THE EMERGENCE OF THE TAIFA KINGDOM OF TOLEDO

The period between the fall of the Umayyads of Cordoba and the emergence of the successor states in the Iberian peninsula is shadowy and unclear. In this article, I attempt '°^ The version of this passage in the edition of the iVâ i published by Ihsan 'Abbas, Rasa'il Ibn Hazm al-Andalusi, II, Beirut, 2nd edition, 1987, p. 58, shows no variations on these versions. It is the version of Seybold, with additions from the others. '̂ "̂ This is not entirely correct: it was a walled city, and appears occasionally in our sources in contradistinction, even in opposition, to Toledo; Toledan rebels against the last Dhu 1-Nûnid took refuge there after the failure of their revolt in 474/1082, and even tried to make themselves independent in the city {Rise and Fall of the Party-Kings, p. 255). See also M.-J. Viguera, «Madrid en al-Andalus», Actas IllJarique de Numismática Hispano-Arabe, Madrid, 1992, pp. 11-35 (I am grateful to the anonymous reader for Al-Qantara for pointing this out to me). '̂ ^ This study was prepared in Spain in the summer of 1998, which I spent working largely in the Departamento de Estudios Árabes of the Instituto de Filología in the Consejo Superior de Investigaciones Científicas in Madrid. It is a pleasure to express here my appreciation of the helpfulness and hospitality of my fiiends and colleagues in this institution, as also of the Spanish Ministerio de Asuntos Exteriores, which provided me with a fellowship under their programme for foreign hispanists which made possible my visit to Spain for this period. I am also, particularly grateful to Professor David Waines and to Dr. Maribel Fierro for reading this paper in draft form and giving me the benefit of their very penetrating criticisms; they are not, of course, responsible for my conclusions. (c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) http://al-qantara.revistas.csic.es 56 D. J. WASSERSTEIN AQ. XXI, 2000 to offer a micro-study of the process in one place. Using literary and numismatic sources, I attempt a reconstruction of events in and connected with Toledo, and of the list of rulers who were active there, in the first two decades of the fifth Islamic century. This list is much longer (though almost all those mentioned in it ruled very briefly) than was pre-


D. ]. WASSERSTEM
AQÍ XXI, 2t)ûi) raises questions about the nature of the transfer of power there from the caliphal-'âmirid system to the successor régimes.How did the transfer occur?Who were the successors?Why were they able to engineer their own takeover and survival?And how was it that it was they, rather than others, who were able to do so successfully?
In this article I offer a study of some of these questions, based on a microstudy of developments in Toledo.Such a study can be attempted only thanks to the survival of stray remarks and scattered facts in the sources at our disposal.They make possible a far more detailed study than has so far been attempted for this city and its history at this time.It may be that similar studies can be carried out for other cities, but it appears particularly desirable to pursue these questions in relation to Toledo both because of the significance of this city in the period before the coUape of the caliphal-'àmirid system, and because of the importance which the city possessed later, under the Dhù 1-Nünids, during the fifth/eleventh century.
Beyond this, a study of this kind also represents an attempt, from a methodological point of view, to see how far we can push the sources on issues which, while they possess great importance, cover very small ground chronologically and appear often to be scarcely covered by the material preserved in the source material which has survived.The Arabic texts of the middle ages which we have generally appear concerned with individuals and with events, rather than with groups and with processes.The specific texts on which this study is based are concerned primarily with religious scholars and scholarship on one hand, and with literary anecdotage on the other.Their overall nature is such as to suggest Üiat they can be of little help with larger questions, and, by extension, that such larger questions may not be capable of the profounder study which they, and we, demand.It emerges from this study that, within certain limits, we can indeed pursue micro-studies of individual developments even in places which are far from the centres of interest of those who composed the texts which are at our disposal today.In the light of the material and the interpretations derived from it which are offered here, this is a very comforting conclusion to be able to draw.

SOME TEXTS AND SOME PROBLEMS
We begin with chronicles.Two important sources for the fifth/eleventh century contain some previously unexploited material which enables us to begin the task of sketching in some further detail of the political history of the taifa of Toledo in this intermediate period.The first is the section of Ibn TdhM's al-Bayàn al-Mughrib dealing with the taifas.^And the second is the A 'mal al-A 'lam of Ibn al-Khatîb.6 In the A 'mal, Ibn al Khatïb says the following: [The B. DM 1-Nün] were not possessed of leadership or of nobility until the reign of al-Mansür Muhammad Ibn AM 'Amir; at that time they advanced, and became famous, and led armies, and settled in the küra (province) of Shantabariyya [= Santa ver].Now when 'Abd al-Rahman b.Matyûh seized power in Toledo, and then after him Abd al-Malik b.Matyüh, and treated its people badly {asá'a al-sira fi ahliha), they deposed him, and agreed to send to Ibn Danniin (sic).And he sent Ms son Ismà'ïl b. 'Abd al-Rahman Ibn Dannun to them from Santaver7 Ibn TdhM has the same story, but in rather more detail, and it is worth giving it in full: They [scil.the B. DM 1-Nûnj advanced in [al-Mansùr'sJ reign, and became famous; some of them led armies and governed provinces (a'mal) and cities; and one of them, at the end of the time of the caliphate (jama *a), was governing the küra of Shantabariyya.And when the fitna broke out in al-Andalus the governor in the city of Toledo and its environs was 'Abd al-Rahman b.Manyüh, but his fate came to Mm during this [period], and 'Abd al-MaUk b.Manyûh inherited his post (nazar), and treated the subjects (ra 'iyya) badly.Now the people of Toledo were from of old people of dissension (fitna) and rebellion against the kings, and they did not like the conduct (slra) of this fatà («young man», «slave») so they deposed him and appointed someone [else] over themselves to manage their affairs.Then they became hostile to him for some reason and deposed him [as well].Then they decided to write to Ibn DM 1-Nûn in Shantabariyya, and he sent his son Ismâ'îl b. 'Abd al-Rahman Ibn DM 1-Nün to them, and this fatà («young man», i. e., Ismâ'îl ^ The section of Ibn 'IdhM's Kitàb al-Baym al-Mughrib fi akhbàr mulük al-Andalus wal-Maghrib dealing with Islamic Spain in the fifth/eleventh century was published by E. Lévi-Provençal, Ibn 'Idârî al-Marrakusî, Al-Bayàn al-Mughrib, tome troisième.Histoire de l'Espagne musulmane au XP siècle, Paris (Textes arabes relatifs à l'histoire de l'occident musulman, 2), 1930 (repr. Beirut, n. d.).

AQ. XXI, 2000
THE TAIFA KINGDOM OF TOLEDO 21 acquired prominence only under al-Mansür (i.e., in the last third of the fourth/tenth century) need indicate no more than that such prominence as they possessed before that time was very local in its effects (Santaver is only some one hundred and fifty kilometres from Toledo, to the north-east).Toledo itself is well characterised by Ibn Tdhârî as a place whose inhabitants were «of old people of dissension and rebellion against the kings», as its history under the Umayyads amply demonstrates.^^And the account given by both writers of the process by which the Dhü 1-Nünids came finally to rule the city tallies well with the facts as deduced by Dunlop.^^Nevertheless, there are difficulties with the two passages: neither makes any mention of Ya'ish, the qàdi who apparently ruled Toledo before the Dhù 1-Nünids, unless we should see a reference to him in one of the two un-named rulers who are mentioned by Ibn 'IdhM.The two B. Matyuh (or Manyuh), in then* turn, do not appear to be known to history as rulers of Toledo through references in other sources (although, as will be seen, one of them at least is not totally unknown), and the same is true of at least one of the two un-named rulers just mentioned (the other, of course, may be Ya'ish).These are not, however, great difficulties: Ya'ish, as we have seen, may be one of those two un-named rulers; the two B. Matyuh (or Manyuh) seem both to have been in power for a very short time, and probably thus had little lasting effect on the city and its fate; silence about them in the sources may reflect an ignorance of their existence bom of the general confusion of the period; the same is even more true of the unnamed rulers.Ya'ish himself comes through the sources as a very shadowy figure, who withdrew after his deposition to a quiet and apparently undisturbed retirement in Calatayud.^"*There seems to be no good argument against accepting the evidence of these two sources (or of their single common source) for events at Toledo in this period, so far as it goes.However, in doing so, as will be seen, we come up against some difficulties.In The Rise and Fall of the Party-King s I gave the beginning of the list of rulers of the taifa of Toledo as follows:'^ AQ.XXI, 2000ca. 403/1012-13 Abu Bakr Ya'îsh b. Muhammad b. Ya'ïsh, al-qàdi.409/1018-19 Abu Muhammad Ismà'ïï b. 'Abd al-Rahmân Ibn DM 1-Nün, Dhü 1-Riyâsatayn, al-Zàfir.This list has now to be modified, by the addition, or the insertion, of at least three, and possibly four, other rulers before Ya'ïsh.Who were these rulers, and what dates are to be assigned as the beginnings and ends of their periods of rule?By about 403/1012-13 Ya'ish may have been in control in Toledo, and he remained in power there possibly until no later than 409/1018-19.There are difficulties with both of these dates, which are discussed below, but for the moment they offer a convenient working framework, and if they do in fact stand in need of correction, such correction probably affects the overall chronology, not the inner detail of my conclusions here.'^Between 399/1009 and 409/1018-19 Toledo, or those who ruled there, played an important rôle in the conflicts around the caliphal title and the power that was still thought to accompany it.(The quietude of Ya'ïsh and of the other rulers of Toledo in respect of these conflicts was as important, in its way, as the participation in this struggle of this strong and potentially significant city would have been.)Two of the major characters in this process, for a short time, were the caliph al-Mahdî and the Slav Wàdih, both of whom were active in Toledo.
