skip to content

Plants on Maimonides’ bookshelf: T-S Ar.41.41

by Gabriele Ferrario


Which were the favourite books of Moses Maimonides? Which titles would have found space on his bookshelf?

Maimonides’ letter to the Hebrew translator of most of his Judaeo-Arabic production, Samuel ibn Tibbon, contains revealing passages regarding the books that Maimonides considered the basis of any solid philosophical education.[1] No wonder the place of honour is occupied by the works of Aristotle, which became available to the Arabic-speaking world thanks to the spectacular effort of Arabisation of Greek sciences conducted under the Abbasid caliphs. Maimonides describes Aristotelian treatises as ‘the roots and foundations of all works on the sciences’. But Aristotle’s philosophy was not always easy to understand for a medieval reader, and Maimonides recognised the utility of later commentaries and systematisations of Aristotelian works produced by philosophers of Late Antiquity and Islam, in particular the works by Alexander of Aphrodisias (2nd–3rd c.), Themistius (d. 390 CE), and Averroes (d. 1198). As much as praising his favourite authors, Maimonides is very keen on downplaying the importance of authors he fancied less, and writes to Ibn Tibbon that reading commentaries by Abū Yaḥyā ibn al-Biṭrīq (9th century), Yaḥyā ibn ʿAdī (10th c.) and by Abū al-Faraj ibn al-Tayyib (11th c.) would be a waste of time. A similarly dismissive approach characterises Maimonides’ stance towards Plato and other Greek classical philosophers: Aristotle said it all, why should one look for anything else? Among Muslim philosophers, Maimonides praises Al-Fārābī (10th c.), particularly for his logical works, Ibn Bajja (the Latin Avempace, 11th–12th c.) and Averroes (12th c.) for his numerous Aristotelian commentaries; he also remarks that books by Avicenna (11th c.) are worth studying, even if they are not as good as Al-Fārābī’s. Maimonides’ outspoken comments on the philosophical production up to his day allows us to think the he must have read – and possibly also owned – at least a portion of the works he is commenting upon.

What could be a better source than the Genizah for finding direct information about which books found space on Maimonides’ bookshelf? Although rich in Maimonidean autographs and material related to his life and activities in Egypt, Genizah manuscripts have not yielded yet much evidence on this topic.[2]

During my work for the Wellcome Trust funded project ‘Medicine in Medieval Egypt: creating online access to the medical corpus of the Cairo Genizah’, I have come across a fragment that may help us move some steps forwards in this direction: T-S Ar.41.41.

The entry provided in the printed catalogue of the Genizah medical fragments reads: ‘Beginning of a translation of Nicolaus’ synopsis of Aristotle’s De Plantis, once belonging to Maimonides’.[3]

The fragment is made up of two separate leaves and the opening of the treatise On Plants is found on the verso of P2.


T-S Ar.41.41 P2 verso

T-S Ar.41.41.P2 (verso), Cambridge University Library

The Arabic text on this page corresponds to the very beginning of the treatise known as De Plantis (‘On Plants’) and was long transmitted as an original Aristotelian work. It opens by stating that life is found in both animals and plants but, while in animals the presence of life is apparent (they are clearly provided with motion, and they use it to satisfy their appetites), life in plants is concealed, and therefore its study requires some more investigation. The fragment proceeds to mention the opinion of two pre-Socratic philosophers, Anaxagoras (5th c. BCE) and Empedocles (5th c. BCE), both argued that plants feel desire and sensation, pleasure and pain. Anaxagoras is reportedly convinced that plants are nothing but animals and that they also feel joy and pain.

The history of the transmission of the De Plantis is complex, fascinating and possibly unique within Aristotelian works.[4] Since Aristotle himself mentions the De Plantis ten times in his own works and the title is listed in the classical repertoires, we can assume that an original Aristotelian De Plantis (Greek: Περὶ φυτῶν) must have existed at a certain point, but was probably already lost by the time of Alexander of Aphrodisia (3rd c. CE), who explicitly says that no work with that title was produced by the philosopher. The Aristotelian De Plantis disappeared from circulation probably during the 1st century CE, but not before Nicolaus Damascenus managed to produce a miscellaneous treatise on plants, conflating material of the Aristotelian De Plantis with Theophrastus’s treatment of the same topic. This re-elaboration by Nicolaus is in its turn lost in its Greek original. In the East the text was transmitted in Syriac and Arabic translations, along with genuine Aristotelian works. A unique fragment of the Syriac translation of the De Plantis is preserved in ms. Gg.2.14 at Cambridge University Library (f. 383), and other passages are quoted in Bar Haebraeus’ Candelabrum Sanctuarii and Butyrum Sapientiae.[5]



