The Calendar of Cordoba for the year 961 by ‘Arīb ibn Sa‘d


The Arabic text of this curious calendar, whose Latin version was published in 1838 by Libri among the documents included in the first volume of his History of Mathematical Sciences in Italy1, is found in the Bibliothèque Nationale in a manuscript written in Hebrew characters whose existence was only noted in 1866 in the new catalogue of Hebrew and Samaritan manuscripts where it bears the number 10822. I deciphered and copied it for my own use but, yielding to the entreaties of friends, I am publishing it here because it helps to elucidate the Latin translation, which being literal is very often unintelligible. However the great lexicographic work which I am presently working on prevents me from addressing all the issues arising from this manuscript. I therefore refer the reader to the information given by the late Mr. Reinaud3 and will only mention briefly a matter I dealt with seven years ago in the German Asiatic Journal4.

This concerns the name of the author of the calendar, for we find that two contemporary writers contend for the honour of having composed it. They are the bishop Rabî’ ibn-Zaid and the secretary ‘Arîb ibn-Sa‘d, both of whom enjoyed the protection and favour of the Umayyad caliph al-Ḥakam II, surnamed Al-Mustanṣir billah. According to Ibn-Sa‘îd al-Maghribi5, the first wrote a book of astronomy entitled Kitāb tafṣīl al-azmān wa-maṣāliḥ al-abdān which was dedicated to al-Ḥakam II. The second, who was also known as a historian, composed a Kitāb al-anwā’ or calendar, from which Ibn al-Awwâm cites several passages in his Treatise on Agriculture.

  • The following reasons lead one to suppose that the calendar was composed by bishop Rabî’ ibn-Zaid:

    1.  The title of the book, which is given at the end of the Arabic version, is exactly the same as that of the work by Rabî’ ibn-Zaid, as attested by Ibn-Sa‘îd .

    2.   The book is dedicated, at least in the Latin version, to the caliph al-Ḥakam II.

    3.   The name of the author, in the Latin version, is Harib filius Zeid episcopus. Harib is not Rabî’, but filius Zeid episcopus is surely ibn-Zaid al-Asquf (the Bishop), as recorded by Ibn-Sa‘îd .

    4.   The author seems to have been a Christian rather than a Muslim for he mentions the feast-days of [Christian] saints but nowhere does he indicate any important day in Islamic history as would surely be noted in Muslim calendars. He also uses certain expressions, such as that found under 25th January in the Latin version, which could only come from the mouth of a Christian, and his book begins, in both versions, without the doxology that Muslims are wont to put at the head of their books.

    However the third and fourth points may be countered with the following:

    1.  The name of the author, in the Arabic version, is ‘Arîb ibn-Sa‘d6 the secretary, and in the Latin version, it is Harib, which is evidently a transcription of ‘Arîb.

    2.   The author, at least judging by the Arabic version, was a Muslim, for when he cites a passage of Qur’an (in this edition, p. 4, lines 6 and 7) he presents it as the word of God, in the Muslim manner. Furthermore, the Christian expressions, such as one finds in the Latin version under 25th January, are not in the Arabic version.

    Should we then accept that the author of the calendar is ‘Arîb ibn-Sa‘d? Before we can do so, we have to be able to answer the following questions satisfactorily:

    1.   Why is it that, in the Arabic version, the work of ‘Arîb bears the title of a book written by his contemporary Rabî’ ibn-Zaid the bishop?

    2.   Why does this ‘Arîb bear, in the Latin version, the name of filius Zeid episcopus, that is to say, the name of the other author?

    3.   The citations that Ibn al-Awwâm took from the Kitāb al-anwā’ of ‘Arîb are only preserved in part in our book, and some of them are at odds with what we read in our calendar, as I have shown elsewhere7. What accounts for these differences if our book is really the calendar of ‘Arîb?
What is certain is that in both versions the titles of the two works and the names of the two contemporaneous authors have been confounded in a very strange way.

