, which Mūsā carried out in 711-712 CE. Now, that may just be presentism talking and it might have made more sense, before their permanent attachment (so far) to governments in the Peninsula in the thirteenth century, to see the islands as prone to African dependency, as under the Vandals and Byzantines, and not Iberian rule. But, this is also our factoid.

You see, when I first saw this report, in Signes I think, I was immediately struck by two things. One was how early it was, as I just said; but the other was that I really ought to have known about it because I own an English translation of Ibn al-Qūṭiyya’s History of the Conquest of al-Andalus.¹⁵ I’m not saying I’ve ever sat down and read it all through and made notes, but I have gone into it quite a lot for gobbets for my Special Subject at Leeds, and if I’d seen this bit I would have grabbed it because of what it implies about the conquest of the Peninsula following on naval raiding rather than being a spontaneous event.¹⁶ So I went and got my copy off the shelf, and this bit isn’t there.

Well, how odd, I thought. It seemed unlikely that Professor Signes had just made this up, so I looked up his reference, which was as I might have expected to the old Castilian translation of Ibn al-Qūṭiyya by Julián Ribera. And that was online then, so it was pretty easy to check that too and find that it’s not there either.¹⁷ But he gave a reference also to Josep Amengual’s two-volume history of the late Antique Balearics, and it was that which I arrived in Cambridge needing to check. And from that it became clear that the reference in question is in the Ribera volume after all.¹⁸ So for a moment it looked as if either David James had missed it out of his English translation, or it wasn’t in the manuscript of Ibn al-Qūṭiyya that he’d used. BUT! Not so! There are actually two texts translated in that volume of Ribera’s, Ibn al-Qūṭiyya and a similar history under the name of Abū Muhammad Abd-Allāh ibn Muslim ibn Qutayba, and our factoid is in the latter, not the former. Well, at least I now had the source. And in some ways it should be a better source, because although based in Iraq, Ibn Qutayba was writing even earlier than Ibn al-Qūṭiyya and brings the gap between source and event, otherwise so large in this saga, down to a mere 150-odd years.

But.

Shamefully, I knew nothing about this Ibn Qutayba, so I did some rapid research. That told me that Reinhart Dozy had written about this very text in the 1880s.¹⁹ And this is the great virtue of the Cambridge UL: having found this out, within ten minutes I could sit down again with Dozy’s work before me, and it gave me pause. Dozy had spent some time with this Ahādith al-Imāma wa’l-siyāsa of Ibn Qutayba and in the end concluded that it was actually nothing of the kind, partly because no such work seemed to be attributed to that author by medieval biographers, partly because it claims to have had the Mallorca report from eye-witnesses but it had supposedly happened 134 years before, and mainly because there is a reference in it to Maroc, a city not founded until 1062 CE, difficult for a ninth-century CE author to have added. Dozy’s conclusion was that the whole thing is an Andalusī “romancing” of the work of Abū Marwān ʿAbd al-Malik ibn Ḥabīb, with some extra Iberian material from who knows where, and that its unique content – of which the notice with which we’re concerned is part – is late eleventh-century at the earliest and probably without earlier basis. So, let’s stack all this up.

1. Some time after 1062 CE, someone decided to write a history of Muslim Iberia, which they based on an earlier work but to which they added some of their own material, including our notice about a 707/708 Muslim attack on Mallorca, and put the whole thing out under a respectable, plausible, but false name. Well, goodness knows they weren’t the last to do that, but then what?
2. In the 1870s, Reinhart Dozy, having got curious about this, went into it and discovered the forgery, and wrote the discovery up as one of about a dozen unconnected little studies in one of his less well-known works. Not many people seem to have noticed.
3. Juan Ribera had noticed, but for reasons that aren’t entirely clear thought that an earlier thought of Pascual de Gayangos that maybe this was more of Ibn al-Qūṭiyya still deserved to be taken seriously enough that the known text of Ibn al-Qūṭiyya and this one should be translated together, and so he did that.
4. Several people, including Josep Amengual, presumably searching quite rapidly for all references to Mallorca they could find, then found this one, but apparently did not realise that it was not actually within Ibn al-Qūṭiyya’s text.
5. Signes then presumably got the cite from Amengual, and repeated it even though he doesn’t seem to have been able to find it himself. And that paper, unlike Amengual’s book, went online and thus other people started to ‘know’ this thing as well. But unfortunately, it’s a dud…