The first of these, al-Mahdî, fled to Toledo, where Wadih held the marcher territories loyal to him, in Jumada 1400/December 1009-January 1010, when he was ejected from Cordoba, but he left the city shortly afterwards to attempt a return to the capital.In Shawwal 400/May-June 1010 he succeeded in this endeavour, only to be assassinated there a few weeks later, in Dhu 1-Hijja 400/July 1010, by Wadih, his main supporter.'^Wadih, who appears to have harboured the ambition to be another Mansùr, then re-installed Hishám II al-Mu'ayyad (?redivivus) on the caliphal throne, and had himself named by him as hàjib.In hopes of removing the other challenger for the caliphal throne, Sulaymàn al-Musta'în, as a serious contender for power, he sent the head of al-Mahdi to his camp, near Cordoba, with proposals for an agreement.The Berbers supporting Sulaymàn were not interested, however, in any arrangement which returned Hishám, especially with a Slav as his main minister, to power, and this overture by Wadih was rejected.Sulaymàn al-Musta'ïn himself displayed much grief over the killing of his rival, and sent the head to a son of al-Mahdï, called 'Ubayd Allàh.^^'Ubayd Allah b. al-Malidï, who was no more than sixteen years old, had been in Cordoba at the time of his father's murder by Wadih.With the help of partisans of his father in the capital, however, he had managed to evade capture by him and now succeeded in making his way to Toledo.According to one source, he was well received by the Tbledans, who even gave him, despite his youth, authority over themselves.This 'Ubayd Allah b. al-Mahdï, if our source for him is to be relied upon, seems thus to be the first ruler of an independent Toledo after the fall of the caliphal-'àmirid régime.
Our information about this young man and his career is slight and confusing.His existence is recorded for us, unlike that of so many other minor Umayyad princelings, for virtually one reason only: because he was the son of al-Mahdï, and in the context of his father's murder.Apart from this context (and not fully detached from it), al-Maqqan preserves some verses by this 'Ubayd Allah in his Nafh aUTib?^Here al-Maqqarï adds that he is said to have been known as «alaqra'» («the bald one») an odd nickname for a sixteen year old, but plausible if AQ. XXI, 2000 he was indeed bald at that age.Beyond this there is obscurity, and that obscurity is deepened by the rest of our scanty material.In a couple of places we hear of someone known as Ghulam (sic) al-Fasïh al-Andalusî who claimed to be this 'Ubayd Allah b. al-Mahdî, succeeding in persuading some people of the truth of this claim, and we are also given some verses (not the same as those in the Nafh al-Tib) alleged to be by him.^'The name «Ghulam al-Fasih al-Andalusî» is awkward: «the boy (belonging to) the Eloquent Iberian»?or even «the Iberian boy (belonging to) the Eloquent (person)»?A name, as such, seems to be lacking.^^ The Naqt al-'Ams of Ibn Hazm offers some evidence which touches on this latter question, but deepens the mystery surrounding this character.In the version of this text published by C. Seybold, we have: This is translated by L. Seco de Lucena: «Un hombre que pretendía pasar por 'Ubayd Alláh al-Mahdí se sublevó en Madrid, afincándose sólidamente en esta ciudad, hasta que lo mataron.No es cierto que fuese el auténtico 'Ubayd Alláh.Por el contrario, me consta que se trataba de un eslavo del droguero conocido por al-Fasíh».^"^Several features of this very short passage are puzzling, or worrying.First, we note the geographical problem: Madrid is not Toledo.While it is not too far from Toledo, it is, for all that, a different place.
It is noteworthy at once, further, that the word ibn is absent between the name 'Ubayd Allah and the title al-Mahdï: a slip? an error?ignorance?The word mamlük attracts attention here too, as suggesting slave-status more definitely than the word ghulàm, which need not mean much more than «youth».The presence of the term «al-'Attàr» is also strange: is this a description of the man's profession? or is it merely a professional label which has passed over into being a name?We have innumerable examples of this process, even for this very label.^^And the word al-ma'rüf, «known (as)» looks a little like a way of dealing with what may have appeared to be an obscurity in an earlier version, such as our other versions, in which only the term «al-Fasïh» occurred.While each of these difficulties is fairly insignificant on its own, together they cast a thicker cloak of obscurity over what is already a difficult scene.
This version of the account of this young man is repeated in the version of the text published by Dayf (pp.58-59) (with the word Cr!, missing in the Seybold-Seco text, correctly in place between the name of 'Ubayd Allah and the title al-Mahdî) and in the version of this text preserved in the manuscript (n.° 5374) of the Naqt in the Chester Beatty Library in Dublin (at f. 4 r) (also there with the word CH in place; there are some other slight differences between the three versions: see Appendix on this);^^ but in both there is a significant additional piece of information which is lacking in Seybold's manuscript: between the word qàma, «rose up», and the word bi-majrït, «in Madrid», these versions have the words 'ala al-Mustakfï, «against al-Mustakfí».This can only be a reference to the later Umayyad caliph (regn.414/1024-416/1025), and given the relationship between Ibn Hazm, the author of this text, and that caliph, if these words are original parts of the text, this would make the passage still AQ.XXI, 2000 more interesting: in such a case we should have an apparent reference to a man claiming (presumably falsely) to be a son of al-Mahdï, rising in revolt some ten to fifteen years after the deaths of the genuine son and of al-Mahdî himself.Could this have been an obscure Mahdist revolt?In the light of what follows, this may be of some relevance.^^*lJbayd Allah b. al-Mahdï is said to have stayed in Toledo for a time, but finally he decided to make an attempt on the caliphal title in his own right, as heir to his father.He was quickly defeated in this attempt, and sent to Wàdih in Cordoba, who had him put to death.^^Wadih himself found able rivals in treachery among his own supporters in Cordoba.One of them, 'All (or Ahmad) Ibn Wada'a, killed Wadih on 15 Rabr II 402/16 November 1011, inaugurating a new stage in the uncertainty and confusion in the political scene in the country's capital.^^Although we have no explicit testimony as to the dates for the arrival of 'Ubayd Allah b. al-Mahdï in Toledo and his departure from there, it is nevertheless possible to establish the outside dates for these events.He must have arrived in the city at some point soon after the killing of his father in Cordoba by Wadih, an event which took place in Dhü al-Hijja 400/July 1010.He is unlikely to have reached the city very much before the beginning of Muharram 401/August 1010.His own killing by Wadih will have occurred before the death of Wadih himself, in Rabí' II 402/November 1011.This gives a maximum of some fifteen months or so for the reign, if it may be so termed, of this youth in Toledo (and/or Madrid).
^"^ Makkï, M. A., «A propósito de la revolución de 'Ubayd Allah b. al-Mahdî en Madrid», Revista del Instituto Egipcio de Estudios Islámicos en Madrid, IX-X (1961IX-X ( -1962)), 255-60, also discusses this impostor, identifying him correctly, on the basis of Dayf's edition of the Naqt (and correcting a mistaken hypothesis of J. Oliver Asín as to the nature of the rebellion which he led); he also translates the verses ascribed to the impostor into Spanish.
^^ Ibn 'Idhari, Bayàn, HI, 100; here the person who defeated and captured 'Ubayd Allah is called «Muhârib al-Tujîbî»: it is not clear whether this is to be understood as a name or simply as a description(?«the Tujibid fighter», «the soldier of the Tujibid»), in somewhat inelegant Arabic.See also, on this whole episode and the period in general, the passages from al-Nuwayri translated by Pascual de Gayangos, in his translation of parts of the Nafh al-Tib of al-Maqqari, The Mohammedan Dynasties in Spain, 2 vols., London, 1840-43, ii, 496-97, and also p. ix there (These passages appear not to be among the texts published in Arabic by Gaspar Remiro).
^^ Lévi-Provençal, HEM^, 318.Lévi-Provençal (following Ibn Hayyán) calls him 'All, but he is called Ahmad by, e.g., Ibn 'Idhari, Bayàn, m, 93.Lévi-Provençal, loe.cit., gives the Christian date as the middle of October, not of November, but this is presumably a slip for Rabí' I, as he gives the correct hijñ date.On Ibn Wada'a see also al-Humaydï, Jadhwat al-Muqtabis, ed. I. al-Abyàri, 2 vols., Cairo andBeirut (al-Maktabat al-Andalusiyya, 7-8), 2nd éd., 1410/1989, n, 499-500;Ibn al-Abbàr, Kitâb al-Hulla al-Siyarâ', ed. H. Mu'nis, 2 vols., Cairo, 1963,1, 282-83, n.° 105, with examples of his poetry, and, for an unpleasant anecdote about his meanness, Ibn Bassam, Dhakhira, ed.I. 'Abbas, Beirut, IV part i, 1399/1979, 53-55.It may be possible to narrow this period down somewhat.There is an obscure hint in our sources that the arrival of 'Ubayd Allah b. al-Mahdi in Toledo may have been rather later than the beginning of 401/August 1010.We have a report that a certain Abu 'Umar Ahmad b.Muhammad Ibn Wasim of Toledo «raided with Muhammad Ibn Tammâm towards Maqueda, and when they were defeated he fled to Cordoba; but the people of Toledo sought to prosecute him (Ar.ittaba 'ühü) during the rule of Wadih {scil in Cordoba), and they won possession of him and crucified him; and he said at that time, "That was written in the Book (kàna dhàlika fi al-kitàb mastiir^'^)"»?^ Ibn Bashkuwàl uses Ibn Mutáhir as his source for this information.^^And Ibn Bashkuwàl adds, from Ibn Hayyân, that the date of this Ibn Wasïm's death was Rajab 401/February-March 1011.^^The edition of the Sila prepared by Codera has an isolated sentence which says: «and the crucifixion of Ibn Wasim was on Tuesday 5 Sha'ban 401/14 March 1011».^^Al-Husaym's edition of the Sila has the same miscellaneous material as Codera's, but with some minor textual differences, and, more importantly, without this sentence or any reference to it, which is strange, and from our present point of view irritating, as it tends to weaken any value that the datum might have.^"^The same biographer, Ibn Bashkuwàl, has an entry on another Toledan, presumably to be identified as the second one mentioned above, «Abu 'Abd Allah Muhammad b.Tammàm b. 'Abd Allah» of Toledo, who «was killed by the people of Toledo in 400/1009-10 or 401/1010-11», the date 400/1009-10 being furnished by Ibn Hayyàn.^^AQ.XXI, 2000 These two sets of remarks are very puzzling.Do they indicate a revolt by some part of the Toledan population in support of 'Ubayd Allah?And if so, do these two executions of prominent men of religious learning contain a hint of some religious element in 'Ubayd Allah's revolt?The implication of Ibn Bashkuwàl's entry on Ahmad b.Muhammad Ibn Wasîm could well be that, at least in Rajab-Sha'ban 401/February-March 1011, Toledo, or some sections of her population, enjoyed good relations with Wadih, then in control in Cordoba, and wanted to maintain such ties.
