Part One. Philosophy. Chapter Three. Jews

Abstract

This chapter provides an English translation, revision, and bibliographical updating of the section of Die Hebraeischen Uebersetzungen that deals with the Hebrew translations of Arabic philosophical and theological works written by Jews. The principal authors considered are Abraham Ibn Daud, Baḥya Ibn Paquda, David al-Muqammaṣ, Solomon Ibn Gabirol, Isaac Israeli, Judah ha-Levi, JudahIbn Malka, Joseph Ibn Aqnin (as well as Joseph Ibn Simeon), Joseph Ibn Ẓaddik, Moses Abulafia, Moses Ibn Ezra, Moses Maimonides, Saadia b. Joseph Gaon, Pseudo-Abraham ibn Ezra, and Karaites thinkers including, Al-Qirqisānī, Joseph b. Abraham al-Baṣir, Jeshua b. Judah, and others. The Hebrew translations are described, important matters for the history of Hebrew translation are noted, and lists of manuscripts and their locations are provided.

|368| §211. [Abraham ben David ha-Levi (Ibn Daud)] In this chapter we deal with a number of treatises that cannot be called philosophical or scientific in the strict sense. Abraham ben David (Arabic: Dā’ūd, in corrupted form, Dior¹) ha-Levi from Toledo, suffered a martyr’s death (shortly after 1180?).² |369| [For his milieu and a possible reconstruction of his life, see Gerson Cohen’s introductory essay to his edition and translation of Ibn Daud’s popular Sefer ha-Qabbalah ( Book of Tradition ) Ibn Daud 1967 , esp. xvi–xlii; on the possible identification of Ibn Daud with the translator Avendauth, see M. Th. d’Alverny 1954 ; for the oppposing view, see the literature quoted in A. Eran 1998 , 301, n. 1 and 4. The misidentification of Avendauth with Johannes Hispalensis is discussed in Robinson 2003 .] Abraham wrote for a friend who had asked him about the problem of free will (1160) the book אלעקידה֞ אלרפיעה֞ (The Exalted Creed),³ one of the first attempts to achieve a compromise between Jewish belief and the Arabic peripatetic philosophy.⁴ ‹On the title cf. Bacher 1892a.› It contains, for instance, an explanation of the ten categories, which are said to be alluded to in Psalm 139 (ed. Ibn Daud 1852, 8), explicitly mentioning the book of Aristotle with the Greek name קטאגוריש as well as an explanation for the meaning of the term, doubtlessly מקולאת in the Arabic text, for which the translator coins the Hebrew term מאמרות (p. 3).⁵ In the introduction the author states that Saadia’s Book of Beliefs and Opinions is the only book that he can recommend to his coreligionists who are interested in theoretical philosophy⁶ and who require serious guidance in their attempts to verify their creed.⁷ But, he adds, Saadia’s book is not sufficient for this purpose, and Ibn Gabirol’s Fountain of Life does not specifically take the Jewish nation into account, etc.⁸ The |370| work consists of three treatises: (1) the basic principles of physics and metaphysics, (2) the principles of the Law (or religion), (3) the “medicine of the soul,”⁹ i.e., practical philosophy, which leads to felicity: ethics, economics, and politics.¹⁰

A copy of the work in the original Judaeo-Arabic was still extant between 1485 and 1520.¹¹ [No portions of the original have been discovered in modern scholarship. An attempt to reconstruct the Judaeo-Arabic terminology underlying the Hebrew translations is part of Eran 1990 .]

Jacob Guttmann 1879, 6–7, says about the work: “The often excessive brevity and conciseness of expression, the lack of animation, and the strictly logical presentation, the dry and highly terminological language make, more often than not, even for a trained scholar, difficult reading. For a person less trained in philosophy or logic, looking for light and entertaining reading, it is utterly unsuitable. The scarce attention the book receives in later Jewish literature is at least partly due to that literary peculiarity and the strictly scholarly character of the book.” [For two recent monographs on the Exalted Faith , see Fontaine 1990 and Eran 1998 .]

§212. Samuel Ibn Motot translated (in 1392) the work of Abraham for the renowned Rabbi Isaac ben Sheshet¹² under the title אמונה נישאה. Its only extant manuscript, Mantua, CI 81 , was first recognized by Luzzatto.¹³ This unedited translation is generally less reliable,¹⁴ but may serve to correct some mistakes of the edited version. Motot displays here the same elegance of expression and generally the same style as in his translation of part of Bat ̣alyawsī.¹⁵

§213. Another translation was made at the same time, perhaps some years earlier or later; its title is slightly different: האמונה הרמה.

Manuscripts: Oxford, Bodl. Ms. Mich. 57 (copied according to London, Mon. 274/2 ); Munich, BS Cod. hebr. 201/8; St. Petersburg, RNL Evr. I 470 , cf. Geiger 1837b, 447 n. 45; chs. 1–5 are lacking, according to the ms. catalogue); Turin, BN 156 A V [no longer extant], cf. Peyron 1880, 156;¹⁶ Vatican BA Cod. ebr. 259 (4 folios missing from the middle); Vatican BA Cod. ebr. 341; ¹⁷ Vienna, Pinsker 23¹⁸ (still not identified); London, Mon. 274/2 ‹(= Ghirondi n. 14 of the printed catalogue [written 1478])›.¹⁹ [Cincinnati, HUC 922 ; ‹London, BL Or. 1069 ›; Oxford, Bodl. Ms. Hunt. Don. 19/2 Moscow, RSL Günz. 270/2 ; Moscow, RSL Günz. 678/1 New York, JTS Ms. 2237 ; New York, JTS Ms. 2238/1 ; New York, JTS Ms. 2239 ; New York, JTS Ms. 2243 ; St. Petersburg, IOS B 451 ; ‹St. Petersburg, RNL Evr. I 468 ›] |371|

This translation was published on the basis of Munich, BS Cod. hebr. 201/8 alone, with German translation, as האמונה הרמה “Das Buch Emunah Ramah … verfasst im J(ahre) 4820 (תת“ך sic!)²⁰ nach E(rschaffung) d(er) W(elt) (1160) mit fortlauf(enden) hebr(äischen) Anmerkungen, und ins Deutsche übersetzt von Simson Weil,” Frankf(urt) a.M. 1852, Selbstverlag (104 Hebrew text, V and 134 German), 1852. This edition, which offers brief explanatory notes, but reveals a lack of proper information about the literature and the linguistic background, does not always exhibit an accurate text. An example of an arbitrary substitution is בשותפים in the section title on p. 81 for אלקדר) ברשות ms. f. 184b, see Munich, BS Cod. hebr. 201/8). The typographical errors are not abundant, but there remain a number of sufficiently problematic places so that inspection of the manuscripts retains some value. A lacuna (41), indicated in the margin of the ms. (f. 145), goes back to the original, cf. Vatican BA Cod. ebr. 259 .²¹ [An edition and English translation based on Oxford, Bodl. Ms. Mich. 57 , and collated primarily with Vatican BA Cod. ebr. 259 and Vatican BA Cod. ebr. 341 , with New York, JTS Ms. 2239 , was published by N. Samuelson and G. Weiss in Ibn Daud 1986 .]

None of the known manuscripts names the translator, except for London, Mon. 274/2 (and the copy, Oxford, Bodl. Ms. Mich. 57 , and Vatican BA Cod. ebr. 259 ), which begins with an unedited little poem.²² The epigram runs as follows: “I have done it (sc. the translation) from the Arabic, I, Solomon ben Lavi,²³ but it needs improvement (צריכה הגהה), for I had only one copy of the Arabic original at my disposal which was full of errors;²⁴ but in order to fulfill your wish, I am sending it to you, imperfect as it is.”²⁵ The last phrase can only be that of Solomon Ibn Lavi who without doubt lived towards the end of the fourteenth century in Ixar (Aragonia).²⁶

This translation contains some Arabic words with their explanations: סאסמ (25), ראבצ (22) which should be read רבצ (28), ךאמס (28, 35), ²⁷ טאראהגנט (sic)²⁸ לאכ (28) קורפ קאירת (49), ²⁹ הרפאט (read הרפמ 55), ןויפא (87 opium, German 110 “aphion”), תידרוי (ibid. “Viridit”);³⁰ peculiar terms are: תותואה,³¹ תוטשפתה, תויוטשפתה (dimension, 5, 10), הפיאש (27).³² Some of the material is dubious.³³

London, Mon. 274/2 and the copy, Oxford, Bodl. Ms. Mich. 57 (for part 1),³⁴ give, alongside the text |372|, an anonymous commentary which (according to Guttmann) confines itself to a paraphrase of the passages “presented” (?), but prefers “to keep silent at passages which in fact would require information.” Guttmann takes the date 1673 of the ms. straightforwardly as that of the composition.

[Krisztina Szilágyi has found among the Cairo Genizah collections a few pages from a commentary to Aristotle’s Physics by Ibn Daud, written in Judaeo-Arabic, with Ibn Daud’s name clearly displayed. Decades ago, Manuel Alonso 1943 , 186, reprinted in Sezgin 2000 , noted that Albertus Magnus refers to a work on the Physics by Ibn Daud. Thus there may have been a Latin version of the text, though it has not been located so far. There is no record of a Hebrew translation.]

Trans. Solomon Motot

Moscow, Mantua Comunita Israelitica Ms. ebr. 81 Abraham ben David Ha-Levi, fols. 1–213.

Trans. Solomon b. Lavi

Cincinnati, Hebrew Union College Ms. 922 (IMHM F 40275), fols. 1–46.

London, British Library Or. 1069 [Margoliouth 900](IMHM F 5940), fols. 1–130.

London, Montefiore 274/2 (Halb. 222) (IMHM F 5238), fols. 1b–68a.

Moscow, Russian State Library Ms. Guenzburg 270/2 (IMHM F 19031), fols. 19a–122b.

Moscow, Russian State Library Ms. Guenzburg 678/1 (IMHM F 43938, F 18562, F 18477), fols. 1a–71b.

Munich, Bayerische Staatsbibliothek Cod. hebr. 201/8 (IMHM F 1137), fols. 108b–205b.

New York, Jewish Theological Seminary of America Ms. 2237 (Halb. 452) (IMHM F 28490), fols. 1–129.

New York, Jewish Theological Seminary of America Ms. 2238/1 (JTSA Acc. 1920) (IMHM F 28491), fols. 1b–116a.

New York, Jewish Theological Seminary of America Ms. 2239 (IMHM 28492), fols. 142a–63b.

New York, Jewish Theological Seminary of America Ms. 2243 (IMHM F 28496), fols. 1–50.

Oxford, Bodleian Library Huntingdon Don. 19/2 (Uri 328) [Neubauer 1283/2](IMHM F 22097), 161a (margin).

Oxford, Bodleian Library Ms. Mich. 57 (Mich. 52.) [Neubauer 1227] (IMHM F 22041), fols. 1–223.

St. Petersburg, Russian National Library Evr. I 468 (IMHM F 52724), fols. 1–128.

St. Petersburg, Russian National Library Evr. I 470 (IMHM F 51318, CD 1018), fols. 1a–51b.

St. Petersburg, Institute of Oriental Studies of the Russian Academy B 451 (IMHM 53730), fols. 1–101.

Turin, Biblioteca Nazionale Cod. hebr. 156 A V, fols. 42–105.

