¹ Imprimerie Orientale Maurin, G., Paris.Google Scholar

² Catálogo de los manuscritos árabes existentes en la Biblioteca Nacional de Madrid, Madrid, 1889.Google Scholar

³ Op. cit., p. 6.Google Scholar

⁴ Sr. Manuel Sánchez Mariana, head of the section of old and rare manuscripts at the National Library in Madrid, told me that the new catalogue numbers were introduced at the end of the last century when more manuscripts came to be housed in the present library building. No words can register my deep gratitude to Sr. Sánchez Mariana and his staff for the help I was given while working at the National Library in Madrid during the months of September and December 1986 and January 1987.

⁵ Guillén Robles gives no date, but this investigation of the Escorial maunscripts took place in all likelihood about the middle of the XVIIIth century. Casiri, or, to give him his Arabic name, Mīkhaīl al-Ghazīrī, arrived in Madrid in 1748, and the first volume of his catalogue of the Arabic manuscripts at the Escorial Library (Bibliotheca Arabico-Hispana Escurialensis), appeared in 1760.

⁶ Catalogue, , p. 244.Google Scholar

⁷ It has now been given no. 1623 in Casiri's catalogue, and that is after it was decided, for the purpose of standardising references, to give the manuscripts in Casiri's catalogue the numbers given to them by Derenbourg in his Les Manuscrits arabes de l'Escurial, Paris, 1884,Google Scholar 1903, 1928 and 1941. Casiri, writing in Latin, had used Roman numerals.

⁸ Derenbourg, , op. cit., p. 41.Google Scholar

⁹ See n 19 below.

¹⁰ “Un valioso códice árabe de concilios españoles recuperado para El Escorial,” the October–December issue, vol. CLXXXIX, no. 4, pp. 681–695.Google Scholar I was guided to this important article by Dr. Braulio Justel Calabozo, Dean of the Facultad de Filosofía y Letras at the University of Cadiz, and Director of the Escorial Library from 1974 until 1982.

¹¹ In fact what Derenbourg calls “résumés in Latin and attempts at translation into Latin” is Casiri's consistent attempt throughout his copy to sum up in Latin all the chapter headings, and to indicate the names of the councils from which the particular canons proceeded.

¹² It is difficult to see what pagination Derenbourg is referring to. The original MS. is in 435 folios and is divided into ten books. Casiri's Latin translation of it is complete, and is in two volumes whose catalogue numbers at the National Library are now MS. 8985 and MS. 8986. The first volume contains the translation of the first three books, while the second contains the remaining seven.

¹³ Guillén Robles is clearly mistaken in his catálogo where he says:

Más adelante en el reinado de Carlos IV, pensóse en hacer una editión de los antiguos códices de la Iglesia de España, y entonces hicieron otra copia D. Elías Schidiac y D. Pablo Lozano(p.243).

This remark is rather baffling, because when Guillén Robles goes on to describe this other copy under MSS nos. 595–596 in his catalogue, he clearly states: “Es una copia hecha por D. Pablo Elías Hodar. … ” More baffling still is the fact that Simonet, writing in 1897 repeats that “Schidiac y Lozano sacaron otra copia del mismo códice … ”, and goes on to state in a footnote that this is “Cód. árabes núms. 595 y 596 del cat. de Guillén Robles.” (See Historia de los Mozárabes de España, Amsterdam, Oriental Press, 1967, p. 725,Google Scholar and n. 4 on the same page.) What Schidiac and Lozano in fact did was to collate the two copies of Casiri and Hodar with the original, and with other Latin collections of concilios. See Blanco, Pedro Luis, Noticia de las antiguas y genuínas colecciones canónicas inéditas de la Iglesia española, que, de ordén del Rey nuestro Señor, se publicará por su Real Biblioteca de Madrid. Madrid, Imprenta Real, 1798, p. 99.Google Scholar See also de Andrés, , op. cit., p. 685.Google Scholar A close reading of the texts might show that both Guillén Robles and Simonet seem to have drawn extensively on the work of Pedro Luis Blanco just cited, but without closely verifying what he had to say.

¹⁴ Derenbourg, , op. cit., pp.41–2.Google Scholar

¹⁵ See de Andrés, , op. cit., p. 683,Google Scholar where he says that this collection gives in their entirety canons which other collections in Latin do no more than refer to. Its title in Arabic, Jamī Nawāmīs al-Kanīsa wa-'l-Qānūn al-Muqaddas, is revealing. See also Blanco, , op. cit., p. 97,Google Scholar where he describes this collection as “completa”.

¹⁶ See Derenbourg, , op. cit., p. 41,Google Scholar and de Andrés, , op. cit., p. 682Google Scholar

¹⁷ For a detailed account of the fire of 1671 and the damage resulting from it, see Justel, Braulio, La Real Biblioteca de El Escorial y sus manuscritos árabes, Institute Hispano-árabe de Cultura, Madrid, 1978, pp. 187–191,Google Scholar and de Andrés, Gregorio, Real Biblioteca de El Escorial, Madrid, 1970, pp. 15–16 and 39.Google Scholar

¹⁸ As quoted by de Andrés, , op. cit., p. 685.Google Scholar In fact de Andrés wonders why Carlos III never came round to ordering the publication of this work, and thinks that it was perhaps the unfavourable attitude of the Royal Librarian at the time, Francisco Pérez Bayer, which prevented publication, (loc. cit.)