Regardless of how we answer these questions, what we have here does suggest that we may be justified in dating the arrival of 'Ubayd Allah b. al-Mahdi in Toledo a little differently.There are two, or even three, distinct, though slightly overlapping, possibilities.The first of these is to see his arrival in Toledo as occurring ca.Muharram 401/August 1010, soon after the death of his father, and his departure in a doomed revolt against Wadih as occurring around Rajab-Sha'ban 401/February-March 1011, less than six months later.In this case, we should see the deaths of Ibn Wasïm and of Ibn Tammám as part of the failure of what may perhaps be termed a Mahdist legitimist revolt in support of this 'Ubayd Allâh.36The second possibility is to see these two executions of learned Toledans as not, or not necessarily, tied to 'Ubayd Allah's revolt.In this case, the young man's arrival in Toledo will still have occurred ca.Muharram 401/August 1010, quite soon after the killing of his father, and he will have departed from there, in his attempt on the caliphal throne, at an unknown date before Wadih's own murder.This is the possibility that we saw before.But this possibility can in fact be split in two: 'Ubayd Allah may have arrived in Toledo at an unknown date before the deaths of the two scholars; or he may have reached the city at an unknown date after these events.Given his own situation, it may be preferable to see his arrival as having occurred sooner, rather than later.But the evidence is insufficient to enable any decisive conclusions to be drawn on this question.
One element in the scraps of information that we have about these two scholars may encourage us, nonetheless, to see their deaths as connected with the fate of this young man.The identities of Ibn Wasîm and of Ibn Tammám, as men of rehgious learning, are what made them worthy of entries in a biographical dictionary of scholars, and hence what made information about them survive.But their character ^^ It would perhaps be just a little excessive to see this as an attempt to create a mahdist (as distinct from Mahdist) movement, but the title adopted by the father of this 'Ubayd Allah, together with what we seem to know about an impostor using his name more than a decade later (see below), may nevertheless afford some justification for the use of this label.as religious scholars, the manner of their deaths, and the nature of the material that we have about 'Ubayd Allah, as rebelling from a base in Toledo (or at least close to it, if we are to accept the reports connecting his revolt to Madrid), as well as what is reported above of the career of the second man claiming to be the son of al-Mahdî, all suggest that the rebeUion of 'Ubayd Allah may have contained an element of some sort of mahdist (? or Mahdist) legitimism.-^^Against this possibility, we should of course weigh the fact that in neither of these two cases are we told explicitly that the execution was in any way connected with 'Ubayd Allah.But this may not be a very strong objection, given the allusive and brief manner in which the biographical dictionaries so often speak even about such matters as these.And the fragmentary nature of at least part of these particular remarks may also hide a good deal.^^TOLEDO AFTER THE UMAYYADS: THE B. MATYÜH It would appear from this that the periods of rule in Toledo of the two B. Matyüh and at least one of the two un-named rulers mentioned by Ibn 'Idhârî (for Ya'ïsh may, of course, have been the second of these) must be placed within a very short period.They must have occurred between the departure of 'Ubayd Allah b. al-Mahdï from Toledo (at some unknown date before Rabí' II 402/October 1011) and the beginning of the rule there of Ya'îsh, probably at some stage in 403/1012-13.39 Now Ibn 'IdhM also tells us that «when thtfitna broke out in al-Andalus the governor in the city of Toledo and its environs was 'Abd al-Rahmàn b.Manyüh».'^^The meaning of the term fitna is fairly clear: normally it means ^^ Cfn previous note.38 Ibn al-Faradi, Ta'rîkh 'Ulamà' al-Andalus, ed. Codera, 2 vols., Madrid, 1891-92, II, 36, n.° 1.524 (= ed. I. al-AbyM, 2 vols., Cairo-Beirut, 1410/1989, II, 882, n.° 1.524), is an entry for Abu Bakr Wasîm b.Ahmad b.Muhammad b.Wasîm, «of the people of Qurtuba», but there is no further information; at Ibn Bashkuwal, Sila, ed. Codera, 2 vols., Madrid, 1883, II, 585, n.° 1.301 (= ed. Husaynï, Cairo, 1374/1955, II, 610, n.° 1.415) «civil strife», but it is applied in the context of Ibero-Islamic history to the period following the collapse of central Cordobán authority.However, it is not always wholly clear just when this period is to be understood to have begun for any specific region or city.In this particular case, however, it seems likely that Ibn 'IdhM used the term in a fairly loose sense, for the activities of 'Abd al-Rahman Ibn Matyûh can be traced with some exactness for just the period in question.
The historian Ibn Hayyân reports that this *Abd al-Rahman Ibn Matyûh was in the army of Sulaymán al-Musta'ïn, outside Cordoba, at the time of the murder of al-Mahdï by Wâdih, in Dhü l-Hijja 400/July 1010.Hostile to al-Mahdï, 'Abd al-Rahman Ibn Matyûh was delighted to hear of his death, and decided to abandon Sulaymán al-Musta'în and change his allegiance.He entered into correspondence with Wâdih, al-Mahdï's killer, and soon reached an agreement with him, by which he returned to Cordoba and appears to have been given a share in power there.We are told that there he «ran Hishám's affair [s] 'Andalousie, Cairo, 1948 (repr. Beirut, n. d.), 88-89, for a life of Ibn Wâfid; for the exact date of the appointment of Ibn Wafid cf.Viguera, M.* J., «Los jueces de Córdoba en la primer mitad del siglo xi (Análisis de datos)», Al-Qantara, 5 (1984) "^"^ The references here to Ibn Wàfid expressing clear dislike of accepting a post under the state, and to Hishám, the ruler, applying pressure to him to accept it are both reflective of conventional attitudes and behaviour; while they may also reflect genuine attitudes, especially on the part of Ibn Wafid, we have no particular reason to suppose that they do, and, in the earlier part of the text translated here, in the reference to «love of poweD> (if this has been correcüy understood), we may have testimony to the pure conventionality of these expressions.We cannot really know.and wealthy house, and he was extremely well-educated both in general culture and in traditions, but he had little knowledge of fiqh and (had to be) compelled to (accept) the ^âçiï-ship; he maintained a good rule, persisting in trying to resign until 'Abd al-Rahman Ibn Manyüh left Cordoba,"^^ and then Hisham removed him from office and brought back Ibn Wàfid, as has been related."^^These statements suggest that 'Abd al-Rahman Ibn Manyüh was in Cordoba at least during the period from the death of Wadih until as late as 22 Rajab 403/6 February 1013, when Ibn Wàfid was returned to the office of chief qadi, shortly before the takeover by Sulaymàn al-Musta'în.They also suggest that he played an important rôle in affairs there at this time, particularly in trying to reach an acconmiodation with the Berbers supporting Sulaymàn.A report of a letter sent by Hisham to Sulaymàn in the year 402/1011-12, in which a proposal for just such an accommodation was made (a proposal which was rejected by the Berbers), occurs in Ibn Tdhàrï; although the exact date in the year 402/1011-12 is not given, it is clear from the political circumstances and other details that this must belong to the period of Ibn Manyüh's power in Cordoba.^^It is possible, however, to show that Ibn Manyüh actually left Cordoba rather earlier than the takeover by Sulaymàn.In Ibn Tdhàrî we have another statement to the effect that he left Cordoba before the end of 402/mid-1012.At the very beginning of Dhü al-Hijja 402/late June 1012, he is reported, together with the leaders of the slaves {'abid: or Slavs?) and the army (jund), to have gone to Hisham II al-Mu'ayyad and to have explained the seriousness of the political and military situation to him in graphic detail.The caliph's reaction, natural in a ruler who had suffered three decades of forced political inactivity under the tutelage of the 'Àmirids, was one of helpless despair; he told them to do whatever they deemed best.Ibn Manyüh interpreted this in exclusively personal terms, and opted for flight.Taking with him a huge amount of treasure, he made for Badajoz.^^ We have what appears to be a confirmation of his activity there at about this time in a report on the early history of that city in the taifa period.^^If the reports of him as ruHng in Toledo are to be accepted at their face value, then the only period during which he (and after him 'Abd al-Malik Ibn Matyùh) could have done so seems to be after his departure from Cordoba at the end of 402/mid-1012 and before the rise of Ya'ïsh at some unknown date in 403/1012-13.This is a maximum of some twelve months.
The extreme brevity of this time for two periods of rule by named persons and at least one by an un-named person (two if Ya'ïsh is not the second un-named one mentioned by Ibn Idhári) raises a difficulty with this interpretation of the material.Another difficulty, more serious than this one (which can, after all, be resolved, if not very satisfactorily, by the assumption of very short periods of rule in Toledo for each of these individuals), is raised by the reports in Ibn Sa'id.