Vatican, Biblioteca Apostolica 259 (IMHM F 307), fols. 1–60.

Vatican, Biblioteca Apostolica 341 (IMHM F 378), fols. 1–76.

1852. Das Buch Emunah ramah, oder: Der erhabene Glaube . Translated by S. Weil. Frankfurt a.M.

1986. The Exalted Faith . Translated by N. M. Samuelson. Rutherford.

§214. BaḤya Ibn Paquda. “BeḤai” is the usual, though erroneous, form of the name which, in the absence of anything certainly better, I have assumed. Some recent authors use “BaḤya,” a version that is less justified³⁵ than “BaḤyé”; the latter is preferred by recent Spanish authors, and for that reason, it is used by Munk³⁶ as well. BaḤiel is also used.³⁷ [Notwithstanding Steinschneider’s preference for “BaḤai,” (in his German transliteration: “Bechai”) we observe present scholarly convention and use “BaḤya.”]

BaḤya b. Joseph Ibn Bakuda or Paquda, the judge, probably lived in Saragossa³⁸ in the second half of the eleventh century, perhaps shortly after Ibn Gabirol, if one of the two knew the work of the other.³⁹ [Attempts by Yahuda and Goldziher 1913 to place BaḤya in the early twelfth century, because of inter alia an alleged dependence on al-Ghazālī, were conclusively refuted by Kokowzow 1927 and Baneth 1938 .] BaḤya composed a work, almost unique of its kind, representing a complete theory of Jewish ethics, showing a certain inclination towards asceticism, and introducing it with a philosophical proof of monotheism. However, this introduction evinces a bias against that type of philosophy which sees speculation as humanity’s highest calling. Nevertheless, it also takes a stand against a strict and formal observance of the law without regard to intention. According to BaḤya, the latter should act as a motive for the former. BaḤya therefore called his work אלהדאיה אלי פראיצ' אלקלוב ואלתנביה עלי לואזם אלצ'מאיٔר (“Instruction for the duties of the heart and admonition for the obligations of the soul, or, for the obligatory intentions”).⁴⁰ This complete title is found in Paris, BN Ms. héb. 756 , which is one more reason to presume this manuscript to be the first redaction, quite different from the Hebrew translation. Oxford, Bodl. Ms. Poc. 96 , written in 1191 by Abraham bar Tahor,^40b has only the first part of the title. Probably the book was called simply “אלהדאיה”⁴¹ and only later the Book of the Duties of the Heart.⁴² [A full discussion of the Paris and Oxford manuscripts, as well as many (but not all) of the Judaeo-Arabic manuscripts stored in the library at St. Petersburg, can be found in the introduction to the critical edition of the Judaeo-Arabic version by A. S.Yahuda in BaḤya 1912, 1–18; see also Yahuda 1904. In conformance with the scholarly convention of his day, Yahuda transcribed the text (but not the biblical quotations) into the Arabic alphabet; however, all known copies of the original are in Hebrew letters. The recent edition of Rabbi Yosef Qafih in BaḤya 1972 is based on the Paris and the Oxford manuscripts as well as Rabbi Qafih’s own manuscript.]

One Hebrew manuscript⁴³ as well as the title of the edition of 1548 indicate that the author of the Arabic book is unknown and that BaḤ̣̣̣ya (b. Asher, in the 1548 edition) is one of the two translators (see §215). This led D’Herbelot to confuse it with [the Sufi work of Abū Ṭālib al-Makkī], Qūt al-Qulūb but this is unfounded.⁴⁴ St. Petersburg, RNL Evr. I 742 ⁴⁵ contains an Arabic compendium of the Karaite physician Daniel b. Moses ([ms. dated] 1681, 22 Tevet 442). ‹For BaḤ̣̣̣ya and al-Ghāzālī, see Bacher 1892b, 56, n. 45; Steinschneider 1862b, 51, 1879f, 72.›

[Daniel ben Moses Fayruz, who also bears the name Yerushalmi, is the author of a treatise entitled כתאב אלמרשד ופראיץ' אלקלוב ומעארף אלנפוס והדאיה אלעארפין. The manuscript noted by Steinschneider is in Fayruz’s own hand. Another copy is found in the same library, listed as St. Petersburg, RNL Evr. I 1684 . Fayruz’s treatise is divided into ten books ( abw $\mathop .\limits^{\bar{{a}}}$ b ) bearing the same titles as those of BaḤ̣̣̣ya’s Duties . However, the subdivision of each book into chapters ( fuṣūl ) differs significantly. The texts must be compared closely. Fayruz appears to depend very heavily upon BaḤ̣̣̣ya, but he has not simply copied anything word for word. Nowhere does Fayruz mention BaḤ̣̣̣ya by name. In the proemium (f. 2a in St. Petersburg, RNL Evr. I 742 ), he states that he has compiled his treatise מן כלאם אלעלמא אלאפאצ'ל החכמים המשכילים הנבונים והטהורים. The passages from the book open with קאל אלמולף, but, according to Arabic usage, that could be Fayruz speaking about himself in the third person. However, the two devotional poems placed at the end of the book are also attributed to אלמולף, and here it seems clear that Fayruz is referring to BaḤ̣̣̣ya, especially since he has included a poem of his own, with his name indicated in the acrostic, at the beginning of the treatise. The first of these poems, נפשי עז תדרכי, is found at the end of BaḤ̣̣̣ya’s Duties . The second, however, ארוממך ה' אלהי ואגדלה שמך, is not. It is relatively rare, but it does display the name BaḤ̣̣̣ya in the acrostic; see I. Davidson 1924 , no. 7524 . Fayruz’s treatise poses some challenging questions concerning late medieval notions of authorship, as well as the place of Sufi-type piety within the Karaite communities of the late seventeenth century.] |373|

§215. The first translation of The Duties of the Heart is in fact one of the oldest translations from the Arabic because the first part may have been translated already in 1161, one year after Ibn Ezra had translated an astronomical treatise (see §357).⁴⁶ The translator Judah b. Saul Ibn Tibbon (or Tabbon?)⁴⁷ of Granada (מרימון ספרד) had perhaps left his native country in 1150, a year that saw catastrophic persecutions. Around 1160 he is mentioned by Benjamin of Tudela as a physician in Lunel where the renowned rabbi Meshullam b. Jacob and his equally renowned son Asher had him translate, first of all, the first part of our book and A Vintage of Pearls (§221). The translation of our book represents, in a sense, an epochal event in the history of Hebrew literature. From now on, the history of the translations can be traced along more or less certain dates in the century following this translation and even within the Tibbon family itself. In contrast, earlier translations show isolated or less coherent, uncertain, and obscure features. Judah, whom his son (in his prologue to the translation of Maimonides’ Guide) calls “father of the translators,” while later authors call him “head of the translators,”⁴⁸ deserves some more of our attention, not to give information about his not very remarkable life,⁴⁹ but rather to analyze the prologue to his first work. This was left untranslated in the most recent German translation (1854), although it came to serve as the model for all translators’ introductions. Of course we shall, in this brief analysis, not stray from our particular subject.

In a short survey of the vicissitudes of the Jewish literature, Judah stresses the fact that the heads of the Oriental academies (the Geonim) and their contemporaries living under Arab rule made wide use of the Arabic language. Its terminology, in contrast to Hebrew, is rich, and moreover, Arabic is familiar to the general reading public, which does not know Hebrew. In Christian countries scholars have restricted themselves to the study of the Bible and the Talmud because other sciences do not exist there. Meshullam b. Jacob, however, has combined the study of the Law with that of the sciences.⁵⁰ He collected, copied [or: translated] (or had copied [or: had translated] והעתיק) works on the sciences of the Law, language, and belief (האמונה), etc. Since he had heard that BaḤ̣̣̣ya had written a work on the theory of the duties of the heart (תורת חובות הלבבות – so reads the complete title in the author’s foreword; the editions leave out the first word), founded on monotheism, he ordered Judah to translate its first book. Judah had previously been asked to translate some |374| works of the Geonim, but he had not let himself be persuaded to do so in view of the demands of an undertaking of this sort, some of which he explains. Not one of the books translated from Arabic into Hebrew has escaped injury at the hands of the translators, he says, and for three reasons:⁵¹ either the translators do not know Arabic thoroughly,⁵² or they have not mastered Hebrew, or they do not understand the author. Their translations thus reflect their opinion, and for two reasons: they are not familiar with the particular discipline, and they did not read⁵³ the book under the guidance of its author or someone who had read it with him. Thus they arrived at an understanding different from that of the author or even found inconsistencies resulting from not reading a work according to its proper arrangement. This, Judah says, has been a pitfall even of great scholars. Finally, Arabic cannot be rendered into Hebrew succinctly, because – as he maintains – Arabic is a comprehensive and clear (צחה) language. In the course of these arguments he remarks, among other things, that the translator becomes the father and author of the translated work. He should strive to translate literally, without adding or omitting anything, avoiding, however, formulations that are difficult to understand, etc. This is in fact a piece of instruction to any translator, informed by common sense and linguistic sensitivity. – One of the examples he gives is the translation of Onkelos. The Bible, Mishna and Torah, however, were commented upon, translated into other languages, and given different interpretations. But since we possess both the original texts and the commentaries and translations, the latter have proven to be useful, in their own right and without detriment. The translator assumes responsibility for the author. BaḤya b. Joseph hesitated to compose his book because he felt insufficiently equipped for writing in Arabic.⁵⁴ How much more restraint, then, should the translator show when approaching his task! Judah goes on to say that most of his contemporaries tend to criticize and reproach (להתעולל ולהתגולל) anybody who creates something new, be it a translation or an original work, a liturgical poem (פיוט) or any other product of a man of reason. He talks about all this in order to let his readers know that he has taken upon himself this translation – and any possible criticism of it – only in obeisance to the order he had received. He endeavored, he says, not to distort the words of the author, translating literally even phrases with which he felt uncomfortable. When he was unable to translate, he pondered over the proper understanding of the passage, then translated to the best of his ability. When he had doubts, he consulted other works of this particular discipline. In this context, Judah regrets the absence of a (special tool) for the translated work, namely a glossary of |375| terms of the external sciences (חכמות חיצוניות)⁵⁵ according to the usage of the scholars involved. In the end Judah asks the reader’s forbearance with the novel constructions (בנינים) from verbs and nouns that he has coined, due to the limitations of the Hebrew language. Earlier colleagues had done much the same,⁵⁶ following the Arabic. Finally, he apologizes for introducing rabbinical formulations (לשון רבותינו) into biblical Hebrew, making use of the former even where he might have found an appropriate biblical formulation.