The differences in content between the two versions also deserve mention. The Arabic text contains some things that are not in the Latin text though, in general, the latter is more extensive. Notably it mentions places in Cordoba and its environs where the saints’ feast days were celebrated, information which is never given in the other text, and, curiously, it is the names of martyrs during the times of the Umayyad princes Abdul Rahman II, Mohammed and Abdal Rahman III that are mentioned here, whereas one searches in vain for them in the other version.

From all this one is tempted to conclude that strictly speaking our calendar is neither that of Rabî’ nor that of ‘Arîb but the work of a third person who, having abridged the two calendars, merged them into one. The notable differences between the Arabic and Latin versions even lead one to think that there might have been two abridgements, or two redactions of a single abridgement. The question is very obscure and the materials we have at our disposal are not sufficient to resolve it conclusively.

The Arabic manuscript from which this edition is transcribed and which was written in the 14th century, is far from correct - one has only to cast an eye over the numerous mistakes that I have noted below the text. Several of them can only be explained by the Hebraic writing system. For example one notes the confusion between the letters ة) ه) and و ,ا and ى, which in the Arabic script are very different, and the frequent transposition of letters which would be inexplicable in Arabic, whose letters are joined to one another. The style of the book is very negligent and rules of grammar are not always respected. I have taken the liberty of correcting the most flagrant mistakes but have left other irregularities which seem to derive from the dialect of the Jewish copyist, or that of the person who abbreviated the two calendars, if we consider our book as such. I have left  شهيد  as a common noun, and plurals such as  طواوس  ,بساتن  ,اسابع  ,عقاقر,  for  اسابيع  ,عقاقير,  etc. A few vowels have been added by me, for the manuscript has none, and I have substituted Arabic numerals for the Hebrew characters indicating numbers. The days which I have left blank are marked in the manuscript with لا كلام عليه or فيه.

The calendar is not without lexicographic interest and if, contrary to my usual practice, I have not added a glossary it is because the words that need explanation will be included in my Supplement to the Arabic Dictionaries, a quarter of which has been prepared for printing.

In reprinting the Latin version I have similarly had to correct the text, mainly by comparing it with the Arabic version, but for the rest I felt I was obliged not to alter the crude Latin in which it is written and to scrupulously preserve the often bizarre orthography of the manuscript. With rare exceptions, I thought it unnecessary to point out the very numerous mistakes made by the translator; they are easily corrected by consulting the Arabic text. The words printed in italic, and enclosed in parentheses [in the Latin version], are marginal notes in the manuscript.

Finally, I would like to thank my excellent friend, M. de Goeje, who, while I was suffering from a long illness, kindly helped me by correcting the proofs as well as contributing many useful and important comments.


1 p. 392 and following
2 Old Collection 462. In the old catalogue (Catal. Codd. MSS. Bibl. Regiae, t. I, p.45), the name of the author was unrecognizable, for it reads: Abul Hacen Adib filii Sahih tractatus de calendario, ubi cursus solis aliaque ad astronomiam pertinentia explicantor.
3 French translation of the Geography of Aboulfeda, Introduction, p. xc and following.
4 Zeitschrift der deutschen morgenländischen Gesellschaft, t. XX, p. 595-609.
5 In Makkari, t. II, p. 125, Leiden edition.
6 In the German Asiatic Journal I believe I proved that these names are correct and that Gharib and Sa‘îd are erroneous. In the Arabic manuscript written in Hebrew characters one finds the second error at the beginning, and the first at the end, though at the beginning the copyist has written this name correctly.
7 In the Journ. Asiat. Allemand, p. 600-603. Adding to the list that I have given there: Ibn al-Awwâm, t. II, p. 130, last l.

(Translated from the Introduction by R.P.A Dozy to Le Calendrier de Cordoue de L’Année 961: Texte Arabe et Ancienne Traduction Latine, published by E. J. Brill, Leiden, 1873)