We learn from these reports that Ibn Wafid, opposed to an accommodation with the Berbers, was dismissed from office on 9 Dhü 1-Hijja 402/2 July 1012, as a result of his differences with Ibn Matyüh over policy towards the Berbers; we are also told that he remained under house arrest thereafter until Ibn Matyüh's departure from the city; and we learn, further, from the same source that, following a period in office by Ibn Bishr, Ibn Wáfíd was recalled to office by Hishàm on 22 Rajab 403/6 February 1013.Unfortunately, we learn from Ibn TdhM that Ibn Matyùh in fact left the city at some stage in Dhû 1-Hijja 402/July 1012, i. e., at around the very time of his opponent's dismissal.This would have appeared to be the obvious time for his opponent to be recalled, not dismissed.
Of the possible ways of resolving this difficulty, none is entirely satisfactory.One is to assume that in fact the Ibn Matyùh whom we find active in Cordoba at this time is not the same as the one reported to have been in power in Toledo; such an assumption could be supported by the fact that we have two different sets of forms for this name: we have forms like Ibn Matyùh (and Manyùh, which differs from Matyùh by no more than a single dot) and we have others like Ibn Munàw^": unfortunately, while this is superficially a very attractive solution to this difficulty, the two sets of forms of the name occur in ways that make it impossible to accept.The form Ibn Munaw*" is the form given by Ibn TdhM on some occasions, but the form Ibn Matyùh/Manyùh occurs not only in Ibn Sa'îd and in Ibn al-KhaGb when discussing Toledo but also in Ibn Tdhan himself on the same subject.No good which he is not otherwise known to have held (though it is not inherently impossible that he should have adopted it during his domination of Hishàm n al-Mu'ayyad; and see Appendix); Ibn Hayyàn tells us that «Sibür al-'Àmiii, one of the associates (? Ar. sibyan) of Fà'iq the khâdim, thefatâ of al-Hakam (scil.al-Hakam al-Mustansir), had declared himself independent in (or "seized control of) Badajoz and the western thaghr min 'amal al-hâjib Ibn M.yJ.w.y.h».It is not entirely clear what the expression min 'amal is intended to mean in this context, but the account does appear to provide an explanation for Ibn Matyüh's choice of Badajoz as a first refuge on fleeing Cordoba.argument can be brought forward in favour of one form of the name as against the others (except that which sees a pun in the shape Manyüh).At the same time, there is no good argument either for seeing two separate individuals behind the two forms; given their rarity and their unusual character as Arabic names, they could very easily have been corrupted in the normal course of transmission of names in Arabic manuscripts once their original forms were no longer current.
The argument against the reading M.t.y.w.h, and for the reading M.n.y.w.h, is perhaps strengthened by Ibn 'Idharfs phraseology: «And when the fitna broke out in al-Andalus the governor in the city of Toledo and its environs was 'Abd al-Rahman Ibn M.n.y.w.h, but his fate (Ar.maniyyatuhu) came to him during this (period)».Would it be stretching the text, or the author's intention, to see in this choice of words a pun on the man's name?^^Another possibility is that we should understand the Arabic form in these names to represent something like Muño (from which we get the modem Muñoz).This would suggest a Christian background for the bearers of these names, which may be of significance in what follows.^*^A second solution is to see the awkwardness as resulting from errors in the sources and in their sources of information about what was, after all, a very confused period, but this is in reality to suggest that no sense can be made of the information in the sources; it is a counsel of despair.
A third possibility, which perhaps commends itself a little more than these, lies in a consideration of the development of events in Cordoba at this time: at the beginning of Dhù al-Hijja 402/late June 1012, Ibn Matyùh, according to our sources, explained the seriousness of the situation to Hishám; the caliph told him to act as he saw fit; on 9 Dhù al-Hijja 402/2 July 1012 he, acting through Hishám, dismissed the chief qàdï Ibn Wafid, who was in part responsible for the failure of his policy of peace with the Berbers; it is at this point that we should place the letter which was sent from Cordoba to Sulaymán and the Berbers, and rejected by them; seeing that the situation was by then completely hopeless, Ibn Matyùh will at this point have thrown up his hands and taken refuge in flight, first to Badajoz and then on to Toledo.The long delay between that time and the re-appointment of Ibn Wafid as chief qàdl mñy then be explicable either as due simply to (fully understandable) indecision on the part of Hisham or as a result of a perceived lack of need for a new qàdl, as Ibn Bishr occupied the post.
The dates of the periods of rule of the two B. Matyûh in Toledo thus become a little easier to define: as has been seen, the first of them did not leave Cordoba before somewhere round the middle of Dhù 1-Hijja 402/mid-July 1012, and will have arrived in Toledo some little time thereafter, possibly after a visit to Badajoz, to cement his relations with Sábür.Unfortunately, we do not know how long they remained in power in Toledo.This is of significance also for the reign of Ya'ish, as will be seen below.If we do not know how long they remained in power, we do, nonetheless, have a hint as to how their rule there ended.In the Naqt al-'Ams of Ibn Hazm, a miscellaneous collection of historical anecdotes, we are told that «'Abd al-Malik b. 'Abd al-Rahman b.Munùh died at the hands of a slave (or: a Slav) who was defending himself».^^This is a very isolated remark, and the details remain obscure.The expression «defending himself» (which Seco renders mistakenly as «que tenia para que protegiese su persona»), lends itself to at least two, not wholly exclusive, interpretations: 'Abd al-Malik may have been making a violent attack on a servant, or he may have been making a homosexual advance which was unwelcome.The Arabic permits either interpretation; and parallels in support of the second one can be found, for this very period.Nevertheless, whatever the correct interpretation of the background, this may well represent something of the end of this obscure pair of rulers in a very confused period.Such an end for the second of them would also help to offer some explanation for the readiness of the local qàdi to step in as ruler.While we do have examples of qadis as rulers, they are few, and the reason is easy to understand.Qadis generally display, or are represented in our sources as displaying, reluctance to assume even the office of judge, because of the fact that such a post would put them in the position of judging their fellow-Muslims.Rule in the state, political power, as being something higher and beyond the status of a judge, would naturally be even less welcome to most qàdis.Things should begin to be clearer as we enter the calmer water of longer, more settled and more solidly documented periods of rule by better-known personages in Toledo.However, experience shows that where Toledo is concerned things are rarely simple.The reign of the qàdi Ya'ïsh presents a number of aspects of interest, and as many matters of difficulty.We do not know exactly when Ya'ïsh came to power in the city.It is not wholly clear that he ruled alone.Despite the reports which survive, we know too little about his exercise of power there during his reign.And we cannot be sure when his reign came to an end or when the Dhü 1-Nùnids came to power.
If the analysis offered here is correct, it may be reasonable to suggest that if Ya'ïsh became ruler of Toledo in the year 403/1012-13, then he will have done so relatively late in the year, as at least three and possibly as many as four other rulers will have preceded him in power there in the same year.The brevity of the period of rule of each of these in Toledo could thus also be part of the explanation for the ignorance of their names in the sources, unless, just as possibly, the shortness of the reigns of each of whatever number there were meant that a larger number than really existed was posited in the sources in error.One such instance could be the one (or two) un-named ruler(s) mentioned by Ibn 'IdhM, one of whom could be a reference (out of place) to 'Ubayd Allah b. al-Mahdï.On this, however, certain knowledge is unUkely unless major new sources become available.
There is, however, a further set of problems concerning the dates and the nature of the period of rule of Ya'ïsh.So far, I have placed the start of his reign in 403/1012-13 because of his apparent involvement as ruler in the murder of Ahmad b.Sa'ïd Ibn Kawthar in this year, and the end in 409/1018-19, because of a report that the Dhü 1-Nûnids acquired power in Toledo in that year.However, it is possible, even perhaps probable, that both of these dates are mistaken, both by a number of years.Ya'ïsh may have begun ruling there some time later than 403/1012-13 (quite possibly later even than 409/1018-19) and the Dhü 1-Nûnids may have taken over as late as 417/1026.And Ya'ïsh may, though less probably, have ruled for at least part of the time in conjunction with at least one other person.The problems arise in information about Ya'ïsh supplied by the qàdi 'lyád and others.
Writing their fathers.They were both honest men (Ar.wa-kànà 'ala safa').Now Muhammad b.Ya'îsh was outstanding among his peers in knowledge (*//m), until there occurred some rivalry between the two of them in the days of the B. Maysara, which led the two of them to quarrel with each other Ibn Maysara leaned towards Ibn Ya'îsh and he put Ibn Kawthar out of favour and removed him to Santarem; then he sent someone who killed him, and his place lay open to Ibn Ya'îsh and he stood alone in the leadership of the town (balad).And when Ibn Maysara died, Ibn Ya'îsh brought his (? Ibn Maysara's or his own?) sons together (or: agreed with his sons) and acquired the leadership of the city (? Ar. wa-qtata 'a al-balad ri'àsaf").And he behaved like the qàdi Abu al-Qâsim Ibn 'Abbàd in Seville and al-Bakri in the west of al-Andalus, and he defended his territory and ruled well (wa-ahsana al-siyàsa)...^^There are a number of difficulties with this passage, most notably that the qàdi Tyâd seems slightly unsure whether he is talking about Ya'îsh or about his father (Muhammad b, Ya'îsh); linguistically the question remains unclear.Fortunately, from a historical point of view things are less ambiguous, for the qadi 'lyad himself tells us (see below) that the father had died as early as 391/1000-01.More generally, the Arabic is not transparently clear at a number of points in the passage, though the overall sense of the passage can be made out.