The translation of the second book of this treatise also has a short prologue. After Judah had translated the first book, Joseph KimḤi translated the remaining books; he then translated the first one as well. Later on, Abraham b. David⁵⁷ asked Judah to translate the other books too. Judah hesitated to do so for personal reasons, namely, a delicate deference to his rival. However, he later disregarded these, because his earlier scruples were no longer valid after he had translated the first book. – We have thus three different versions of the translated book: one contains the translation of the first book by Ibn Tibbon and the rest by KimḤ̣̣̣̣i, and two other versions, in which the book in its entirety is the work of one of the translators alone. Judah asks the copyists to indicate, in the beginning, the name of the translator, so that one translator not be held responsible for the mistakes of the other.⁵⁸

§216. Before continuing our discussion of Ibn Tibbon’s translation we would like to supply the meager extant information that we have about the translation of his rival and to which we shall return, under Ibn Gabirol (§221). This translation was less successful and was forgotten later on. We would know only the little that Judah relays had not Zunz (1838, 318) by chance recognized a fragment of the seventh treatise in Leipzig, UBL B. H. 39 . This was edited, with comparative notes, by Jellinek in his edition of BaḤya 1846, xiv–xxvi.⁵⁹ [KimḤi’s name does not appear to be on the fragment in the Leipzig ms., and so the identification is far from certain. This is also true for New York, JTS Ms. 1912 , which includes an excerpt from the first treatise and which Alexander Marx identified as belonging to KimḤi.] The fragment, without the notes, is also edited in the supplement to the edition of 1871 |376| [and in Tsifroni (BaḤya 1948 ), 627–37.] Some variant readings in the margin of Tibbon’s translation in the earlier editions go back, according to Jellinek, apparently to KimḤi’s translation. [Some of these go back, according to Yahuda in his edition, to variants in the Arabic.]

Judah informs us about the character of his own work. Fully aware of the difficulty of his task, he studied the relevant disciplines by himself, then strove to render the meaning faithfully by (morphological) imitation of a given term, even though this meant subjugating the special character of the Hebrew language to that of the Arabic. Arabic words which he retains as such are only few, and they are accompanied by a translation, viz. יהלאלא י'צאירלא, יעיבטלא םלעלא, (read: 2, ed. 1846a; f. 4, ed. 1854c), אלגדל (read: 28, 1846a, resp; f. 13, 1854c). We do not know how much of the technical terminology (especially that of the first treatise), which is otherwise unknown before the twelfth century, goes back to his predecessors. The same applies to the Arabicizing syntax which, however, is not as stilted and obvious as in the school which follows his procedure. There, as in all imitations, it borders upon caricature. [For more on the translation technique of Judah Ibn Tibbon, see M. Sister 1937 .]

KimḤ̣̣̣̣i is a professional philologist and interpreter, possessing a sense of the subtlety of the Sacred Word. He does no more than to render the meaning of his Arabic author in elegant, chosen words, easy to comprehend. Perhaps it was the serious scholarly milieu, formed by Samuel Ibn Tibbon’s translation of Maimonides’ Guide and gradually having grown accustomed to the new arabicized Hebrew, which gave preference to Judah’s translation, while the edifying and ascetic character of the work secured for it a cherished place in the general public. The Book of the Duties of the Heart became a work of devotion; it was copied and edited, printed repeatedly, commented upon and translated. Contemporary scholarship has devoted philological and historical studies to it. The next paragraph lists the most important of these. I shall mention here a compendium (קיצור), probably by a grandson of the same Meshullam who commissioned the translation. This compendium, in which the philosophical part is almost entirely left out, has been ascribed so far to its second editor Jacob da Fano (1614). The first edition remains unknown until today.⁶⁰

[The ascription by Benjacob to “Jacob da Fano” is an error, perhaps because the editor of the 1614 Prague edition, which differs from the 1520 Constantinople edition, was Jacob b. Abraham Katz, confused with Jacob b. Joab Elijah da Fano. Several abridged versions of the Hebrew translation are known. Steinschneider refers here to the one that was by far the most popular, judging from the large number of manuscript copies and printings. It has recently been studied by I. Ta-Shema 1982 , who identified the author as Asher ben Shelamiah, an important Provençal talmudist whose mother was the daughter of the same Meshullam who commissioned Ibn Tibbon’s translation. Ta-Shema (14 n. 5) lists nineteen manuscripts and mentions the abridgements by MenaḤ̣̣̣em ben Aaron Ibn ZeraḤ̣̣̣ (Spain, 14th? cent.), and by Jacob ẓ ahalon, the latter called, מרגוליות טובות; to these we may add an abridgement by Immanuel ben Joshua Serero of Fez, bearing the title פוקח עורים (New York, JTS Ms. 2290 ); portions of the Duties of the Heart were incorporated by Aaron ben Gabriel of Trebic in his lengthy ethical compilation, in Hebrew and Yiddish, צמח חי, written in 1756 (Moscow, RSL Günz. 545 ).]

§217. The Book of the Duties of the Heart in Ibn Tibbon’s translation was first printed, without title-page, at Naples⁶¹ in 1489. Other editions worthy of mention are ed. Isak Benjacob with an introduction by Ad. Jellinek (Leipzig 1846); ed. R. Fürstenthal, with commentary and German translation (Breslau 1835; ed. Em. (Mendel) Baumgarten (German translation) and Abraham Geiger (introduction: “The Ethical Foundation of the Book of the Duties of the Heart”) (Vienna 1854); a second edition with a different German translation (paraphrase) (Vienna 1856); and David Slutzki (Warsaw 1870), whose installments 7 and 8 contain |377| a collection of philosophical writings, under the title of חכמת ישראל, with an appendix containing the [extant] fragment of KimḤ̣̣̣̣i’s translation as well as some corrections on the basis of the Arabic text of B. Goldberg. These, however, should not be accepted without exacting control.⁶² Generally, the editions have not made use of the manuscripts, which are numerous. We mention in particular Paris, BN Ms. héb. 671/1 , which, according to the catalogue, offers a great number of variant readings.⁶³ [Steinschneider’s emphasis on the importance of this Paris ms. is based upon the enthusiastic description in Zotenberg’s catalogue 1866 , 105. However, Zotenberg remarks only that this manuscript exhibits many variants from the editio princeps (Naples 1489). In the absence of a critical edition, it is impossible to state how significant these may be. Zotenberg further notes that this manuscript has invocations in rhymed prose not found in any manuscript or printed edition; however, it is unlikely that he actually consulted all or even the majority of the numerous manuscripts.]

In the interest of textual criticism people have begun to consult the Arabic text. In addition to B. Goldberg’s publication and some other references, published here and there,⁶⁴ Jehiel Judah b. Joseph Moses Levenson has begun to compare the Arabic original with all editions in a Hebrew booklet calledcתובבל ייח.^64b ‹Cf. Levinsohn 1885, not available to me.› The author does not indicate the Arabic ms. used by him, but he mentions in his foreword a part of the book (Arabic or Hebrew?) in the library of Cairo. We call attention to the fact that the Paris ms. presents an [Arabic] recension different from the one translated by Judah Ibn Tibbon, and there is no justification for always preferring it.^64c [The mystery concerning the manuscript utilized by Jehiel Judah Levinsohn can be cleared up, thanks to New York, JTSA Ms. 2240, which contains two separate items bound together: a copy of the Venice 1506 edition, with Hebrew and Judaeo-Arabic marginalia in Levinsohn’s hand, and a large fragment of books two and three in Judaeo-Arabic, again with marginalia of Levinsohn. On f. 10 of the Judaeo-Arabic manuscript, Levinsohn writes that he received the Judaeo-Arabic manuscript in Alexandria as a gift from Faraj Ḥayyim MizraḤ̣̣̣i, who had obtained it from the Genizah: במתנה לקחתיו מכבוד הרב כ“ש מה“ו פרג חיים מזרחי הי“ו באלכסנדריא אשר נמצא בידו ממקום הגניזה בעיר (?) מצרים]

Ibn Tibbon’s text served as the basis for more recent translations into various languages, viz., into Portuguese 1670, Ladino (not Latin) 1712a, Spanish 1610, in Hebrew letters 1713, Italian 1847, Judaeo-German 1716. Already in 1765 a German translation was begun. This is the first attempt by a Jew to make a Jewish text available in German, if we do not take into account paraphrases of a German narrative in Hebrew letters.⁶⁵ Jo. Ge. Chr. Adler seems to have begun a German translation,⁶⁶ and Jakob Roman, in 1643, in collaboration with the Flandrian physician Leon Sia, wanted to publish the Arabic text in Arabic letters with a Latin translation, based upon the Arabic text.⁶⁷ [A nineteenth century translation of part of the book into Dutch is found in New York, JTS Ms. 3702 . The original Arabic text has been translated into French (BaḤya 1925 ), Spanish (BaḤya 1994 ), and English (BaḤya 1973 ); other translations are generally based on the Hebrew versions.]

From among the Hebrew commentaries two were especially popular: the one by ManoaḤ̣̣̣ Haendel 1596, a scholar versed in a great number of various disciplines, and the other, read until our day, by Israel Zamosch (BaḤ̣̣̣ya 1809.)

BaḤ̣̣̣ya’s work was also discussed in the introductions to some of the editions mentioned here, as well as in histories of literature. David Kaufmann has prepared a monograph, accepted by the Viennese Academy,⁶⁸ in which he |378| presents the basic ideas of the work and looks into some problems of textual sources. This is not the place to enter into details.

For an annotated list of the Hebrew versions of Ḥovot ha-Levavot and their translations through 1950, see A. M. Habermann 1951 . Habermann comments on the Hebrew (and the emendations made on the basis of the Arabic version) based on the 1550 Constantinople edition, which he feels to be of great significance. The principal editions and scholarly translations of BaḤya’s work since Steinschneider’s day are the following:

• 1912. Edited by A. S. Yahuda. Leiden.

• 1948. Edited by A. Tsifroni. Tel-Aviv.

• 1950. Les Devoirs du coeur . Translated by André Chouraqui. Paris.

• 1972. Edited by Yosef Qafih. Jerusalem. (Judaeo-Arabic with Hebrew translation).

• 1973. The Book of Direction to the Duties of the Heart . Translated by M. Mansoor.

§218. David (Dawud) Ibn Marwan b. Marwan אלרקי (from Raqqa?),⁶⁹ the Babylonian, called אלמקמץ or אלמקמאץ (usually al-”Mukammaz”, according to Fleischer “Mikma‘s”), is probably the earliest Jewish theologian to write a philosophical work after the manner of the mutakallimūn. Not much is known about this person.⁷⁰ An author of the eleventh century says that Saadia Gaon (died 941) did know something of David (personally?), but the matter is, according to our source, not proven or established (ברור).⁷¹ He is praised for his knowledge of non-Jewish literature, especially of medicine. Perhaps he is to be identified with the Jewish Mutakallim Abū‘l-Khayr Dā’ūd b. Mushāj (corrupted from Marwān, or MuqammaṢ?) whom the Christian Ibn Zur‛a (died 1008) mentions, or David al-Karamsi (אלקרמסי, died 945/6) mentioned by al-Mas‛ūdī.⁷² The Karaites considered him as belonging to their sect, because the two Josephs, in particular Hadassi, quote him.^72b

[There is now a book-length study of al-MuqammiṢ, including an edition of the surviving portions of the Arabic text of the Twenty Chapters , with English translation and analysis by Sarah Stroumsa 1989 . Stroumsa argues that there is no textual evidence linking him to the Karaites. We shall take note here only of those publications pertaining to the Hebrew translation that are not included in Stroumsa’s study; for fragments of the Hebrew translations Stroumsa relies upon Halberstam’s edition of Judah ben Barzillai’s commentary on Sefer Yeẓira , Judah ben Barzillai 1885 , on material brought by I. Ginzburg 1930 and by G. Vajda 1956 , 311.]