¹⁹ “Casiri”, says de Andrés, “trying to make known such a sensational manuscript, and hence to get it published, had obtained from Carlos III the permission to take the manuscript to his home in Madrid in 1751 where he worked over a number of years translating it into Latin” and preparing his Arabic copy of it. The manuscript remained in the National Library in Madrid after Casiri had used it, until its return to the Escorial in 1966. Ibid. p. 682.

²⁰ See de Andrés, , op. cit., p. 685.Google Scholar It should be pointed out that the Escorial manuscript, which is on parchment, has about two dozen folios at the beginning either totally blotted out or hardly legible. Damp has similarly affected various other sections of this massive codex, and made reading them a laborious task. In view of this, and in view of the excellent copies by Casiri and Hodar preserved at the National Library in Madrid, it could well be said that the Escorial's gain by the return of the original MS. in 1966 was not a measure of the National Library's loss. Casiri's copy is in the greater part of it a supreme model of Arabic calligraphy, and a delight to read.

²² This work (see full title and details in note 13 above) is now kept in the printed manuscripts section at the National Library in Madrid, under number 3905 MSS IMP.

²³ So called, according to Blanco, (p. X of the introduction), because they were composed in the reign of the Visigothic kings.

²⁴ p. 168.

²⁵ Full name, Manuel de Godoy y Álvarez de Faria (1767–1851).

²⁶ See op. cit., p. 686.Google Scholar Although Godoy's name is associated with the foundation of various cultural institutions, he is pictured by most historians as primarily intent on self-aggrandisement. H. R. Madol called him “the first dictator of our time”. It was Godoy apparently who used the title of generalísimo for the first time in Spain, and that as commander of the Spanish army in 1801. His meteoric rise to power and the favour he found at Court were apparently the result of a sensational scandal. See on this Diccionário de historia de España, edited by Bleiberg, Germán, 2nd ed., Madrid, 1968, vol. II, pp. 213–218.Google Scholar

²⁷ Ibid. p. 218.

²⁸ Among them the statesman and author Melchor Gaspar de Jovellanos (1744–1811).

²⁹ Op. cit., pp. 686–687.Google Scholar

³⁰ Colección de canones de la Iglesia Española (with translation into Spanish by D. Juan Tejada y Ramiro), Madrid, 1849,Google Scholar Introduction, pp. X–XI.

³¹ This work was first published in Madrid in 1903. But the preface points out that it was ready for publication in 1897, and that the author saw the first proofs of it before his death in that same year. In a footnote in his other well-known work, Glosario de voces ibéricas y latinas usadas entre los Mozárabes, published in 1888, n. 2, p. XIV,Google Scholar Simonet speaks of this work as “que aun yace inédita”. All references here are to the Amsterdam, Oriental Press reprint, 1967.

³² Op cit., p. 728.Google Scholar

³³ Ibid., p. 720.

³⁴ Ibid., p. 724.

³⁵ Ibid., p. 726.

³⁶ Folio 333 recto in the Escorial MS; folio 227 in Casiri's copy. The date is in fact given as 1087 of al-tarīkh al ṣufri, apparently the copyist's term for the era española which was widely used at the time. Cf. Derenbourg, , op cit., p. 43Google Scholar and Dozy, R., Supplément aux dictionnaires arabes, I, p. 836.Google Scholar

³⁷ Changing Malik to Mālik in the first hemistich, kitābun li-‘Abdi 1-Māliki l-usqufi 1-nadbi, allows the poet to avoid substituting a mafa‘īlu foot, with a final short or open syllable, for mafā‘ālun, a licence categorised as qabīḥ (jarring and crude) by all prbsddists. Casiri, perhaps unaware of the poetic necessity, corrected the name in his copy to read Abd al-Malik. The four verses in question heap encomiums on the bishop in a manner reminiscent of Arabic poetry in what has been termed by the Arabs ‘uṣūr al-inḥiṭāṭ (the period of decline). But it is the kind of fulsome praise characteristic of Arabic poetry at various stages, and which still occurs in what is being written today.

³⁸ Folios 279–280 of Casiri's copy, MS. 4877, and pp. 1225–1227 of Hodar's copy, MS. 4906, both at the National Library in Madrid, as already indicated. Casiri copied out each folio of the compact original into two folios of his own copy.

³⁹ Vincencio, Vicente and the Latin Vincentius have been suggested as the original name. At least four or five centuries before the Aljamiado texts, MS. 1623 is an excellent guide for linguists intent on discovering how the Romance letters were realised in Arabic at that early stage. Since Spanish v was invariably transcribed as in Arabic (e.g. tavernas transcribed in Ar. as and as Spanish c was realised (e.g. Barcelona transcribed in Ar. as the surmise that is the Spanish Vincencio or the Latin Vincentius is not a mere conjecture. The Spanish s is invariably realised as throughout this MS.