Nonetheless, the passage seems to have led Prieto y Vives into error on the rulers of Toledo in this period.^"^This scholar appears to have understood the references to «leadership» in the passage to indicate political leadership, rulership.This seems to be mistaken.The reference to Ibn Maysara (whoever he was) indicates clearly that he was the actual ruler of Toledo at the time in question; and the context in which Ya'îsh and Ahmad b.Sa'îd Ibn Kawthar acquired their «leadership» makes this clear too: they are said to have acquired it after the deaths of their fathers, and as successors to them.Now Sa'îd, the father of Ahmad, died in «around 400»;^^ and Muhammad b.Ya'îsh, the father of our Ya'îsh, died, according to the qàdi 'lyád himself, in 391/1000-01.^^From AQ. XXI, 2000 this, and from the lengthy accounts offered here of the learning of both pairs of fathers and sons, it emerges quite clearly that the «leadership» in their city acquired by the two sons of these two men after their deaths was simply preeminence on account of religious and legal learning.
Given what we know of the learning and piety of the two men, such an action as ordering his colleague's death may well strike us as unlikely in Ya'ish.Ibn Bashkuwal tells us that when the man sent to carry out the murder of Ibn Kawthar came in «he found him reading the Qur'an; he understood that he wanted to kill him and he said: 'I know what you want; carry out your orders'; and he killed him.But it was given out publicly that he had been ill, and died.May God have mercy on him».Ibn Bashkuwal adds, depending here on Ibn Hay y an, that he died, of poison, in prison in Santarem in 403/1012-13.The story reported above could easily be a retrojection of this, though it is not absolutely clear from the way the Arabic is expressed that the story of the murder is intended by the sources to be ascribed to the year 403/1012-13, like the story of the poisoning.The story of Ibn Kawthar's reaction to the presence of his murderer also fits well with other information that Ibn Bashkuwal provides about him: apparently the family was wealthy, and, according to a report derived from 'Abd Allah b.Sa'id Ibn Abi 'Awn, some forty students, including himself, used to come to Ahmad every year, during the months of November, December and January (sic), and he would feed them all.^°On the basis of this text, we cannot be entirely certain that Ya'ïsh did in fact order his colleague's death.Regardless of any individual, personal responsibility by Ya'ïsh for this killing, moreover, involvement in engineering a man's death need not, of course, indicate share in the political power responsible for the carrying out of the murder itself.
The remarks associating Ya'ïsh with Ibn Kawthar raise another matter as well.They suggest that, if Ya'ïsh was actually ruling Toledo at the time when his colleague was murdered, then he was doing so in cooperation, or in association, with Ibn Kawthar.This would not be an impossibility: we know of the existence of colleges of rulers (though with three members, not two) both for Seville and for Cordoba at around this time.^^However, it would be extremely unlikely, and the sources do not seem to support such an interpretation.Nevertheless, there are hints, like the expressions found in the report in the qâdï 'lyád itself, that if ^^ The pairing of the first two of these is striking, given the relevance of Ibn Kawthar and our concern with pairings, but the presence here of «others» tends to lessen the possible significance of the pairing.Unfortunately, no date of death is given for this man.However, the wording in all of these cases suggests simply that a person who nazarafi l-masà'il 'ala a l-qadi was subordinate to that qadi within a professional hierarchy of legal offices; it does not necessarily tell us anything about any position or authority on the part of the person or of the qüdi outside that professional, legal, hierarchy.All of these examples, in fact, are AQ.XXI, 2000 the same in type; and it is on one of these, alone, that Prieto's case in fact rests for the suggestion that Ya'ïsh ruled in conjunction with someone else.
We have no other information, so far as I know, about the Banü Maysara.The name is rare, though not unique.^^But may it be the case that we have here another obscure reference to the Banù Matyüh?Such a view would gain some support from the fact that the place to which Ibn Kawthar is said to have been sent by «Ibn Maysara», Santarem, lies in the far south-west of the Iberian peninsula, and was at this time probably under the control of Badajoz, then ruled by Sàbùr, with whom Ibn Matyüh seems to have had, as has been seen, very close relations of mutual obligation.•' ^ If this is the case, then this text raises another difficulty, of particular relevance here, with regard to the date of the beginning of the reign of Ya'ïsh in Toledo.I have assigned this to the year 403/1012-13, on the basis of the attribution to him of responsibihty for the killing of Ibn Kawthar in that year.^^According to Ibn Bashkuwal, Ibn Kawthar administered the legal apparatus (waliya al-ahkàm) of Toledo together with Ya'ïsh, but «then [Ibn Kawthar] became a burden to [Ya'ïsh], and he arranged to have him killed (dabbara 'ala qatlihi)»?^According to the text of the qàdi 'lyâd, it was not Ya'ïsh who was directly responsible for this, but «Ibn Maysara» (although Ya'ïsh seems to have been the ultimate beneficiary of this action).If 'lyâd is right in attributing this killing to Ibn Maysara (and if Ibn Maysara is indeed to be identified with Ibn Matyüh), and not to Ya'ïsh, then it seems to be necessary to place Ya'ïsh's assumption of power in Toledo somewhat later than 403/1012-13, though it remains impossible to know how much later.Such a change in the date of the start of the reign of Ya'ïsh has (at least in theory) implications also for the dates of the reigns of all those rulers, following 'Ubayd Allah b. al-Mahdi, who preceded him in that rôle in Toledo.This is because 'Ubayd Allah is the last of these rulers, before Ya'ish himself, to whose activity we can assign a fixed date of any sort.Though we can assign something approaching a fixed date to the arrival of the B. Matyüh in the city, this is in fact no more than a terminus post quern, since they arrived there simply at some stage following their departure from Cordoba.All of these rulers, including Ya'ïsh for most of his reign, may well have ruled for longer or shorter periods scattered over the entire period from the departure of 'Ubayd Allah from Toledo up to the deposition of Ya'ïsh, or, more precisely, up to our first datable attestation of Dhü 1-Nùnid rule there.
If Ya'ïsh indeed began ruling later than the year 403/1012-13, then it may be possible to assign to this series of rulers, whom it has appeared necessary so far to fit into the very short period of the year 403/VII 1012-VII 1013, a rather longer period.However, we have no real way of knowing more exactly when Ya'ïsh took over power in the city.Things are not totally hopeless, for we can be pretty sure that the Dhü 1-Nünids were in control there by about 418/1027.We hear of them there at this time, and we hear of Ya'ïsh's own death around that date too (see below).But this does mean that the period of these rulers' activity can not, may not, be too tightly defined.
If it is difficult to know when Ya'ïsh assumed authority in Toledo, it is also difficult to know exactly when he lost it.According to another source, al-Sabtï, also quoted (ibid.)by 'lyàd, Ya'ïsh began as a good ruler, but later became corrupted by power, and as a result was eventually overthrown by the Toledans, his son, 'Abd Allah, being killed in the process, in the year A\1I\026P He is said to have died in Calatayud either in 418/1027 or in Safar 419/March 1028; the first date is given by Ibn Mutahir and the second by Ibn Hayyan.^"^Ya'ïsh is described as generally a good ruler by the qñá 'lyád:^^ ...he defended his territory and ruled well (ahsana al-siyàsa).In all this he did not call himself by the title of prince {bi-sm al-ri'àsa), contenting himself with AQ.XXI, 2000 that of «faqihyP^\ and he did not give up (wearing) the attire of the 'ulamà'; but he gave power and the title [of prince] to his son 'Abd Allah.An example of his harshness was [his] prohibition on women going outside the gate(s) of Toledo behind funeral processions in crowds...^^ This example of reluctance to take on the titles of rulership is striking, not least because it is so early in the context of Andalusi political fragmentation after the collapse of Cordobán rule.While Ibn 'Abbâd offers another example of a qàdi taking power in a political vacuum, it is not entirely clear whether in fact he refrained from adopting titles implying rulership.In Cordoba, by contrast, somewhat later, after 422/1031, we find the first of the Jahwarids behaving in very similar fashion to that described here, refraining from adopting royal titles, and making a point of continuing to live in his own house, and not moving into the royal palaces (though, given what we know of the devastation caused during the preceding twenty years of violence and plunder, and of the general poverty of Cordoba at this time, there may well have been other, more severely practical reasons for this aspect at least of this ruler's modesty and restraint).As in general in such cases, the first Jahwarid's modesty as to titulature was not maintained by his successor.This aspect of Ya'ish's rule may, however, have another relevance here.If he was indeed so modest as to the practice of rulership, and if he was really so deeply concerned about public and private morals, perhaps we should think again about the possibility of his having been involved, directly or otherwise, in the death of his close colleague (and, probably, friend) in the city in 403/1012-13.As against this, other cases show that a powerful ruler may easily content himself with such a title as faqïh; and they show similarly that, whether or not he remains extremely modest and pious in personal terms, he may have little or no difficulty in imposing his ideas as to personal modesty and pious behaviour on others with extreme harshness.
However this may be, the reconstruction which a combination of all these data makes possible suggests that we should see the royal career of this qàdi as having a ^^ This title (if it was intended as such) is striking, in the context of the modem use of the same title for a ruler in Iran.(1957), pp. 253-316, at pp. 288-90, § § 32-33 (sic).In connection with Ya'îsh's prohibiting women to follow funeral processions, it is striking, to say the least, that at around the same time (actually shghtly different structure, at least chronologically, from what has traditionally been thought: in this schema Ya'îsh will have started ruling at an unknown date after 403/1012-13.403/1012-13 is to be rejected as the date for the beginning of his reign, as resting solely on the date of his possible involvement, as ruler, in the murder of Ibn Kawthar.In consequence, we have no reason to associate his rise to power with that year at all.All we can say with any certainty is that he came to power afterwards.His rise to power may have followed the success of a plot aimed at ridding himself of a colleague and potential rival, Ahmad b.Sa'id Ibn Kawthar; such an action will have had the effect, at least, of demonstrating his capacity for rule.And his rise will also, more certainly, have followed the departure or death of the second of the B. Manyùh.As has been seen, the second of the B. Manyüh seems to have died violently and suddenly, and it may be right in such a case for us to see the accession of Ya'îsh as ruler as a response to a very difficult situation which called for immediate reaction.It could of course equally be the case, if we have a plot aimed at ridding the qcidi of his legal colleague, that we have here, in the death of the previous ruler, an echo of another such conspiracy.At all events, he will then have ruled Toledo efficiently, if with some harshness, from whenever this occurred until the people of the city revolted, possibly because of his harshness, kilhng his son 'Abd Allah, in 417/1026 and exiling him to Calatayud, where he died a year or two later.In this case, it will have been only at this stage that the Dhù 1-Nünids were summoned by the Toledans to take power in the city.