David composed an Arabic work, cited according to the title עשרון מקאלאת (Twenty Chapters),^72c in which he proves the unity of God and defends the Jewish law against its adversaries. [Stroumsa 1989 , 22 and esp. n. 56, establishes that the title cited is actually ‘Ishrūn Maqāla and not ‘Ishrūn Maqālāt ]. He also supplies information about some sects and schools. Only small portions of the work were translated into Hebrew. Judah ben Barzillai al-Barẓeloni (around 1130, probably in Provence) included some snippets in his commentary to the Sefer Yeẓira (Judah ben Barzillai 1885) “as they are translated from Arabic.” [Stroumsa 1989 , 39, observes that some of the passages cited in translation are not found in the extant portions of the original.] Halberstam’s edition of this extremely important text was based upon a transcription of the only known manuscript, whose present status, or whereabouts if it exists, remains unknown. Another short passage belongs to an unidentified treatise. [Steinschneider must be referring here to the passage cited by Judah ben Barzillai 1885 , 66; unlike the other citations brought by the latter, the name of the book is not specified here. For a list of other writings attributed to this author, see Stroumsa 1989 , 20–23.] This citation along with the ninth treatise [of al-MuqammiṢ’ compilation] was published by S. D. Luzzatto 1846, 72–78. Half of the tenth treatise was added to this by Fürst 1847.⁷³ A fragment of the sixteenth chapter was unknown (151).

These fragments are important for us because they were translated no later than the beginning of the twelfth century. Also their subject matters are of interest, for example, the mention of Aristotle, “the philosopher” in Fürst 1847 (632, 633), the Dualists (םינשה ילעב, לעב 632, 633), the Christians (622, 632–43), the classification of the sciences, beginning with |379| theology (620).⁷⁴ We name a few particular terms, for instance, the nomina departiculativa: הימכ, תוכיא, תוינא (629, 643, 643), later יצורח רבד and its opposite ינחור (631), תונרפכ, םינרפכ (622, 633); the Arabic word הנדסה (explained on 629); for חידות מכחישותz(632, 647), or מכחשות (Judah ben Barzillai 1885, 80) read תודמ?74b רוגמה םלוע (Judah ben Barzillai 1885, 151).

§219. Solomon ben Judah Ibn Gabirol (Gabriel), Arabic: Abū Ayyūb⁷⁵ Sulaymān Ibn YaḤ̣̣̣yā, Ibn Jabīrūl; Latin: Avi-Gebrol, Avicebrol, Avicebron, Albenzubrun, Jubeyzol⁷⁶ from Córdova (living around 1040–50 in Malaga), is the most original philosophical author of both Arabic and Jewish literatures, although he is better known as a Hebrew poet.⁷⁷ He is the author of the following:

1. Fons vitae, a philosophical work that develops Plotinian theory⁷⁸ – he probably knew the work of Plotinus through the pseudo-Aristotelian Theology of Aristotle (§128) – towards an original system in which the human will has almost the same function as in Schopenhauer’s philosophy.⁷⁹ It should be stressed that Ibn Gabirol does not refer to an explicitly Jewish tradition and its sources, an omission that earns for him the reproach of Abraham ben David (§211).⁸⁰ This was no doubt the main reason why the Jews neglected the philosophy of their first poet and that Christians authors who did not suspect a Jew under the garbled form of his name were attracted to him, right down to our own day, when Munk “has rendered the history of the human mind an excellent service by showing that this Avicebron who plays such a great |380| role in medieval Christian philosophy was none else but the Jew Solomon Ibn Gabirol from Malaga.”⁸¹ As a matter of fact, we owe to Munk almost everything that we know of this work, and it will suffice to refer for details to Munk’s Mélanges de philosophie juive et arabe 1859. [See now Roberto Gatti’s edition 2001, which includes a new edition of Shem Tov b. Joseph Falaquera’s extracts, concerning which see below. For an English translation of Falaquera extracts, see Manekin 2007.]

According to Munk (152), the Arabic original is not extant; its title was probably ינבוע אלחיאה. [Not a few quotations from the original (though not always, as it appears, exactly word for word) have been recovered from Moses Ibn Ezra’s Kitāb al-Ḥadīqa ; see Pines 1957/8 , 218–33; Fenton 1976 .] However, some extant manuscripts have a Latin translation from the Arabic by Johannes Hispalensis with the help of Dominicus Gundisalvus⁸²; two Paris manuscripts, (Bibliothèque Nationale, ancien fonds 6552 and Mazarin 510) were discovered by Munk and Dr. Seyerlein (of Ulm), who gives a report in an article in the Theologische Jahrbücher, ed. by Bauer and Zeller, vols. 15 and 16; a third ms. is in the Colombina in Sevilla, and I found a fourth in Erfurt.⁸³ [These four manuscripts were edited by Clemens Bäumker in his critical edition of the Latin text 1892. Schlanger 1970, 16, cites two others: Vatican, Biblioteca Apostolica, Urb. Lat. 1427, and Madrid, Biblioteca Nacional, oriental ms. 95–21, as well as two Latin epitomes (which are very close), the first of which was edited by Bäumker as the Epitome Campiliensis (on which see Sturlese 1979 .) Translations of the Latin have been made into English, French, German, Italian, and Hebrew.]

Shem Tov b. Joseph Falaquera in his Moreh ha-Moreh displays 15 quotations under the name of Ibn Gabirol. Only in one of them (1837, 96) does he supply as well the title of the book מקור חיים.⁸⁴ [Shiffman’s conjecture in 2001, 259, was anticipated by Steinschneider.] Nowhere, however, does he refer to a translation proper.

Munk detected in Paris, BN héb ms. 700 extracts conforming to the citations in that commentary, probably translated by Falaquera later on, which he, under the title of לקוטים מן ס' מקור חיים, edited along with a French translation, an introduction, etc., in his Mélanges 1859. [Another copy has since been identified in Parma, Biblioteca Palatina 2626 (De Rossi 1283), ff. 90a–107a, and this was utilized by Gatti in his critical edition 2001.] In his very brief preface the translator says that, in his study of the text, he has discovered that the author adopts the ideas of ancient thinkers⁸⁵ as they are found in the book of Empedocles on the “five substances” (or rather “the fifth substance,” according to Munk).⁸⁶ [This is a reference to the pseudo-Empedoclean work that is no longer extant in Arabic; a medieval Hebrew translation of excerpts was published in Kaufmann 1899, 18–51. For a different hypothesis see de Smet 1998. Its relationship to Fons Vitae is examined in Schlanger 1968.]

We have already had occasion (§2 and §12) to deal with the terminology of Falaquera. We shall limit ourselves here to a few expressions. He often uses פלוש and derivations for “penetrate,”⁸⁷ הגון for ראוי; he forms למות instead of למהות, according to Arabic למיה (par. 17, f. 28b of Munk’s edition);⁸⁸ נאמרות (f. 10b, par. 7; cf. his commentary on the Guide, f. 66) for “Categories”; ההזון (f. 14a, par. 27 from הזנה!);הרבנות כלומר ההאלוהות (f. 14b, par. 30, see Munk 1859, 55); …ההשקפה ב (f. 26b, par. 20), מפורקים מן “abstract,” “abstracted” (Arabic:אמרה,(מתפרק (f. 33 par. 56 and 35b par. 71) מונח בין ידיך (f. 36a, line 6 from the bottom). The Arabic words דהר (f. 10b and f. 13),⁸⁹ אניה and גנסות (f. 25, par. 12) occur for מג'אנסה֞ (Munk 1859, 95, cf. 69); the Greek genos has found its way into Aramaic.⁹⁰ [Schlanger 1968 is a comprehensive monograph on Fons Vitae, its doctrines and its sources; it was translated into Hebrew 1980.]

Principal Editions and Translations

1892. Avencebrolis (Ibn Gebirol) Fons vitae ex arabico in latinum translatus ab Iohanne Hispano et Dominico Gundissalino ex codicibus Parisinis, Amploniano, Columbino. Edited by C. Baeumker. Monasterii: Aschendorff.

1970. Livre de la source de vie (Fons vitae) . Translated by J. Schlanger. Paris: Aubier Montaigne.

2001. Fons vitae = Meqor hayyim. Translated by R. Gatti, Testi e studi di filosofia ebraica medioevale; 21. Genova: Il melangolo.

§220. 2. כתאב אצלאח אלאכלאק, “The Refinement of Character Traits” (= Ethics),⁹¹ which correlates the virtues |381| with the five senses. The author affirms that this is his original idea,⁹² representing it at the end of his introduction by drawing a table or diagram, similar to the square of opposition that can also be found in books on logic. The Hebrew poem on the four elements, inserted after the table in the edition of 1562, exhibits the acrostic אריה⁹³ but it does not appear in one of the old manuscripts or in the edition of 1550. [The incipit of this poem is: אש ומים איך דבקו]. Ibn Gabirol apologizes (1562, f. 5) for not being able to furnish either intellectual (שכליים) proofs, i.e., those based on formal logic (חכמת הדבר?) or biblical analogies (ההקשות), although he had meant to do so, since “the power of the flesh is weak” כי כח הבשרי חלש כ“ש לאיש אשר כמוני מה שאני בו מריב ההקנטה (?) ומיעוט השגת החפץ. This means that he did not intend to write a wholly popular treatise. Nevertheless, neither his language nor his method is strictly scholarly. He has, however, inserted quite a number of biblical passages that the editors have not always indicated as such. Part 3, chapter 3 on trust in God is not much more than a collection of Biblical verses; sayings (from Ḥunayn Ibn IsḤ̣̣̣āq’s Maxims of the Philosophers (see §200)) are also inserted, and what is particularly noteworthy, many verses from Arabic poetry. I promised to publish these, but I have since given my transcriptions to a student of mine. [The student remains unidentified; perhaps it was A. Löwenthal, who in 1896 published Ḥarizi’s translation of Ḥunayn’s Maxims, with a German translation and with comparisons to Ibn Gabirol’s Ethics .]

Oxford, Bodl. Ms. Hunt. 382 probably gives the date of composition as [4]828 (1068), but the first letter of the number has become illegible. The earliest Hebrew sources for the most part point to Saragossa, [4]805 (1045).⁹⁴ [Another fairly complete Judaeo-Arabic ms. is New York, JTSA ENA 4038 (identified by Y. Tobi; see his description in 1993, 291–318.)] A few fragments from late manuscripts have also come to light. [The Judaeo-Arabic original was published, along with an English translation, by S. Wise 1901.]