⁴⁰ Derenbourg, , op.cit., p. 43,Google Scholar and Blanco, , op. cit., p. 133.Google Scholar

⁴¹ This tone of philosophical desperation is not unfamiliar in Arabic ascetic or zuhd poetry.

⁴² As quoted from his ‘Idiculus luminosus’ by Pidal, Ramón Menéndez, in El idioma español en sus primeros tiempos, 9th ed., Madrid, 1979, p. 32.Google Scholar

⁴³ loc. cit.

⁴⁴ He states that he finished (copying) Book 8 on the first Sunday in Lent. As he had completed Book 7 on 17 October, 1049, this is most likely to be in Lent of the following year, 1050. See Simonet, , op cit., p. 724.Google Scholar

⁴⁵ This assertion can be made notwithstanding the fact that two of the four verses at the end of Book 8 are copied in a clumsy manner which obscures the sense and mars the metre.

⁴⁶ Simonet, , op. cit., p. 728,Google Scholar and Glosario de voces ibéricas y latinos usadas entre los Mozárabes, Madrid, 1888, Introduction, p. XIV.Google Scholar:

⁴⁷ Historia de los Mozárabes, p. 738.Google Scholar

⁴⁸ Derenbourg, ,op. cit.,p. 43,Google Scholar n.5.

⁴⁹ All folio references are to Casiri's copy, MS 4877 at the National Library in Madrid. As Casiri copied each folio of the compact original into two folios, no reference is made here to recto or verso. See n. 37 above.

⁵⁰ See Suras II, 154, V, 53, XXXIII, 33, XLVIII, 26 and XLIX, 6.:

⁵¹ It is very rare indeed that Casiri is misguided by the compact script of the original MS., the deterioration it has suffered, and the many pitfalls it presents as a result, but this term appears as in his copy, which neither in this context nor otherwise would make sense.

⁵² (Casiri's Latin translation of Codex 1623), MS. 8986 (vol. II), National Library, Madrid, folio 9.

⁵³ Maḥāarīb al-qirā is one version used suggesting the locale of reading-desks or pulpits (cf. folio 154).

⁵⁴ A distinction is made between and the latter term applying to ‘schismatics’(cf. folio 5).

⁵⁵ See, for example, the reference to Khārijiyyat Āryūsh or “The Arian heresy” (folio 279), with the final s in the name of Arius (of Alexandria realised, as usual, as sh in Arabic.

⁵⁶ One can also compare with this rūs al-ahilla (folio 280) for what would normally be ru'ūs al-ahilla. For more detail on the dropping of the glottal stop in Spanish Arabic, see Corriente, F., A Grammatical sketch of the Spanish Arabic dialect bundle, Madrid (Instituto Hispano-Árabe de Cultura), 1977, pp. 58–60.Google Scholar

⁵⁷ The reference to the zajal, apart from anything else, indicates that it was well known long before the time of Ibn Quzmān.

⁵⁸ See notes 30 and 45 above.

⁵⁹ Glosario. Introduction, p. XXXII.Google Scholar

⁶⁰ See, for example, the ode rhyming in b addressed by Al-Nābigha al-Dhubyānī to Al-Nu‘mān, King of Ḥīra. While sha‘ath has become obsolete, muhadhdhab is still the everyday Arabic term for “refined”, “well-bred”.

⁶¹ I am indebted to Miss Alison Farrow, Head of Classics at St Martha's Senior School, Barnet, for her enlightening comments on the Latin terms examined here.

⁶² Hebrew ‘atzeret. My colleague, Dr L. H. Glinert, of the School of Oriental and African Studies, tells me that certain Spanish pronunciations of the Hebrew ‘as n could account for the presence of the n in the Arabic term. See Nakhla, R., Gharā'ib al-Lugha al-‘Arabiyya, 2nd ed., Beirut, 1959, p. 212.Google Scholar

⁶³ In a muswashshaḥ by Al-A‘mā al-Tuṣīlī (d. 1126). See Gómez, E. García, Las jarchas romances de la serie árabe en su marco, Madrid, 1965.Google Scholar

⁶⁴ Historia de España y de la civilizacion española, Barcelona, 1900, vol. I, p. 343.Google Scholar

⁶⁵ Ibid., p. 341.

⁶⁶ The Legacy of Islam (ed. SirArnold, Thomas and Guillaume, A.), Oxford University Press, Reprint, 1960, p. 10.Google Scholar

⁶⁷ R. Briffault, after pointing out that “the domains of Provence remained an appanage of the Kingdom of Aragon until the annexation of Languedoc to France in 1229”. See his work, The Troubadours, translated by the author; ed. Koons, L. F., Bloomington, Indiana University Press, 1965, pp. 58–59.Google Scholar (French original, Paris, 1945).

⁶⁸ Ibid., pp. 64–67.