A REVISED AND EXPANDED TOLEDAN KING-LIST The list of rulers of Toledo between the fall of the caliphal-'amirid régime and the assumption of power there by Ya'îsh, to be followed in due course by the Dhü 1-Nünids, should thus read somewhat as follows (hijri and civil months are indicated by small Roman numerals, years in the normal manner): Ya'îsh, al-qàdi Abu Muhammad Ismà'îl b. 'Abd al-Rahmân Ibn Dhï 1-Nün Dhù l-Riyasatayn, al-Zàfir CONCLUSIONS From the table given above it can be seen that between 400/1009 and some time in 403/1012-13 (or possibly some unidentifiable later date) we can trace at least four and possibly as many as seven different rulers in Toledo, not to mention the eight or nine months of 402/1011 -12 (if not more) for which we have no information at all.Given the importance of Toledo, both as a frontier defence for Islamic territory in the Iberian peninsula and as a major centre of caliphal-'àmirid power before the collapse of the Cordobán régime, this may appear either as surprising or as easy to understand.It would be easy to understand as reflecting the importance which different factions in the struggle for the caliphate attached to the city; and surprising as suggesting that, despite its importance and strength, it was relatively easy to win control of it.
A glance at the identities of the various rulers suggests that the different factions involved in the political struggles of the period in fact attached little importance to Toledo: the partisans of al-Mahdî, Wàdih and 'Ubayd Allah all found aid and refuge there, suggesting that it was an area of residual Umayyad support at the end of the Mansùrid domination of the cahphal heritage, but the B.
Matyüh, who also spent some time in power there, are explicitly described in the sources as hostile to that branch of the Umayyad house which these ephemeral rulers represented.The B. Matyùh were in all probability 'Àmirid clients, and we should see their activity, in general terms, as part of the process by which so many former clients of the 'Àmirids sought to ensure their futures in the new political framework in al-Andalus after the collapse of the centralised system there.In more particular terms, too, it may be possible to see an explanation for the interest shown by the B. Matyüh in a refuge in Toledo: if their name does reflect, as was suggested earlier, a form originating in Muño, vel sim., this would indicate a Christian background for the family.In an area bordering on Christian territory it would come as little surprise to see an attempt by ex-Christians, or their descendants, to have a more significant share in local politics with the benefit of potential support from local Christians or their Muslim descendants.
No other identifiable group of participants in the struggle for power in al-Andalus at this time seems to have devoted much energy to acquiring control of the city and the huge territory which it dominated.Hostility to outside control, by whomever exercised, may have played a part in persuading the Toledans to accept any and every pretender to local authority there, but there may have been another element at work too.
In a study of the significance of Christian involvement in the affairs of al-Andalus at this time published a decade and a half ago, the late Peter Scales demonstrated that the threat presented by the Christians in the frontier areas was extremely serious.^^Between 399/1009 and 401/1011 they recovered, as a direct result of their involvement with pretenders to the caliphal throne, considerable territory and, even more importantly, a large number of strategically valuable fortresses marking the frontier between the worlds of Islamic and Christian military and political dominance in the Iberian peninsula.Many of these had been built or strengthened in the last decades of the fourth/tenth century.Their loss at the start of the fitna can, in retrospect, be said to mark the definitive beginning of the Christian recovery of territory in a visible way from an enfeebled al-Andalus.It seems likely that the Toledans, in the front line of the defence of Islam in the Iberian peninsula from the second/eighth century onwards, became aware almost immediately of the threat which this realignment of forces between MusHm and Christian in Iberia, in territory very close to themselves, portended for themselves as part of an Islamic polity in the peninsula.Regardless, therefore, of what happened in the old capital, far away to AQ. XXI, 2000 the south, and regardless, too, of the identities of the claimants to supreme authority in their Islamic polity as a whole, they entrusted authority over themselves to anyone who appeared likely to offer them a certain stability and security against further Christian encroachments.
The correctness of such a view of the security, or lack of it, of Toledo in the face of the Christian threat is confirmed for us dramatically not only by events later in the fifth/eleventh century, when Toledo was the first major city in the peninsula to be recovered for Christianity by Alphonso VI, but by the account preserved in the Primera Crónica General of that ruler's motives for attempting the reconquest of Toledo.According to that source, while dozing one day as an honoured prisoner in a palace garden in the city, he overheard two Dhu 1-Nünid courtiers discussing how easily the city could be taken.^^Whether or not this story is true is unimportant; what is significant about it is that its invention and use reflect both the reality of the situation and the awareness of this reality on the part of the Christians around the Christian monarch.
This study offers a number of benefits.On one hand it offers, via a microstudy of the sources for one very narrow and limited series of events, the potential for a surprisingly detailed account of developments in a particular place of importance during a particular period of critical significance.At the same time, and at a time when political history remains somewhat unfashionable, it is worth pointing to some benefit which a micro-study of this sort can have in other areas.This is precisely because of the nature of the sources at our disposal, as being concerned with events, on the political and military levels, rather than with processes.Such a method, by concentrating on the detail of the sources, brings out those elements which have the capacity to illuminate broader issues.Here the broader issues include the nature and extent of factional politics in the city of Toledo.Leadership there varied a great deal in this short period, and it is worth asking which element was the tail and which the dog in these cases.As has been seen, very many of these rulers came to Toledo from outside.May it be the case that they came merely as invited figureheads for local factions, at least at first, or should we rather see them as a réponse to a total lack of local leadership cadres in a city which, despite resistance to Cordobán authority, had nevertheless been ruled fairly effectively from the capital for a considerable time by the fall of the 'Amirids?Clearly a far more intensive study of Toledo, both under the rule of governors appointed from Cordoba and under Dhu 1-Nünid rule, might answer some off these questions, but success in such a study demands the discovery of new sources.The sources that we possess at present permit us to do no more than point to a number of individuals who will have been part of such factions; and we can occasionally point to the actions of a particular group or even to the local elite acting as a united front, for example at the end of the taifa period, against the ruler.But in general this represents the limit of our ability to penetrate the composition of the social and political elite of the MusHms in Toledo.
A further question relates to the B. Matyùh.As has been seen, we can extract a few details about them from our sources, though we cannot be sure, especially in the absence of any independent confirmatory testimony for the bulk of it, how far these may be trustworthy.But it looks significant in the present context that so much of our material for these rulers of Toledo actually concerns their period of activity in Cordoba, for it is that material, and its analysis, which enables us to date their arrival in Toledo.We have virtually nothing at all on them as rulers in the city itself.While this does perhaps confirm that the idea of the unity of al-Andalus was at this time, just, still somehow alive in the minds of the participants in the political struggle in the Iberian peninsula, it does little for our understanding of the political situation in Toledo itself.Our sources, generally seen as metropolitan in their focus, remain in this case also profoundly Cordobacentred in their interests and concerns, at least for this period, and such information about other cities as emerges, as in this case, is likely to be influenced by this characteristic of the texts.We should like to know much more about these B. Matyûh: did they have any connection with Toledo before the end of the caliphal-'amirid regime's existence?Who were they and what sort of background did they have?What of their careers before their adventures in Cordoba and Toledo?Did they have connections with others?And did they fit somehow into a factional system or were they merely adventurers thrown up by the extraordinary circumstances of the day?It is worth stressing how extraordinary those circumstances must have been in the eyes of contemporaries and participants.To none of these questions can we find answers in the sources.
Another set of such questions relates to the possible participation in political activity in Toledo by non-Muslims, Christians.Excluded from the political sphere in Islamic societies with great success by Muslims from the very beginning, did they succeed here in acquiring a share in the turbulent public life of this marcher town, on the frontier between Islam and Christendom?The overall history of the city in the period from 92/711 onwards might make us expect such a development.If, on the other hand, Christians, as Christians, did not succeed in acquiring such a share in pohtical life, did a different strategy produce political dividends for them?Did conversion to Islam act as a means to the maintenance and preservation AQ.XXI, 2000 of earlier political significance for local people?Can we see recent converts to Islam from Christianity and their descendants here, more than elsewhere, influencing public life in their world?Given the position of Toledo, on the border with Christendom, such a situation would be of great interest.And given the relations of Toledo over the preceding three centuries with the central government in Cordoba, it would be of still greater interest.If recent converts to Islam and their descendants did not, on the other hand, acquire influence and exercise some degree of power in this area, at times like this, then we are bound to ask different questions, related to the shape of the society of Islam in the Iberian peninsula.
Beyond all this, it is possible to point to other aspects of such a study which are less positive.First among these is the nature of our sources.In the preceding pages quite a large amount of space was devoted to 'Ubayd Allah the son of al-MahdL But it is noteworthy that a good deal of the material about him supplied by our sources is not only anecdotal but literary in character.Such information about his political career as the sources provide is incidental to the material that they contain about him as a poet.This sixteen-year-old youth seems to have had time in his short life to produce some indifferent poetry, and also apparently to inspire someone else to claim his identity a decade and more after his death, someone who also left some lines of verse.It is largely on account of those lines of verse that we know what little we do know about both of them.Without the verses, we may be fairly sure, our sources would not have devoted to either of them even the small amount of space that they do.The reason for this is partly because so much of the written sources for this period is constituted by a few anthologies (those of Ibn Bassám, al-Maqqarî, and some others like the Mughrib of Ibn Sa'id), which are explicitly literary in intention and in character; their value for us as historical sources in areas beyond the literary goes far beyond the intentions of their authors, but the works themselves and the material which they contain emerge from and largely reflect literary concerns.