This small treatise was translated under the title תקון מדות הנפש by Judah Ibn Tibbon (between 1161 and 1167?) and was published, together with BaḤ̣̣̣ya’s book on ethics, in Constantinople (around 1550),⁹⁵ then in Riva di Trento in 1562 in quarto format, together with Ḥunayn’s Maxims of the Philosophers and the pseudo-Aristotelian De pomo, having a title page with the words גורן נכון (Exodus 2:6) in the top line – a scholarly pun of the editor which was taken up in the reprint of this collection, Lunéville (Ibn Gabirol 1807), in quarto format. S. Silbermann published the book of ethics on its own in Lyck (Ibn Gabirol 1859a), in duodecimo format under the inappropriate title ס' גורן נכון והוא תקון מדות הנפש, based upon a copy of the edition of 1807. B. Goldenberg and L. Dukes furnished Silbermann’s edition with some corrections, on the basis of the original and one manuscript [of Ibn Gabirol’s book], but [consulted] two manuscripts of Ḥunayn’s collection of sayings.⁹⁶ Added to it is a letter of the translator to Asher, son of Meshullam [see the reference in §215 to the study of I. Ta-Shema in the supplementary note ad loc .], which I published first in the Michael Catalogue (Steinschneider 1848), 336 – Munk (1859, 168) knows only the Paris copy⁹⁷ – which may be regarded as a dedication to our book, and which at the same time completes the foreword of the translator |382| to BaḤ̣̣̣ya’s work (§215). When Judah read the first treatise of BaḤ̣̣̣ya together with Asher, he, Asher, told Judah that a small treatise of Ibn Gabirol already contains the subject matter of the other nine treatises of BaḤ̣̣̣ya; and the praise that Judah heaps upon the little book [of Ibn Gabirol] is remarkable. The designation of the sources as מוסרי הפילוסופים has already been stressed. Asher hoped that a translation of Ibn Gabirol’s little treatise would replace the treatises of BaḤ̣̣̣ya. Judah endeavoured to fulfill the wish of both the father and the son by translating according to the same principles. [The “father and son” refer to Meshullam, who commissioned the translation of BaḤ̣̣̣ya’s Duties of the Heart , and Asher, according to Steinschneider’s supposition of their relationship; see the article of Ta-Shema cited above.] Those places, however, where Ibn Gabirol quotes Arabic verses, Judah left blank, intending to fill them with analogous Hebrew verses by Ibn Gabirol, [Samuel] ha-Nagid, or other poets, or else to compose appropriate verses on his own. He had hoped to put them in the margin when he would find the time. Apparently Judah never found the opportunity to keep this promise. The editors have obscured the gaps that necessarily appear wherever the quoted verse of the poet (המשורר) does not follow.

Among the names of the sages who are quoted we encounter one which is corrupt in all editions and manuscripts, but which I could establish, namely Buzurjmihr (I, 2).⁹⁸ The book הקוטי, mentioned also in I, 2, is an Arabic versification of the Psalms by Ḥefeẓ al-Quti, still extant.⁹⁹ [This has now been edited with a French translation by Marie-Thérèse Urvoy Ibn Albar al-Quṭi 1994.]

In the editio princeps and in several manuscripts the translation starts with a poem¹⁰⁰ which, in the other editions, is placed at the end. In the editio princeps, the poem is ascribed to the translator.

Concerning subject matter and style, the book on the whole offers few difficulties, but we find a number of peculiar words, such as הקנטה (see HUe, 381), אלפלכייה֞) הפלכיים) as it has to be indubitably read,¹⁰¹ הנפש הדברנית (1807b, f. 7, line 1), usually (המדברת (אלנאטקה֞. The Arabic words מלקה (V, 3) and סירגא (towards the end) are corrupted from …? [The text in HUe is blank; from Wise’s edition of the Arabic we can supply מוקא] and חליים המשופם ?כירגא (towards the end, missing in the Lyck edition), perhaps ought to be read העפושיים?

One encounters, however, terms that are specifically coined – we do not know by whom first – to denote the virtues, e.g., סבלנות (I, 2 f. 10 b), that, in Choice of Pearls (chapter 10; §221 ff.) are rendered ציקנות ;סבל, synonymous with כילות stinginess (V, 2).¹⁰²

§221. 3. (A Choice of Pearls). We meet here with a delicate literary problem, feeling like a traveler who takes a road with the conviction that he will face |383| insurmountable obstacles, but who cannot and will not stay behind. Let me first of all confess that I do not wholeheartedly consider the book that I shall review next to be an authentic work of Ibn Gabirol. The reasons for and against [its attribution to him] will soon become evident. The structure of this paragraph will necessarily deviate somewhat from that of others.¹⁰³

[The current consensus appears to reject the attribution of the Choice of Pearls ( MivḤar ha-Peninim ) to Ibn Gabirol and, though the matter is difficult to assess, it seems that Steinschneider’s detailed and forceful arguments here had a telling effect. A. M. Habermann 1944 , in his very full bibliographical survey of the Choice , its printings, recensions, additions to the text, and translations, cautiously notes that he does not see sufficient evidence to justify the attribution to Ibn Gabirol. More recently Yehudah Ratzaby 1988 denies Ibn Gabirol’s authorship of the text with a new argument: his own study reveals that the Sirāj al-Mulūk of Abū Bakr al-Ṭurt ̣ushi (d. 1126) is a major source for the Choice and, on chronological grounds, it is very doubtful that Ibn Gabirol could have made use of that work. (Note Ratzaby’s additions to Haberman’s bibliography on 98 n. 9.) In a dissenting article in the same journal 1989 , Sarah Katz speculates that al-Ṭurt ̣ushi may have used Ibn Gabirol’s work. The latest scholarship moves away from the question of authorship: H. Ben-Shammai takes no stand in 1991 , going so far as to declare that the question of Ibn Gabirol’s authorship is not the important one in investigating the book.]

A collection of aphorisms, almost all of them anonymous, arranged in 64 chapters (“gates”) according to subject matter, is extant in a great number of manuscripts too many to be enumerated here, and in many editions, some of them accompanied by explanatory notes or a translation.¹⁰⁴

The first edition was published by Soncino (in Italy) in 1484, with a short commentary, probably by Samson Munay ‹from Joigny? See below, §575, n. 144› [The commentator is indeed Samson of Joighny, according to Richler 2001 , 194] called תודמ ירעש¹⁰⁵ in Parma, BP Cod. Parm. 2908 , fols. 199a–240b; the ms. was copied in 1338 and also includes the additions (תופסות).¹⁰⁶ This edition is, by the way, the first product of the press of the renowned Soncino family. The second edition of Ibn Gabirol’s work, 1739 and 1767, covers the Hebrew text with a Judaeo-German translation. There follow printings with a German translation (Ibn Gabirol 1842), a plagiarized version of which appeared in 1844; a Latin version of 299 aphorisms, published by Jo. Drusius in the third part of his Apophthegmata 1591 and 1612; and a collection of 750 aphorisms, fully vocalized, with a Latin translation done by Jacob Ebertus and his son Theodor, published in Frankfurt a.O. 1630. H. Filipowski edited the Hebrew text on the basis of one manuscript, without the additions Ibn Gabirol 1851). B. H. Ascher produced an edition of 652 paragraphs or aphorisms (Ibn Gabirol 1859b) – their arrangement not wholly correct^{106 b} – on the basis of five manuscripts, again without the additions, but with an English translation, an introduction and many instructive notes, particularly concerning parallel passages (part of them due to L. Dukes).¹⁰⁷ – I have presented a sample of rhymed translations with notes in my book Manna (Steinschneider 1847), nos. 87–125. I shall skip the commentaries and the Hebrew compendia.

There is neither an old manuscript nor any edition up to the nineteenth century that names an author or translator. The conjecture of Gaulmyn pointing to Jedaiah ha-Penini (thirteenth century, see Wolf 1715, I, 440) as author was accepted by a recent edition (Ibn Gabirol 1842) and its plagiarized version; it does not merit a refutation. L. Rosenthal 1875, 178, no. 927, strongly rebukes this error and, before presenting an idea to be mentioned presently, says: “All this is wrong. It is Solomon Ibn Gabirol who collected the sayings of the Arabs and translated them into Hebrew, as every discerning person is aware of.” This contention, however, is merely a false conjecture of Filipowski (1851) and has already been rejected in Steinschneider 1852, 2321.

Just as Judah Ibn Tibbon preserved for us the memory of his rival Joseph KimḤi in his translation of the Duties of the Heart, so, too, the latter has |384| preserved the name of the author and the translator of the Choice of Pearls – if the foreword is authentic and [textually] correct. This is the next point to be discussed.

Joseph KimḤi (or KamḤi?)¹⁰⁸ b. Isaac, father of the two famous grammarians David and Moses, alias “Maestro Petit,” a name which probably remained within the family,¹⁰⁹ [originally] from Spain, [but living] in Narbonne (perhaps also in Lünel?), was well-versed in Arabic.¹¹⁰ He composed a Hebrew grammar,¹¹¹ commentaries to some books of the Bible, etc. We have mentioned him already as translator of the Duties of the Heart. He put the sayings of the Choice of Pearls in verse form under the title שקל הקדש (The Holy Shekel). [The book is extant in eleven manuscripts: Basel, UB R III 2 ; Budapest, MTA Ms. Kaufmann 291/4; Budapest, MTA Ms. Kaufmann 528/4 ; Cambridge, UL Add. 377/8 ; New York, JTS Ms. 1495/2 ; Oxford, Bodl. Ms. Mich. 146 ; Oxford, Bodl. Ms. Opp. 639 ; Oxford, Bodl. Ms. Poc. 50 ; Paris, BN héb 983/3 b; ‹Parma, BP Cod. Parm. 2620 ›, and St. Petersburg RNL Ms. Evr. II 104/9 . Steinschneider lists six of these, including Parma, BP Cod. Parm. 2620] (according to 1872d, 31, Perreau 1889, 23, deficient);¹¹² and Rabinowitz Supplement 1887 under 120, now Kaufmann.^112b Is the copy, which Buxtorf obtained from Jacob Roman in Basel? [ It is.]

The importance of the problems relating to this book as well as the absence of a complete edition, a scholarly desideratum, gives us reason to indicate here the extant extracts in detail. [In 1919 Hermann Gollancz edited the text on the basis of Oxford Bodl. ms. Pococke 50, with reference also to Oxford Bodl. ms. Oppenheim 639, and provided an English translation in KimḤi 1919 . Gollancz raised the possibility that KimḤi was familiar with the Arabic version of the MivḤar ha-Peninim and that he consciously adapated its maxims to his own style.]

There are twenty-seven numbered aphorisms from Oxford, Bodleian Library Mich. 146, of which no. 16 combines two which we call a. and b. [(= ed. Gollancz, nos. 193 and 194)] which were printed by Dukes in 1842, 99, called here “Dz.” [The ms. is a nineteenth-century copy of aphorisms of Shekel ha-Kodesh , made perhaps by Dukes in Hamburg, from transcriptions of M. H. Bresslau – see letter d below – that were in turn made from the Oppenheim ms.]

• (a) Foreword, introductory poem and 33 numbered aphorisms (The final homonym of chapter one is XII!), in their entirety from Paris, BN ms. héb 983 in Ben Asher, Aaron b. Moses and Jacob ben Nissim 1846 = Dl.

• (b) 8 unnumbered aphorisms in a little article dealing with our book by Dukes 1850c, 507, 508, 521, 522 = Dll.

• (c) 33 unnumbered aphorisms in a small collection, edited by H. Edelmann 1852, also with an English paraphrase by M. H. Bresslau, 51 ff; there is no indication which Bodley manuscript was used in this edition (12 aphorisms thereof already in Dz. and Dl.) = ED.

• (d) Joseph KimḤi inserted more than 30 sayings in his commentary on Proverbs, incompletely edited under the ridiculous title ספר חוקה (Breslau 1867). = Pr. (with numbering of the chapters and verses which the ignorant editor had not indicated). [The title comes from the last words of the introductory poem of the commentary; the butt of Steinschneider’s scorn here is Dov Ber of Dubrove 1867. The commentary was critically edited on the basis of the extant manuscripts by Ephraim Talmage 1990.]