Now, of course, this does not leave us with any kind of definitive answer. As I said in my article, you can even just about have it all: it could be that there was a 707/708 raid which sent the poor Mallorcan mulūk off to Syria, possibly even resulting in a pact to the then-governor of Ifrīqiya, which the islands then, finding that not keeping them safe, repudiated in favour of an approach to the Carolingians that didn’t last long, then went independent and possibly piratical for a bit before being reigned back in by ‘Abd al-Raḥmān II, but were not finally conquered till either the 850s-880s or 902/903.²⁰ Or maybe those two dates are even for different islands in the archipelago and both true! It would be a bit weird that not one of our writers seems to have known other stories if they were all true, but they’re not actually incompatible, and even if the 707/708 story is from three hundred years later, that’s still closer to the supposed facts than Ibn Khaldūn or Ibn ʽIḏārī. But what that story is not is the work of Ibn al-Qūṭiyya. And that is a dead factoid, thankyou very much.

---

1. What that means, of course, is that if you’ve read Jonathan Jarrett, “Nests of Pirates? ‘Islandness’ in the Balearic Islands and la-Garde-Freinet” in al-Masāq: Islam and the Medieval Mediterranean Vol. 31 (Abingdon 2019), pp. 196–222, DOI: 10.1080/09503110.2019.1600101, at pp. 199-209 and esp. 206-209, none of what follows is going to be new to you, sorry. But given firewalls and time, I’m betting that mostly you haven’t, and I can forgive you.

2. Covered ibid. pp. 198-212, but see also Luca Zavagno, “‘Going to the Extremes’: The Balearics and Cyprus in the Early Medieval Byzantine Insular System” in al-Masāq: Islam and the Medieval Mediterranean Vol. 31 (Abingdon 2019), pp. 140–157, and Miguel Ángel Cau Ontiveros and Catalina Mas Florit, “The Early Byzantine Period in the Balearic Islands” in Demetrios Michaelides, Philippa Pergola and Enrico Zanini (eds), The Insular System of the Early Byzantine Mediterranean: Archaeology and History, British Archaeological Reports International Series 2523 (Oxford 2013), pp. 31–45.

3. On the Santueri seals see Juan Nadal Cañellas, “Las bulas de plomo bizantinas del Castillo de Santueri” in Bolletí de la Societat Arqueològica Lul·liana Vol. 72 (La Palma 2006), pp. 325–340; they are the only part of the evidence from an extensive archæological dig that has been made public.

4. Jarrett, “Nests of Pirates”, pp. 199-204.

5. This is changing as radiocarbon dating begins to make a difference, as witness Miguel Ángel Cau Ontiveros, M. Van Strondyck, M. Boudin, C. Mas Florit, J. S. Mestres, F. Cardona, E. Chávez-Álvarez & M. Orfila, “Christians in a Muslim World? Radiocarbon dating of the cemetery overlaying the forum of Pollentia (Mallorca, Balearic Islands)” in Archaeological and Anthropological Sciences Vol. 9 (Cham 2017), pp. 1529–1538, DOI: 10.1007/s12520-016-0325-0, but as the title notes they had found Christian burials, not Islamic ones. The real problem is non-differentiation of pre- and post-conquest coarse-ware ceramics, which absent Islamic fine-wares makes it very hard to tell when a site stopped being used. Once this changes, it will probably no longer be viable to hypothesize non-occupation of any of the islands.

6. Dolors Bramon (ed.), De quan érem o no musulmans: textos del 713 al 1010. Continuació de l’obra de J. M. Millàs i Vallicrosa, Jaume Caresmar 13 (Vic 2000), &section;374; Jarrett, “Nests of Pirates”, p. 207.