A second feature, related to this limitation of our materials, which is highlighted by such a study is the narrowness of the section of the population which it is possible to examine.In one sense, our concern here was with the political class of al-Andalus, in particular with that part of it active in Toledo.What emerges from this study is that, at least so far as concerns this city in this very short period, while we can illuminate some comers of the political world of Toledo, this is so only for those at the very top and for others when they happen, for example by virtue of their identity as members of the scholarly elite, to interact with them.Fortunately, in this case, this occurs several times, but it is not always thus, and there are all too many cases, even for this period in Andalusî history, for which such help is not forthcoming from the sources.second category, n° 346b, is better represented: there is a specimen in Berlin;^^ and there is an otherwise unpublished specimen in the Musée du Bardo, Tunis, reported to Miles by M. FaiTugia de Candia.^'Prieto reports the type, claiming that it is «very rare».^^On this type we find the name Sa'id b.Yüsuf, disposed, as on the silver, above and below the obverse field inscription.
The name 'Abd Allah is found also on some specimens of the year 401/1010-11 .^^The name SaTd b.Yüsuf is found otherwise on coins of 403/1012-IB,^^ and 404/1013-14.^5And it appears on a single specimen of al-Qasim Ibn Hammüd of the year 411/1020-21.^^Because of the chronological distribution of the other specimens known bearing these two names, it seems natural to suppose that the coins of 402/1011-12 bearing the name of 'Abd Allah should belong to the first part of that year, and those bearing the name Sa'îd b.Yüsuf to the later part of the year.It seems possible, however, that this neat schema needs to be slightly disturbed, by the insertion of a single specimen between the two groups, reflecting some minting activity on the part, or in the name, of our Ibn Manyüh/Matyüh.
Miles records one coin type, taken from the catalogue of the collection of the Museo Arqueológico Nacional, in Madrid, which may belong to the period of Ibn Manyüh/Matyüh and his domination of Hisham II al-Mu'ayyad in the latter part of 402/1011-12.This is Miles' n.° 351f (a dirhem).Miles' entry here is simply a transcription of that in the Madrid catalogue.^^There the coin is described as having a slightly unusual lay-out for the (otherwise perfectly normal) obverse field inscription; its reverse field ver type attested by surviving exemplar(s).There are many other cases in Prieto's work where this, or something similar, seems to have happened.See also, for more on this, my  , 2 (1934), 299-327); 345n-hh (some score or so of specimens, together with reports in Vives and elsewhere of nearly fifty specimens whose present locations are not known; but of course there may be some considerable overlap between this last category and those whose present locations are known).
^^ Cf.Miles, Coinage, n° 348a (a dinar unique, in the collection of Vives: where is this coin now? = Prieto n.° 13c: «very rare»); 348c (3 specimens); 348d (2 specimens); 348e (over a dozen specimens, but all now not in known locations); 348f (a single specimen now in the Ashmolean Museum, Oxford).^^ Cf.Miles, Coinage, p. 543, n.° 360n (one or possibly two specimens, with the name disposed above and below the field inscription on the reverse, unlike all the other cases considered here, where it appears on the obverse).
^^ This coin is in the Academia de la Historia in Madrid (reported by Vives, n° 807 = Prieto n.° 66).
^' ^ De la Rada y Delgado, J. de Dios, Catálogo de monedas arábigas españolas que se conservan en el Museo Arqueológico Nacional, Madrid, 1892, n° 287.AQ.XXI, 2000 inscription is legible with the single exception of the personal name of the caliph mentioned on the coin.This name, Hisham, can, however, be inferred from the fact that he is identified as «al-Mu'ayyad billàh».According to Rada, there are no marginal legends on this coin («sin orlas»), and the coin looked to him «un-Spanish» («no parece español»), a fact which may help to explain why he did not actually restore the name of Hisham on the specimen.Beneath the reverse field inscription on this coin there is another unusual element: this is an inscription, read by Rada as m.f.w.y.h.Rada, followed by Miles (who did not see the coin himself), marked this reading with a query.
As with so many issues of the period of the beginning of the fifth/eleventh century, the unusual on this specimen need not brand it as not genuine.If the name has been read correctly, there seems to be good reason to see here a reference to our Ibn Manyüh/Matyüh.The slight apparent mis-spelling of the name (or, just possibly, a misreading by the cataloguer) need not occasion surprise.There would be no significant difference at all between a. fà' and a nun or ñ ta' in the script used on these coins, particularly given the small size of the writing; diacritical dots to distinguish between otherwise similar letters regularly do not appear on such coins, so that the similarity between what was read by Rada as a^' and a nun or â ta' is all the greater; and the inversion of the third and the fourth letters, if not itself simply a mis-reading, might reflect no more than the fact that the name was unusual.The placing of the name beneath the reverse, as against the obverse, however, may indicate something much more significant in the present context.A name placed beneath the obverse field on such coins generally refers to a mint-official.A name placed beneath the reverse field, by contrast, generally refers to someone completely different, and more important.During the preceding thirty five years or so, this had been a Mjib (al-Mansûr or one of his sons), a local governor (al-Mu'izz; only in north Africa), or the heir to the caliph (only in the case of issues of Sulayman al-Musta'ïn; here the influence of the pattern offered by the Mansùrids is obvious).In all these cases, the placing of the name in this position was deliberate: here it was in very close proximity to the name and titles of the caliph himself The model for this was provided by the issues of the first Mansûrid hàjib, Muhammad Ibn Abï 'Àmir.The absence of the element «ibn» from the name on this coin need not be a cause for concern in this case, any more than it is on all the specimens of al-Mansür himself There it is quite normal to find nothing more than «'Àmir» in the relevant place on the reverse.^^All this may well indicate that Ibn Manyüh/Matyüh saw himself, as others at this time saw themselves, as aspiring to fill the administrative and political vacuum created by the departure of the Mansürids.That he failed in the attempt, and disappeared very quickly, need not invalidate such an interpretation of his behaviour.While he was in power, he will have assumed, or at least hoped, that he would stay for ever.
If the suggestion made here is correct, and this coin is a solitary representative of a coinage made for or by Ibn Manyüh/Matyüh, then this is a fact of considerable interest.It does more than just enable us to re-assign a coin from one section of Miles' great work to another.The coin also attests to the importance attached even by ephemeral rulers or administrators in Cordoba at this time to the continuation of orderly government and the minting of coins, to their continuing use of standard formulae for their inscriptions, and also, perhaps, to a perceived need for coins.It also means that we can date the coin itself to somewhere in the second half of the year 402/early 1012;^^ and further, it means that we should probably date all the Sa'id b.Yûsuf issues to the period following the departure of Ibn Manyüh/Matyüh from Cordoba, in other words to the very end of the year.This is of even greater interest, for the number and variety of such remnants of the issues with the name of Sa'id b.Yûsuf seem to indicate a large amount of minting, especially for what must have been a very short period at the very end of the year.The fact of such minting with the name of Sa'îd b.Yüsuf seems in its turn also to confirm the departure of Ibn Manyüh/Matyüh from Cordoba before the end of the year.The anonymous work entitled Fath al-Andalus, published nearly a century ago by J. de González, contains a passage dealing with the early history of the taifa of Toledo which appears to mention yet another early ruler of this state.'^°The passage tells us that in the year 424/1033, the people of the city of Toledo chose Ibn Dhi 1-Nün as their ruler on the death of «Ibn Masàf».
Because of the large number of discrepancies between the statements in this work and what is known from other sources, the Fath al-Andalus has received very little attention from historians of al-Andalus in the century and more since it was first printed.^^^ I have attempted to show elsewhere that some at least of the difficulties presented by the work can be explained as resulting from a conflation at some stage of fuller accounts which contained material substantially the same as what is known to us from other sources.^^^In this case, too, I suspect that a similar process has occurred: «Ibn Masàf» is quite plausible as a corruption, or a mis-reading, of «Ibn Munaw'"» (though less so, it is true, of Ibn AQ.XXI, 2000 Manyüh/Matyüh).If it is a mis-reading of this type, then we should be faced with the statement that Ibn DM 1-Nün was the successor of Ibn Matyüh, who is also described as his sihn or relation by marriage, in this passage, and that he succeeded him in the year 424/1033.Such a statement omits the information that there was at least one other ruler, Ya'îsh, in between Ibn Matyüh and the Dhû 1-Nûnids, but this can be accounted for, as I have suggested, by the assumption of a conflation having occuixed, at some stage, of more substantial and more correct information than appears in the work in its present form.Such a statement also, it is true, appears to make the predecessor of the Dhû 1-Nûnids rule until 424/1033, which cannot be the case: I am at a loss to explain this, but especially where dates are concerned, accuracy is perhaps the last virtue to be sought in such a text as this.