A few of these items were published by Dukes in an article on the last-named book 1850a, 358, 378, 389, 391. Seven of them appear in a and b as well. There is not a single saying quoted in the various works that does not belong to one of these series.

All manuscripts of the Shekel ha-Kodesh begin with a foreword. |385| We had the unusually good chance to be able to make use of five sources from which the foreword was printed,¹¹³ but this embarras de richesse does not make things easier, nor does it furnish us with any conclusive evidence. On the contrary, every manuscript presents us with a different picture and, apart from a short passage where KimḤi speaks in first person, they are all different from one another. When we discuss the main points we notice that the most complete recension (R = Rabinowitz [now Budapest, MTA Ms. Kaufmann 528/4 ]) lacks the Hebrew book title which occurs in the next recension and even in the abbreviated recension (B = Uri ms. [now Oxford, Bodl. Ms. Poc. 50 ]).

• (a) The author’s name, Solomon ben Judah בן גיברול (Gibrol), occurs only in R. and in Buxtorf [now Basel, UB R III 2 , in fact, it appears in Oxford, Bodleian Library Oppenheim 693 as well.]

• (b) The Arabic title, corrupt almost everywhere, is in R. מנתכ'ל אלג'והר, probably for מנתכ'ל אלג'ואהר; in Mich. 846 it appears corrupted as מתנדל; מבקאר in Opp. [= Oppenheim 683? But the reading there is מכ'תאר ] is indubitably to be read, with Buxtorf, מכ'תאר (and consequently מובחר, “chosen”).¹¹⁴

• (c) The name of the Hebrew translator of the מבחר הפנינים is given, in O. and R., as Judah b. Saul b. Tibbon “from Sevilla” (אשביליה). This note is very suspect and perhaps an addition on the part of the copyist. [A. Marx 1927 , 436–37, argued on the basis of another manuscript that the note is indeed by KimḤi. The manuscripts that mention Ibn Gabirol as author of the MivḤar ha-Peninim and Judah Ibn Tibbon as its translator represent a later version of Shekel ha-Kodesh made by the author, which was partly rearranged according to the order of MivḤar ha-Peninim .]

• (d) All manuscripts have the passage: “I, Joseph b. Isaac (ben, in R.) KimḤi, have translated it into (or, from) the Hebrew language.” This could well mean that KimḤi has versified the Hebrew translation, done in prose, by Tibbon. In Paris, BN héb 983, we find after this passage: “I found (the sayings) in prose (oratione soluta מפורדים, or scattered? ‹cf. 885›), partly in Arabic, partly in Hebrew.” R. has: “I have added sayings from other books.”¹¹⁵ One thereby understands better what follows.

• (e) KimḤi collects sayings that are related to each other, i.e., those that treat the same subject; or else he arranged his additions according to the base text. He divided everything into 22 chapters, according to the number of the Hebrew letters, and concluded every chapter with a distich in homonyms (instead of rhymes) that indicates both the contents and the number of sayings – perhaps an imitation of the Tarshish of Moses b. Ezra (according to Dukes 1850a, 728, n. 3). E. Carmoly, who published a list of 22 chapters, dared to state that KimḤi’s book has nothing to do with the Choice of Pearls and that all those who identify the two books speak so only from hearsay!¹¹⁶ Has he done more than copy the table of contents? Unfortunately, we have before us only the extracts, noted above, whose second series alone indicates the titles of Kimhi’s chapters. The first series indicates the titles of the printed book (with two errors)¹¹⁷; the others do not have anything like this. We would understand even if a scholar more punctilious than Carmoly, after having read the foreword of the Paris manuscript, without comparing carefully |386| and having perused the printed book, would have doubted that the latter is reproduced in the former.

The published portions and quotations from KimḤi’s version contain, not counting identical passages, only approximately 100 sayings; one must spend several hours in order to identify even less than half of the sayings in both books.¹¹⁸ [Gollancz published a table comparing the aphorisms of the Shekel (in his edition) to similar ones in the Choice (in Asher’s edition), see 1919 , 126–29; see, however, Marx 1927 , who argues that there are two recensions of Sheke l] . This, however, is sufficient proof of the fact that KimḤi made use of the Choice by including, in his divergent arrangement, at least part of the titles not only of the chapters but also of individual sayings. This is not prima facie evident to someone studying the arrangement of the Choice, where the arrangement of the sayings under rubrics that are ethical terms is very vague. For example in the Choice we find chapter headings such as “commendable habits” (in Asher’s English, urbanity!) (ch. 38), “guidance to the good path” (ch. 42), “rules (testaments?) of the wise man for his son” (ch. 43), etc. The author has not paired contrary terms, and the identical aphorism is listed as no. 116 in chapter 5 and as no. 648 in chapter 64. Perhaps KimḤi thought (cf. the end of the foreword) that he had grouped together the material which, to him, seemed to belong to the same subject. All the more, then, must we assume that he did not compile his work in full independence [of the Choice ], since, even in the extracts, many sayings have the same order of arrangement as in the Choice of Pearls, e.g., Dl. 7–9 = Vint. 15–17. On the other hand, the conflation of various chapters and the re-arrangement of individual aphorisms do not allow us to indicate the exact relationship between the parts of the Choice and those parts that have been added from other sources. For instance, KimḤi counts, in the distich at the end of chapter one, 140¹¹⁹ lines whereas the Choice has 75; perhaps, though, part of this chapter has been taken from others.

Another question cannot be answered: Is the redaction of the Shekel ha-Kodesh based on the Arabic original or on the only Hebrew translation? Dukes corrects the reading of one aphorism of the Choice on the basis of the Shekel,¹²⁰ but that does not prove dependence: KimḤi could well use the same Hebrew word as the translator of the Choice. The same applies to the similarity between the few aphorisms that we could compare. (The reading ובית שחת in KimḤi [Dll. No. 8, 522] seems to be, however, a variant reading of מבלי צוות no. 544, Choice, 108, and not מאין, as on 176.)KimḤi may have known, and made use of, the prose translation, as the foreword to R. seems to indicate; but that does not at all mean that he did not know the Arabic text. On the other hand, the mutual divergences in the known identical aphorisms of both books are not proof of their origin from an Arabic text, for KimḤi, like any versifier, could and had to work in almost boundless freedom. Finally, his source might be another Arabic or Hebrew text: Nos. 36 and 37, for instance, dealing with the wise and the rich, are attributed in Ḥunayn’s Apophthegms to Diogenes¹²¹; KimḤi (Dz. no. 8) applies, so it seems, the [rule] omnia mea mecum porto. On his commentary to Proverbs 22:1 (ed. Dubrov, KimḤi 1867, 30, [ed. Talmage, KimḤi 1990 , 111] KimḤi adduces no. 540 of |387| the prose version of the Choice, in a slightly different version¹²²; and that in a way confirms the argumentum a silentio that this commentary was composed before the Shekel, which is never mentioned in it. The Shekel is at any rate not a translation proper but a strongly paraphrasing versification. One example may show this: The simple saying no. 202 “Covetousness is the companion of blindness (delusion)” becomes in KimḤi (ed. Edelmann 1852, no. 3, [ed. Gollancz, no. 144, 1919] a distich four times as long.¹²³

We have analyzed the relation between the Shekel and the Choice from a historical perspective without trying to characterize the former, either philologically or aesthetically, on the basis of the extracts. Nevertheless, we cannot refrain from remarking that pithy sayings generally lose more than they gain when versified and amplified.¹²⁴ Now we have to return to the Choice.

Mss of Shekel ha-Qodesh

Basel, Universitätsbibliothek R III 2 (Basel 37 (Cat. Allony & Kupfer)) (IMHM F 2569, F 8857), 1a–37b.

Budapest, Magyar tudományos akadémia 291 (IMHM Fiche 78), 256–63.

Budapest, Magyar tudományos akadémia 528 (IMHM Fiche 32), 139–76.

Cambridge, University Library 377 (SCR 774) (IMHM F 16296), 662–91.

New York, Jewish Theological Seminary Ms. 1495 (Ms. 6322/Ms. Acc. 1134) (IMHM F 39179), 61b–75b

Oxford, Bodleian Library Ms. Mich. 146 (Ms. Mich. 846) (Neubauer 1180) (IMHM F 16639), 116a–19a.

Oxford, Bodleian Library Ms. Opp. 639 (Ms. Opp. 1404) (Neubauer 1180) (IMHM F 19137), 18 fols.

Oxford, Bodleian Library Ms. Pococke 50 (Uri 497) (Neubauer 1976) (IMHM F 19138), 1b–32b.

Paris, Bibliothèque Nationale héb 983 (a. f. héb 245) (IMHM F 30343), 10a–21b.

Parma, Biblioteca Palatina Cod. Parm. 2620 (De Rossi 1393) (IMHM F 13536), 29a–35b.

St. Petersburg, Russian National Library Evr. II A 104 (IMHM F 64120), 8 fols.

§222. People have grown accustomed to attributing the lost Arabic original of the Choice to Ibn Gabirol¹²⁵ without taking into proper account the fact that we have no other testimony than the passage of KimḤi. That passage remained unknown to all Jewish scholars who used the Choice, expressly or tacitly, and their number is great, their authority, considerable.¹²⁶ They are led by Judah Ibn Tibbon, who knows the book and quotes some of its sayings without referring to either the author or the translator.¹²⁷ Falaquera mentions the Choice as an example of those books that were composed by the “wise men of the nations,” an expression that must be understood to refer to the origin of the aphorisms. That expression proves at least that Falaquera did not think of Ibn Gabirol as the compiler. We have to admit that all this is strange and requires an explanation, if one believes the – not unsuspect – note in KimḤi’s foreword according to the unique ms. R. to be correct. [The note appears in several manuscripts.]

This is not all. Ibn Gabirol’s Ethics contains a number of sayings from the Choice of Pearls without any indication of their source,¹²⁸ nor do they always exhibit the same wording – and this is of importance for [clarifying] the problem of the translator. Sometimes one can correct the text on the basis of these quotations.¹²⁹ In the interpretation of the word דאגה there is a contradiction between no. 140 and Ibn Gabirol III, 2. In the whole text of the Choice of Pearls no actual author’s name or book title is to be found |388| (for no. 242 Aristotle, see above 258; no. 637 “a book on medicine”), while there is no dearth of names and titles in the Ethics. We emphasize four sayings in which Ibn Gabirol names his authorities twice, as does Ḥunayn in his Apophthegms (which apparently are used in the Ethics): 118 (III, 2 Socrates), 617 (II, 4 Plato), 170 (I, 2 Buzurjmihr, cf. 382), 538 an Indian king (I, 1 Azdashir instead of Ardashir which may easily be explained by the similitude of the Arabic letters).¹³⁰ Even if Ibn Gabirol had compiled the Choice of Pearls in his youth for his own use, as has been maintained, no reason for the omission of the names can be adduced. Furthermore, the arrangement of the aphorisms, already mentioned, does not fit with the systematic nature of Ibn Gabirol.