7. Juan Signes Codoñer, “Bizancio y las Islas Baleares en los siglos VIII y IX” in Rafael Durán Tapia (ed.), Mallorca y Bizancio (Palma de Mallorca 2005), pp. 45–99 at pp. 84-85; Jarrett, “Nests of Pirates”, pp. 208-209. The opinion of Ibn Khaldūn is mine, not Signes’s!

8. Basically, the Byzantinists see it as Byzantine till the last possible moment (as witness Signes), the Islamicists see it conquered as early as possible, and the Catalans see it as taken over by the Carolingians and thus effectively gathered into the future Catalonia with some unfortunate Islamic interludes that however serve to justify Aragonese ‘reconquest’. There are, admittedly, exceptions to this in every group except the Byzantinists.

9. Jarrett, “Nests of Pirates”, pp. 205-206; the text is in English in Bernhard Walter Scholz and Barbara Rogers (edd.), Carolingian Chronicles: Royal Frankish Annals and Nithard’s Histories, Ann Arbor Paperback 186 (Ann Arbor 1972), online here, pp. 1-128, s a. 798.

10. Jarrett, “Nests of Pirates”, p. 208.

11. Signes, “Bizancio y las Islas Baleares”, p. 85.

12. Jarrett, “Nests of Pirates”, p. 208, accessible to me from Ibn Idari, Historias de al-Andalus, transl. Francisco Fernández González (n. p. n. d.), pp. 81-82.

13. On the situation of Pamplona see Juan José Larrea & Jesús Lorenzo, “Barbarians of Dâr al-Islâm: The Upper March of al-Andalus and the Pyrenees in the Eighth and Ninth Centuries” in Guido Vannini & Michele Nucciotti (edd.), La Transgiordania nei secoli XII-XIII e le ‘frontiere’ del Mediterraneo medievale. Trans-Jordan in the 12th and 13th Centuries and the ‘Frontiers’ of the Medieval Mediterranean, British Archaeological Reports (International Series) 2386 (Oxford 2012), pp. 277–288.

14. Signes, “Bizancio y las Islas Baleares”, pp. 46-54; Josep Amengual i Batle, Els orígens del cristianisme a les Balears i el seu desenvolupament fins a l’època musulmana, Els Trebals i els dies 36 & 37 (Palma de Mallorca 1991), 2 vols, vol. I pp. 441-453; there are other people I could cite, as well, but these notes are crowded enough and anyway I do at Jarrett, “Nests of Pirates”, p. 207 n. 64.

15. Muḥammad ibn ʻUmar Ibn al-Qūṭīyah, Early Islamic Spain: the history of Ibn al-Qutiya, transl. David James (London 2011).

16. I have, in fact, sat down and read all of it, while in bed in Venice in fact, and it has many excellent stories in it one of which will be a future blog post; but I don’t have notes. Anyway, while I’m trailing future blog posts we might note that if this thing were in there it would not be the only evidence for early raiding preceding the invasion of Iberia, as I will later disclose, and that was why I didn’t initially think anything need be wrong with the idea.

17. Signes, “Bizancio y las Islas Baleares”, p. 93 n. 5, cites Julián Ribera (ed./transl.), Historia de la Conquista de España de Abenalcotía el Cordobés, seguida de fragmentos históricos de Abencotaiba, etc., Colección de obras arábigas de historia y geografía que publica el Real Academia de Historia 2 (Madrid 1926), online here as of 10th July 2016 but sadly no longer, but he gives no page reference, I suspect because he could not find the passage in question either.

18. Amengual, Orígens del cristianisme, vol. I pp. 442-443, citing Ribera, Historia de la Conquista de España, p. 122, which is correct.

19. R. Dozy, Recherches sur l’histoire et littérature de l’Espagne pendant le moyen âge, 3^ème ed. (Paris 1881), 2 vols, vol. I, online here, pp. 21-40.

20. Jarrett, “Nests of Pirates”, p. 210.