There is another possible way of explaining at least some of the difficulty in this passage: after reporting that Ibn DM 1-Nün took over in Toledo, it tells us that the city kànat qablu li-sihrihi Ibn Masàf.Could it be that, confused and confusing as it is, the text actually preserves the name of Ibn Manyüh twice?Could sihrihi be, in origin, a misreading of Manyüh?A sad might easily be mis-read as a rmm, in a name which is wholly unknown, and similarly a hci ' might be mis-read as a rà '.In such a text as the Fath, this is a possibility not lightly to be dismissed.The similarities and the differences between the three versions are striking.'^-^All seem to agree on the falseness of the claims of the person who rebelled; but his real identity remains a mystery, for the expression «the slave of the eloquent druggist», which seems to be what the text in its various versions is trying to say here, tells us nothing, and cannot have been much more informative to any medieval reader.Madrid, as the location for such a rebellion, is also puzzling, since it was not a place of any importance in this period;'^"^ and the reference to al-Mustakfí, as has been noted above, is also strange, given the long period, a decade and half, between the time of al-Mahdfs and his real son's genuine activities and the reign of this caliph.But, if the name of al~Mustakfí really was in the text written by Ibn Hazm, then it is likely to have been correct, since this writer knew that caliph very well, having spent some time in gaol during his reign, after serving as a vizier in the extremely short-lived administration of his predecessor al-Mustazhir (16 Ramadan 414/2 December 1023-3 Dhü 1-Qa'da 414/17 January 1024).io5ABSTRACT The period between the fall of the Umayyads of Cordoba and the emergence of the successor states in the Iberian peninsula is shadowy and unclear.In this article, I attempt AQ.XXI, 2000 to offer a micro-study of the process in one place.Using literary and numismatic sources, I attempt a reconstruction of events in and connected with Toledo, and of the list of rulers who were active there, in the first two decades of the fifth Islamic century.This list is much longer (though almost all those mentioned in it ruled very briefly) than was previously suspected.
Because of the importance of Toledo as a frontier city, it is particularly important to know something of the process of the transfer of authority there at this time.Most of the local leaders seem not to have atributed much importance to the city; the local population, on the other hand, seems to have been willing to accept virtually any ruler who might protect them against the threat of Christian encroachment.The study shows the potential value of micro-studies in illuminating broader issues, such as factional in-fighting in such cifies, but it also brings out the metropolitan bias and other limitations of our sources.
In three appendices I look at numismatic evidence for two of these newly identified rulers, at a textual crux in the anonymous Fath al-Andalus, and at a difficult passage in the Naqî al~ 'Ariis of Ibn Hazm.
UMAYYAD TWILIGHT IN TOLEDO: AL-MAHDI, WÀDIH AND 'UBAYD ALLAH B. AL-MAHD!Direct Umayyad rule in Toledo, defined as rule by the governors appointed by and responsible to the caliphal-'àmirid régime in Cordoba, may be said to have come to an end with the revolution which brought Muhammad Ibn 'Abd al-Jabbar, al-Mahdî, to power, in 399/1009.At that time al-Nàsir 'Abd al~Rahmàn (Sanchuelo) b. al-Mansür Muhammad Ibn Abï 'Àmir, the hàjib, was killed, and the caliph, Hisham II al Mu'ayyad, was either killed or deposed.The state fell apart, with individual local governors «seizing what lay in their hands».'^

( c )
Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) http://al-qantara.revistas.csic.esAQ.XXI, 2000 THE TAIFA KINGDOM OF TOLEDO 29 , is an entry for Abu Bakr Wasîm b.Ahmad b.Muhammad b.Nàsir b.Wasîm al-Umawî al-Hantamî of Cordoba; his teachers, in Cordoba and in the east while on a pilgrimage, are listed; we hear about his activity in writing and teaching in Cordoba, up to his death there in the year 404/1013-14.Are these the same man (the absence of the name of the great-grandfather in entry n.° 1.301 [= Husaynï n.° 1.415] need not be a difficulty here)?And do they represent a son of our Ibn Wasîm?On the family see also Mann, «Familias» (see n.° 35), 269-70.3^ The B. Matyüh are mentioned also by 'Abd al-Majîd Na'na'î, al-Islàfnfï Tulaytula, Beirut (Dâr al-Nahda al-'Arabiyya), n. d., 59. ^ Cfn n° 8 supra.(c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) http://al-qantara.revistas.csic.esAQ.XXI, 2000 after the killing of Wâdih and [of] ' Alï b.Wadà'a (of which many details could be given), until Hishám's power grew weak, and Sulaymán [al-Musta'ïn] entered[Cordoba]  against him, for his second reign» ."^^ This suggests that Ibn Matyûh was the effective ruler of Cordoba at least from the death of Ibn Wadá'a (and of significance there from as early as soon after the death of al-Mahdï) until the takeover by al-Musta'ïn, in other words from approximately the beginning of 401/ca.August 1010 (the death of al-Mahdï) or from approximately Rabï' II 402/October 1011 (the death of Wâdih at the hands of Ibn Wadá'a) until Shawwâl 403/April-May 1013.We have some further information about Ibn Matyùh's activities in Cordoba and elsewhere, however, which not only confirms and amplifies what Ibn Hayyân tells us but also enables us to date his departure from the capital with greater accuracy.In the Kitàb al-Mughribfi Hula al-Maghrih of Ibn Sa'ïd there are a couple of references to 'Abd al-Rahman Ibn Matyûh."^^There, in a section entitled «Qudàt al-fitna» {«Qàdis, of [the time of] the civil strife»), we have a life of Abu Bakr Yahyâ b. 'Abd al-Rahman Ibn Wâfid.He was appointed as chief qàdî, according to Ibn Hayyân in his Kitab al-Qudàt, quoted here by Ibn Sa'ïd, by 41 Cfr.n° IS supra."^^ Ibn Sa'ïd's Kitàb al-Mughrib fi HuB al-Maghrib has been published by S. Dayf, 2 vols., Cairo, 1964^.The passages quoted here are in I, 156-57.For the history of this work see Ef, III, 926 (art.«Ibn Sa'ïd al-Maghribï», by Charles Pellat).In the first edition of Ibn Sa'ïd's text, published by Dayf in 1953, the surname of this 'Abd al-Rahman is given in two forms: at p. 156 he is called Ibn Munïr, and at p. 157 Ibn Manbûh; the relationship of each of these forms to Ibn Manyüh/Matyüh is clear, though the basis for the editor's (silent) correction of them in the second edition is not.(c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) http://al-qantara.revistas.csic.esHishámII al-Mu'ayyad at the outbreak of the BerbQvfitna: the appointment in fact occurred on 5 Jumada í 401/14 December 1010."^^In the text of Ibn Sa'id, however, there also occurs the following passage:He travelled to the East, and went on a pilgrimage, and met the learned, and established contact (wiii them) [Ar.tahakkaká^\ and among those whom he met was Muhammad b.Abï Zayd, the faqih of the Maghrib in Qayrawàn, and he stayed in contact with him (? wa-lam yazal yasil sababahu) until Ibn Abï Zayd died."^^But he was harmed during his tenure of office by love of power (? suMn) and his obstinacy in rejectkig a peace agreement with the Berbers, who had destroyed the people, and in this he differed from 'Abd al-Rahman Ibn Manyüh, the client (mawla) of Ibn Abï 'Âmir, the ruler of Hishàm's affairCs)."^Now this (? he) was the reason for his dismissal on Wednesday 9 Dhu al-Hijja 402/2 July 1012.And he stayed in his house [under some form of house arrest?] until Ibn Manyüh left Cordoba, and the ' Amirid clients (mawàïï) ran affairs; then Hishám called Ibn Wafid back on Thursday [sic] 22 Rajah 403/6 Febmary 1013 to die ^âçf-ship and die office of prayer-leader after he had made clear his dislike of tiie post and [after] Hishám applied pressure to him [to return]."^^In the next entry in this work, a life of Abu al-]VIutarrif 'Abd al-Rahman b.Ahmad b.AM al-]VIutarrif Ibn Bishr, who served as chief qàdi from 407/1016 until 419/1028, there is another reference to our 'Abd al-Rahman Ibn IVIatyüh: again quoting Ibn Hayyàn, Ibn Sa'ïd tells us that Ibn Bishr ... was given the post of qad without the prayer-leadership for the period between the two «reigns» of Ibn Wàfid... His origins were from Beja, from a noble "^^ Cfr.Lévi-Provençal, HEAf, III, 143 (n. 4 to 141), where there is a list of all those who held the office of qàdi al-qudàt in Cordoba from 291/904 up till 429/1038; al-Nubàhî, al-Marqaba al-'Ulyà, ed.E. Lévi-Provençal, Histoire des Juges d (c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) http://al-qantara.revistas.csic.esAQ.XXI, 2000

APPENDIX 2 :
A PASSAGE IN THE ANONYMOUS FATH AL-AND ALUS
^ Could this be a mistake for tahaqqaqa, or even tafaqqahal For tahaqqaqa cf.Ibn Hazm, Ya'ish was ruling at this time, then he may well not have been ruling alone throughout his reign.For example, we have a report of Hisham b.Ibrahim b.Hishám al-Tamîmî, who died as a martyr in 419/1028, who is reported to have nazarafi l-masà'il for «Muhammad b.Muhammad Ibn Mughith and Ya'îsh b.Muhammad».^-It is possible, of course, that Muhammad b.Muhammad Ibn Mughith should be regarded as one of the un-named rulers mentioned serially by Ibn TdhM, but such information about him as we have suggests otherwise: it describes only his intellectual interests, and tells us nothing about any official posts, other than membership in the shüra, that he may have held.^^More puzzlingly still, Muhammad b.Muhammad Ibn Mughith is reported here to have died in Jumada II 444/October 1052.Since his brother, Ahmad, who prayed over him at his funeral, died in Ramadan 459/July-August 1067, this date appears (c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0)http://al-qantara.revistas.csic.esreasonable:^'^ but it looks rather late in the context of his relation to Ya'îsh, particularly if Ya'îsh was given to ridding himself of unwanted colleagues in less than wholly legitimate ways/^-*" We also have another case of similar type: this time it is a biography of a man who filled the same position in the service of the father of Ya'îsh, though unfortunately we have no dates in this case at all.^^A last case brings us back to Ibn Kawthar: this is an entry for Abu al-Walîd Hishâm b. 'Umar b.Muhammad b.Asbagh al-Umawî Ibn al-Hanashî of Toledo, who similarly nàzara fi l-masà 'il, in his case 'ala «Ibn Tammâm^^ and Ibn Kawthar and others».