It is equally strange that the name of Judah Ibn Tibbon can be found only in that passage of KimḤi, which makes him an inhabitant of Sevilla, while everywhere else, including the epigraphs of his son Samuel and his grandson Moses, we read from Granada (מרמון ספרד). The language – one can hardly speak of style in the context of such aphorisms – offers no criterion for our problem: nothing is conspicuous, and there are no hints as to the native country or the date of the translator. Some forms, such as הפצקה or תופצקתה (no. 121), רומח (172) have always been coined anew in scholarly works. B. Asher (Ibn Gabirol 1859b, 176, n. 544) emphasizes the Aramaic word צוות (society) which also occurs in BaḤya’s Ethics (X, 3).

[Habermann and Ratzaby, both of them not without some hesitation, accept the identification of Judah Ibn Tibbon as the translator. Ratzaby points out that Judah Ibn Tibbon’s formulations of five aphorisms in his translator’s introduction to Bahya’s Duties of the Heart are identical to their formulation in the Hebrew translation of the MivḤar ha-Peninim . But this shows at best that Judah Ibn Tibbon was familiar with that translation, not that he was its author. On the other hand, Ratzaby also calls attention to a British Library ms. (without identifying it, but it is Add. 18684 (Marg. 514/2)), in which a certain R. Solomon Ibn Tafu is named as the translator; this was noted already by Steinschneider in 1852 , 2322.]

4. De anima? A book on the soul, translated into Latin, perhaps composed by Ibn Gabirol, partly translated from Latin into Hebrew (see above, p. 21) is in the course of publication. [The work was published in 1891 by Steinschneider’s student, A. Loewenthal 1891 ; for the question of its authorship, see above, §4d.]

§223 (Israeli.) Abū Yaʽqūb IsḤāq b. Sulaymān al-Isrā’īlī (died around the middle of the tenth century; see §479), called in Hebrew sources Isaac ben Solomon Israeli, in Latin sources Ysaacus, is one of the most famous physicians of the Middle Ages. Maimonides (see above, p. 41) values him only as a physician, not as a philosopher. Jacob b. Reuben,¹³¹ a contemporary of Maimonides, bases his interpretation of some biblical verses on Isaac’s authority, “for he was a scholar of the Torah and of the seven disciplines¹³²; all the wise men of his time as well as those who came after him (קמו תחתיו), those who profess the Torah¹³³ or another [revelation], call him a ‘philosopher,’ not as Abraham Ibn Ezra |389| does, a ‘babbler’ (מהביל).¹³⁴ For all his books were well received by the wise men, and they took their arguments from them.” Arabic bibliographers name some titles of his philosophical works,¹³⁵ e.g., a book on logic which seems to be lost. On the other hand, the Jews preserved some fragments, in Hebrew translation, of books and treatises that remained unknown to the Arabs, probably because they were written only in Hebrew letters for Jewish readers. We shall first deal with two books known to the Arabs.

[A. Altmann and S. Stern, Isaac Israeli: A Neoplatonic Philosopher of the Early Tenth Century , Oxford, 1958 , is an authoritative study to which we shall refer in the following as either Altmann 1958 or Stern 1958, depending upon the section. (Altmann wrote on the Mantua Text and the Book on the Elements ; Stern on the Book of Definitions , the Book of Substances , and the Book on Spirit and Soul. ) With the exception of the commentary to Sefer Yeẓira , no longer attributed to Israeli, the writings discussed below have attracted little scholarly interest during the past half century.]

1. The Book of Definitions and Descriptions (כתאב אלחדוד ואלרסום).¹³⁶ [See Stern 1958, 3–78, for a discussion and full translation of this work.] In the beginning of Isaac’s works, usually given the title Opera Ysacii ¹³⁷ (Lyons, Israeli 1515), we find a Latin translation of this little treatise.¹³⁸ Its title is (f. 2): “Collectiones ex dictis philosophorum de differentia inter descriptiones rerum et diffinitiones (sic!) earum et quare philosophia fuit descripta et non definita” etc. It ends, “Hic finiuntur collectiones Ysaac” etc. This title is most probably a literal translation of the text, because it is similar to the titles of other works by Isaac, which he considered to be compilations of the views of the ancient philosophers. In the beginning of the Tabula there is the passage, “Liber definitionum ysaac heben amaran philosophi,” a confusion of Isaac with his teacher, the Arab physician IsḤāq b. ‛Imrān (vulgo ‛Amrān, died around 900),¹³⁹ which already occurs in a Hebrew compilation of the thirteenth century.¹⁴⁰ The Latin translator is not named. The list of the translations of Gerard of Cremona contains in the section on Physica (because the author is a physician!), no. 55: “De descriptione rerum et diffinitionibus etc.” The switch in the wording of the title leads us to surmise that the author of the list did not have the book before him. Some manuscripts contain the printed book under the name of Gerard.¹⁴¹ There is no direct evidence that Constantinus Africanus translated either this or the following (§225.2) book, as he had done with the other printed works of Isaac. [Steinschneider refers here to Israeli’s medical writings; see §479.] This book has yet to be compared with other works of these two translators. |390| Thus we have not found a decisive argument in favor of either one of them.

We face here two difficulties: the loss of the Arabic original, and the peculiar character of this treatise in its Latin translation, which reads more as an agglomeration than as a development of ideas. Indeed, it may rather be called a compilation of fragments. [Part of the Arabic was recovered from the Cairo Genizah and was published by Hirschfeld 1902; for criticism of this edition see Stern 1958, 3.]¹⁴² Constantinus was less faithful as a translator than Gerard, because he abridged the text, whereas Gerard’s Latin, as has been observed, sounds much like Arabic.¹⁴³ Our book begins with the explanation of four Arabic terms: anitas (הליה֞), quidditas, qualitas, and quaritas (למיה֞).¹⁴⁴ Do these terms occur in the other translations of Constantinus? Had he occasion at other places to make use of them? I do not want to decide whose translation it is; but the comparison of our book with the next (§225) yields some clues for choosing one or the other opinion. ‹The printed edition is incomplete compared with ms. 14,700 and the same Vienna 2325, cf. Bäumker 1892, 126. [?]›

[Gerard de Cremona has been securely identified as the translator of the Latin text found in the printed edition and several manuscripts; an edition of his translation was made by Muckle 1937 –38. Stern made several suggestions for improvements to Muckle’s edition in the textual annotations to his translation 1958. In addition, there is an anonymous Latin abbreviated version which is based on Gerard’s translation; it also was edited by Muckle 1937 –38.]

§224. The Book of Definitions was known in Spain already in the eleventh century, for the Qāḍ ī Ṣā‛id al-Andalusī lists it in his article on Isaac Israeli, which the translator of the Book of the Elements ¹⁴⁵ reproduces. Maimonides quotes it under an abbreviated title (§222).¹⁴⁶

A treatise (חבור) by Isaac, translated by Nissim b. Solomon, is contained in: London, Mon. 305/4 , Milan, BA C 116 Sup. , Oxford, Bodl. Ms. Mich. 335 , Parma, BP Cod. Parm. 2105 , Vatican, BA ebr. 236/9 . Oxford, Bodl. Ms. Mich. 335 contains roughly a third of our treatise, namely until the eighth condition (f. 2³, l. 4 of the Latin).¹⁴⁷

The explicit of London, Mon. 305/4 displays: בין החקיקה והתכלית נשלם מחברת של (יצחק בחקיקת הדברים ותכליתם והפרש (והפרק. [The text of Nissim’s Hebrew translation was published by H. Hirschfeld in the Steinschneider festschrift, 1896 . Stern 1958 , 6, announced the forthcoming publication of his own edition of the text, to appear in Hebrew Union College Annual , promising to refute the claim of J. L. Teicher that Nissim worked from the Latin, rather than from the Arabic. However, that publication did not appear in HUCA , nor anywhere else; it is not listed in the bibliography of S. M. Stern’s writings by Latham and Mitchell 1970 .]

We do not know for sure the date and the origin of the translator. A Nissim b. Solomon אגוסטרו) אגוסתרו?) copied mystical works of Eleazar of Worms in Paris, BN héb 850 .¹⁴⁸ [That Paris manuscript dates from the sixteenth century; hence it is very unlikely that the copyist, Nissim ben Solomon אגושתרו, is the translator of Israeli’s writings.] Nissim b. Moses b. Solomon from Marseille, a philosophical writer, lived in the beginning of the fourteenth century¹⁴⁹; Isaac Latif (end of the thirteenth century) quotes a passage from the book גבולי הדברים (The Definitions of Things) by Isaac, apparently from the Arabic text or from another source.¹⁵⁰ [Altmann 1957 , 236 n. 2 (cont. from 235), surmises that Ibn Latif cited directly from the Arabic. This same passage is cited by Gershon ben Solomon; see below, supplementary comment on §225. 2.] A historical passage (Latin f. 3⁴, last line and f. 4¹) may serve as an example of the two translations: Ysaac testific.: quidem me vidisse in egypto |391| huius modi. dominibatur enim eidem quidam ducum de filiis tolonis qui dicebatur filius talix ¹⁵¹; et magnificatum est eius imperium, et relaxata est memoria eius et terruerat ipsius exercitum: et erat dux exercitus eius quidam magister minutorum nomine bidel; ms. Halb. אמר יצחק אני ראיתי הדומה לזה במצרים שגבר עליה איש מבני טולון בנו [הנקרא?] אל כליך שגדלה מעלתו ושמו עלה ורבו חיילותיו והיה מושל על חיילותיו שחור כופר שהיה שמו כלאל (sic).

[A second Hebrew translation was identified in two St. Petersburg manuscripts, St. Petersburg, RNL Evr. II A 388 and St. Petersburg, RNL Evr. II A 412 by A. Borisov; the full text was published by A. Altmann 1957 .]

Treatise (Part of the Book of Definitions) , trans. Nissim b. Solomon

London, Montefiore 305/4 (Halb. 361) (IMHM F 5255), 13a–27b.

Milan, Biblioteca Ambrosiana C 116 Sup. (Kennicott 189) (Bernheimer 14) (IMHM F 12263).

Oxford, Bodleian Ms. Mich. 335 (Ms. Mich. 82) (Neubauer 1318) (IMHM F 22132), 45b–54.

Parma, Biblioteca Palatina Parm. 2105 (De Rossi 1246) (IMHM F 13324), 191b–95b.

Vatican, Biblioteca Apostolica ebr. 236 (IMHM F 292), 52a–62a.

Trans. Anonymous

St. Petersburg, Russian National Library Evr. II A 388 (IMHM F 64676), 2 fols.

St. Petersburg, Russian National Library Evr. II A 412 (IMHM F 64765), 2 fols.

§225. 2 (Book of the Elements) כתאב אלאסתקצאת (or, פי אלאסתקצאת, in Moshe Ibn Ezra, al-MuḤāḍara wa-l-Mudhākara, f. 79b), lost in the original. In the Lyon edition of the Latin works of Isaac it follows the Book of Definitions (f. 4³) as “Liber aggregatus ex dictis Philosophorum antiquorum de Elementis secundum sententiam Aristotelis et Hippocratis et Galeni de quorum aggregatione et oppositione solicitus fuit Ysaac Salomonis filius israelita.” This work appears also as one of Gerard of Cremona’s translations, no. 54 [in Wüstenfeld 1877 ], and some manuscripts name him as translator.¹⁵² No manuscript of the Arabic original is known; it is mentioned both by Ṣa‛id and Maimonides in connection with the previous work, and Moses Ibn Ezra quotes from it.