Gg 2.14, f. 383 (recto), Cambridge University Library


An early Arabic translation of the work was prepared by Isḥāq ibn Ḥunayn (9th c.), who was persuaded he was enriching the list of Aristotelian works available in Arabic, and this translation was corrected by the famous Sabian scientist Thābit ibn Qurra (9th c.). This is the version preserved in the Genizah fragment T-S Ar.41.41 and in at least five other manuscripts, which were still unknown to Steinschneider at the end of the 19th century, when he composed his Die arabischen Übersetzungen aus dem Griechischen.[6] At the time, Western scholars were not aware that an Arabic Kitāb al-Nabāt, the translation of our De Plantis, had been listed in a catalogue of the Yeni Cami Library in Istanbul dated 1309 H. = 1882/3 CE. The Yeni Cami manuscript – now in the Süleymaniye Library – was first noticed and described by Maurice Bouyges in 1923, and later edited and translated by A.J. Arberry in 1933.[7] After the Second World War, a further 3 manuscripts of the Arabic De Plantis were discovered in Tehran and another one in Tashkent.[8]

The De Plantis reached the Latin world through its illustrious pseudoepigraphic paternity, and was translated in the 13th century by the English scholar Alfred of Shareshill, who also authored original treatises and scholastic commentaries on Aristotle and Boethius.



Ii.II.10, f. 220 (recto), Cambridge University Library


This translation proved very popular up to the 16th century, and was published in two early incunabula: 1489 and 1496.[9]

The history of the transmission of the De Plantis is further enriched by its Hebrew translations. The first one was produced by Shem Tov ibn Falaquera around the middle of the 13th century and included as part of the fourth section of his Deʾot ha-filosofim. Shem Tov appears to be aware that the text he is translating is not an original composition by Aristoteles, and attributes its compilation to ‘the Alexandrinians’ who – according to his view - had prepared an abridgment of Aristotle. Another Hebrew version of the De Plantis, possibly based on a different Arabic source than the one used by Shem Tov, was finished on the 8th of Nisan 5074 (= 1314 CE) by the Provençal Jewish philosopher Qalonymos ben Qalonymos, and is now preserved in at least nine manuscripts.[10]

But the intricacies of the transmission of this work do not end here. The next phase sees the production of a Greek retro-version based on the Latin translation by Alfred of Shareshill. Following the remarkably naive idea of a restitution of the text to the purity of its Greek style, an anonymous 13th – or 14th – century scholar, probably from in Byzantium, decided to re-translate into Greek the Latin translation of the Arabic translation of the miscellaneous text that Nicolaus Damascenus produced on the basis of the soon-to-be-lost Aristotelian De Plantis: so much for purity of Greek style! It is dubious that this Greek version resembled in any way the lost Aristotelian Greek text, but the work gained some fame among humanists and it superseded the Latin translation: it is now extant in at least 18 mss, it was printed in 1536 and also forms the basis of two humanistic Latin translations, published in 1542 and 1543.[11] In sum, the complex history of transmission of this text can be visualised as follows:[12]



Stemma of the transmission of the De Plantis (reproduced from Drossaart Lulofs, H.J. and Poortman, E.L.J. (eds.), Nicolaus Damascenus De Plantis. Five Translations, (‘Aristoteles Semiticus-Latinus’), Amsterdam-Oxford-New York, North-Holland Publishing Company 1989)


Up to the middle of the 16th century, no substantial alternatives to the attribution of the De Plantis to Aristotle had been proposed. A thorough study of the work was later conducted by E.H.F. Meyer in his Nicolai Damasceni de plantis libri duo Aristoteli vulgo ascripti (‘The two books on plants by Nicolaus of Damascus generally attributed to Aristotle’), which was published in Leipzig in 1841. Meyer was able to rectify centuries of misattribution of the work and to begin the reconstruction of the history of the transmission of the De Plantis along the aforementioned lines, thanks to a bibliographic note found in Hajji Khalīfa’s Kašf al-ẓunūn (published mid-17th century), where he retrieved a reference to Nicolaus Damascenus as compiler of our botanical treatise.

This is in brief the historical background of the complex transmission and fame of the text preserved in T-S Ar.41.41, that ended up on Maimonides' bookshelf.

But how could Isaacs be sure that the manuscript of the De Plantis survived in T-S Ar.41.41 belonged to Maimonides himself? In this case, the answer is very easy: Maimonides signed the reverse of the first page of the fragment.



T-S Ar.41.41 P1 (verso), Cambridge University Library


On the recto of the same page, the Arabic text from the first section of the De Plantis is badly rubbed and only few faded words can be read. Anyway, Maimonides’ signature on the verso is very clear and can easily be identified by comparison with other autograph documents he signed. Let’s take as an example a very famous letter of recommendation penned and signed by Maimonides with his complete name.


T-S 12.192 (recto), Cambridge University Library


The letter had an official aim and so Maimonides signed it with his patronymic as well. When it comes to his first name, though, there is no doubt that the hand that wrote the letter also signed the verso of T-S Ar.41.41 with a rudimental ex libris, meaning possibly: ‘[this book belongs to] Moses’.


Cropped signatures of Maimonides from T-S 12.192 (above) and T-S Ar.41.41 (below)



One last point may now be proposed: despite his passionate praise of the fundamental role of Aristotelian treatises as the ‘the roots and foundations of all works on the sciences’ and almost the only philosophical texts that really need studying, it is very likely that Maimonides did not actually know that the copy of the Aristotelian Kitāb al-Nabāt he owned and proudly signed was not actually by Aristotle: it was the result of corrections by Thābit ibn Qurra on the Arabic translation made by Isḥāq ibn Ḥunayn of a botanical treatise composed by Nicolaus Damascenus on the basis of a lost Aristotelian De Plantis and of Theophrastus’ botanical works.


My research on the medical fragments in the Cambridge Genizah Collections is supported by the Wellcome Trust Research Resources Award, Medicine in medieval Egypt: creating online access to the medical corpus of the Cairo Genizah.




[1] The Hebrew text of this letter was published in Marx, A., ‘Text by and About Maimonides’, JQR 25 (1935): 378-80 and in Shilat, Y., Iggerot ha-Rambam, Jerusalem 1988, pp. 530-554. For a study on the influence of this letter on later Jewish philosophy, see Harvey, S., ‘Did Maimonides’ Letter to Samuel Ibn Tibbon Determine Which Philosophers Would Be Studied by Later Jewish Thinkers?’, JQR, 83, 1/2 (Jul. - Oct., 1992): 51-70.

[2] The Leverhulme-British Academy project ‘Maimonides in the Genizah’ by Dr. Amir Ashur is still making exciting discoveries of Maimonidean fragments, including autographs.

[3] Isaacs, H.D., Medical and Para-Medical Fragments in the Cambridge Genizah Collections, Cambridge, CUP 1994, p. 29 (nr. 364).

[4] For a detailed treatment of the transmission of the De Plantis, together with critical editions of its Syriac, Arabic, Hebrew, Latin and Greek versions, see Drossaart Lulofs, H.J. and Poortman, E.L.J. (eds.), Nicolaus Damascenus De Plantis. Five Translations, (‘Aristoteles Semiticus-Latinus’), Amsterdam-Oxford-New York, North-Holland Publishing Company 1989.

[5] Bakoš, J., Le candélabre des sanctuaires de Grégoire Aboulfaradj, Turnhout, Brepols 1974. Takahashi, I., Aristotelian Meteorology in Syriac: Barhebraeus, Butyrum Sapientiae, Books of Mineralogy and Meteorology, Brill, Leiden 2004.

[6] Steinschneider, M., Die arabischen Übersetzungen aus dem Griechischen, Leipzig, Harrassowitz 1893, nr. 77.

[7] Bouyges, M., ‘Sur le De plantis d’Aristote-Nicolas à propos d’un manuscrit arabe de Constantinople’, Mélanges de l'Université Saint- Joseph, Beyrouth, 9.2 (1923): 72-93; Arberry, A.J., ‘An Early Arabic Translation from the Greek’, Bullettin of the Faculty of Arts – University of Egypt, Part 1: vol. 1.1 (May 1933): 48-76; Part 2: vol. 2.1 (May 1934): 71–105. Another edition of the Arabic text is found in Badawi, ‘A., Aristatālīs fī n-nas, Cairo, Maktabat al-Naḥḍah al-Miṣrīyah 1954: 241-281.

[8] Teheran: Majlis-i Sénat 8416; Ilāhīyāt 293 ğ/32; Malek 4655. Tashkent: University Library 23/5. The critical edition in Drossaart Lulofs – Poortman, cit., takes into account all the available mss.

[9] On early editions of Aristotelian and Pseudo-Aristotelian works, see Kraye, J., ‘The printing history of Aristotle in the fifteenth century: a bibliographical approach to Renaissance philosophy’, Renaissance Studies, 9,2 (June 1995): 189–211.  

[10] Oxford: Oppenheimer Add. 4o 10; Huntington 576; Paris: BN 1341; BN 1005; Parma: 2272 (De Rossi 216); 2093 (De Rossi 776); New York: JTS 2444; Budapest: Kaufmann 285; Vatican: 290. For a complete description of the Hebrew mss. see Drossaart Lulofs – Poortman, cit.

[11] For details on the Greek retroversion and its Latin translations, see Drossaart Lulofs – Poortman, cit.

[12] I have reproduced here the stemma proposed by Drossaart Lulofs – Poortman, cit., p. XVI.


If you enjoyed this Fragment of the Month, you can find others here.

Contact us:

The zoomable images are produced using Cloud Zoom, a jQueryimage zoom plugin:
Cloud Zoom, Copyright (c) 2010, R Cecco,
Licensed under the MIT License