Abraham Ibn Ḥasdai translated the book for the famous grammarian, the elderly David KimḤi (around 1210–30) under the full title (ספר מחובר ממאמרי הראשונים (הקדמונים ביסודות על דעת הפילוסוף ¹⁵³מאבוקרט וגאלינוס ממה שעיין בקבוצו (וחבורו) יצחק בן שלמה הישראלי הרופא which literally matches the Latin title. Despite Maimonides’ critique it was studied. It is found in:

Manuscripts¹⁵⁴: Oxford, Bodl. Ms. Hunt. 576 (without foreword and defective in the end)¹⁵⁵; Oxford, Bodl. Ms. Mich. 600 ; Breslau, Saraval 19. [This manuscript is no longer extant, to our knowledge.] Leiden, BR Cod. Or. 4751 ³ (f. 37); Munich, BS Cod. hebr. 43/3 ; Paris, BN héb 325/3 ; Paris, BN héb 930/6 (the translator’s introduction is missing from both manuscripts)¹⁵⁶; last page of Paris, BN héb 1144 [f. 45a]; Parma, BP Cod. Parm. 2628 (no proemium); Parma, BP Cod. Parm. 2611 ; Parma, BP Cod. Parm. 3023 ; Parma, BP Cod. Parm. 3524 (formerly Foa 14?; Steinschneider 1865c, 67, Perreau 1889, 61, n. 45); Vatican, Urbinati ebr. 53 ; – Fischl was in possession of a ms. in 1881 (no proemium).¹⁵⁷ [Milan, BA T 30 Sup. 1; Berlin, SPK Or. oct. 516/4 ; St. Petersburg, RNL Evr. Arab II 2615 ; Rome, BC 2916 ; St. Petersburg, IOS C 14 ; St. Petersburg, RNL Evr. II A 109 ; St. Petersburg, RNL Evr. II A 195/6 ; Zürich, ZB Heid. 184 ] This list includes the translation by Abraham Ibn Hasdai and the anonymous one, perhaps Moses Ibn Tibbon. (See Israeli 1884 ). Steinschneider assumed that there was only one translation. What makes the matter trickier is that in some manuscripts the poem by the translator Abraham b. Samuel precedes the anonymous translation. The anonymous version is found in the following manuscripts: Oxford, Bodl. Ms. Mich. 600 Milan, BA T 30 Sup.; Munich, BS Cod. hebr. 43/3 Oxford, Bodl. Ms. Hunt. 576; Berlin, SPK Or. oct. 516/4; Parma, BP Cod. Parm. 2611; Parma, BP Cod. Parm. 3524; Paris, BN héb 1144.

The complete manuscripts begin with a poem of 12 stanzas, beginning קחו (ראו) ספר בא בחידות,¹⁵⁸ in which the translator, Abraham b. Samuel, is named. This is followed by the latter’s foreword in rhymed prose. Part of this has been published, though not very accurately, on the basis of the Leiden manuscript alone.¹⁵⁹ I had intended to undertake a complete edition, but since Mr. Fried has announced a publication of the complete translation, I confine myself to a short analysis, neglecting the textual variants, which are unimportant for our purpose. [The text was published on the basis of the Leiden manuscript by S. Fried in Israeli 1884 . Altmann 1958 , 133–50 presents an excerpt from the work, with preliminary note, translation, and comments.] The translator starts with the modest declaration that he would not have dared to undertake such a work had he not been commissioned by the great scholar, the “Ornament of the Elderly,” David KimḤi, who wanted to |392| disseminate scholarship. He gave a second reason for taking on the translation, as follows: The author, Isaac b. Salomo, the Israelite, wrote the book in Arabic, but he could well have written it in Hebrew, for he was a great scholar in our Law and one of the most renowned men of our nation – has he not composed a commentary on the book Yeẓira, thereby erecting a temple (monument) to himself?¹⁶⁰ One of the Arabic scholars, Ṣā‛id b. AḤmad b. Ṣā‛id from Cordova, is the author of a splendid work, elegantly written and well-organized, which mentions all famous scholars from all nations beginning with the earliest times, everyone according to his language, his origin, his religion and profession,¹⁶¹ his rank and position, among them the Jewish scholars about whom he was well-informed. In this work there is the following account (I am reproducing the gist only): Isaac etc., student of IsḤāq b. ‛Imrān, the Arab, called סם סע̋ה =) סם שעה), was a physician in the service of ‛Ubaydallāh al-Mahdī, King of Africa¹⁶²; he combined the science of logic with other sciences. He lived more than a 100 years, he did not marry, nor did he seek wealth.^162b He composed valuable works, e.g., the book On Foodstuffs (מאכלים, Diaetae); On Fevers, which cannot be weighed with gold and silver; On Urine; the priceless Book of the Elements, and many more, such as the Book of Definitions and Descriptions, the Paradise of Wisdom on metaphysical (or: theological)¹⁶³ problems, and others. He died in 330. [This passage from Ṣa‘id al-Andalusi is translated from the Arabic original in Altmann and Stern 1958 , xviii; there the date is given as 320 (= 932).] ¹⁶⁴ Abraham continues: Since Isaac worked under the order of the ruler, he composed his works in Arabic. For a long time, he remained unknown to the Jewish scholars. Abraham hopes to have his translation inspire other and better translators to render the other works of Isaac. A praise of the Book on the Elements follows, and the foreword ends with this remark: Many translators discuss in their introductions those passages (or rather: those circumstances) that caused them to slip, begging their readers’ indulgence; Abraham feels obliged all the more to ask the readers to correct his mistakes.

The book is divided into three treatises: The first sets forth the views of “the philosopher” (Aristotle), the second those of Galen (Hippocrates |393| is quoted on f. 8⁴ med., cf. III, 9, 4); the third treats the number and qualities of the elements. The style is almost always engaging: the author presents the possible objections (quod si contradixerit, etc.) in order to refute them. In the manner of the Arabs and Jews, his own remarks follow after the opinions he quotes, introduced by the formula: “Isaac says.” In the Hebrew one finds instead, “the author says.”¹⁶⁵

According to the catalogue description of Paris, BN héb 325/3 , the Latin translation differs “considerably” from the Hebrew. I have compared only a few passages, e.g., the end of the book, and have found them to be the same, almost word for word. One may wish to conclude from this that Gerard is the translator (and hence also of the Book of Definitions). In other places there are deviations which, however, do not go as far as the alterations which Constantinus otherwise allows himself. – So far, I have not been able to compare the quotations in Gershon ben Solomon¹⁶⁶ and his contemporary Hillel b. Samuel¹⁶⁷ with a Hebrew manuscript. [The citations in Gershon ben Shlomo’s Sha‘ar ha-Shamayim were matched to the Hebrew edition of the Book of Definitions by James Robinson 2000 , 258 n. 48, 259 n. 62, and especially 262 n. 80. The citations in Hillel of Verona, Tagmulei ha-Nefesh, were matched to the Hebrew edition by G. Sermoneta 1981, 13, 14, and 194.

As mentioned above, fragments of Israeli’s Kitāb al-Jawāhir (Book of the Substances) preserved in St. Petersburg, RNL Evr. Arab II 2615 , were edited by S. M. Stern in Isaac Israeli . There is an additional text by Israeli, extant only in Hebrew translation in a single manuscript at Mantua (Mantua, CI Ms. ebr. 28) , and it has come to be known as “the Mantua text”. Steinschneider took this to be a pseudo-Aristotelian text and discussed it in §123 above. Gershom Scholem first suggested identifying Israeli as the author, and his surmise was confirmed by Alexander Altmann, who published the full text with an accompanying analysis, first in the Journal of Jewish Studies, then in Altmann 1958, 118–32.

Lynn Thorndike 1923–1966 , 658 n. 3 records a Liber Ysaac de differentia spiritus et animae , in ms. Venice, San Marco 179, ff. 57–59, 83. Thorndike gives the impression that Israeli is just one in a long line of scholars to whom QuṢta b. Lūqā’s tract on the difference between soul and spirit was misattributed. He did not know that Israeli actually wrote a monograph on the same theme; hence there is cause to inspect this manuscript, in order to see if it contains a Latin version of Israeli’s book. No Latin translation is noted by either Steinschneider or Stern.]

Book of the Elements, trans. Abraham Ibn Ḥasdai

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4751 (Warner 13) (IMHM F 31909), 37a–52a.

Paris, Bibliothèque Nationale héb 325 (a.f. 158) (IMHM F 20237), 145–61.

Paris, Bibliothèque Nationale héb 930 (IMHM F 31966), 139a–54a.

Parma, Biblioteca Palatina Parm. 2628/6 (De Rossi 207) (IMHM F 13544), 46b–56b.

Parma, Biblioteca Palatina Parm. 3023/8 (De Rossi 771) (IMHM F 13752), 145a–57a.

Rome, Biblioteca Casanatense 2914 (Sacerdote 167) (IMHM F 747), 85a–101b.

St. Petersburg, Institute of Oriental Studies IOS C 14 (IMHM F 69265), 2a–14a.

St. Petersburg, Russian National Library Evr. Arab II 2615 (Evr. Arab II 1197) (IMHM F 61142), 17 fols.

St. Petersburg, Russian National Library Evr. II A 195/6 (IMHM F 64329), 1 f.

St. Petersburg, Russian National Library Evr. II A 40 (IMHM F 65703), 30 fols.

Vatican, Ms. ebr. 53/3 (IMHM F 692), 63b–92a.

Zürich, Zentralbibliothek Heid. 184 (Allony and Kupfer 144) (IMHM F 2720, F 10392).

Trans. Anonymous

Berlin, Staatsbibliothek (Preussischer Kulturbesitz) Or. oct. 516 (Steinschneider 201) (IMHM F 1971), 89b–116.

Milan, Biblioteca Ambrosiana T 30 Sup. (Bernheimer 100) (IMHM F 14617), 108b–36b.

Munich, Bayerische Staatsbibliothek Cod. hebr. 43 (IMHM F 1150), 21a–52a.

Oxford, Bodleian Library Huntingdon 576 (Uri 408) (Neubauer 1316/1) (IMHM F 22130), 1–56 fols.

Oxford, Bodleian Library Ms. Mich. 600 (Mich. 600) (Neubauer 1368/2) (IMHM F 19402), 69a–85a.

Paris, Bibliothèque Nationale héb 1144/2 (Paris, Oratoire 140) (IMHM F 15104), 45a.

Parma, Biblioteca Palatina Parm. 2611/5 (De Rossi 423) (IMHM F 13312), 61b–77a.

Parma, Biblioteca Palatina Parm. 3524/3 (Parma Perreau 45, Parma Stern 55) (IMHM F 14031), 57b–82b.

§226. 3. A treatise מקאלה פי ישרצו המים, doubtlessly written in Arabic¹⁶⁸ and referring to Genesis 1:20, is known through a number of quotations¹⁶⁹ dating from the twelfth century, as well as through the small fragment of a Hebrew translation that S. Sachs identified in Munich, BS Cod. hebr. 47/6 under the wrong title ס' יצירה, and published twice.¹⁷⁰ Perhaps the translator could not, or would not, translate more than this initial part. According to the title, or the introductory remark of the translator, this treatise was actually a responsum to somebody who had detected a contradiction in the biblical tale of the origin of the birds. The text breaks off at a passage where the author starts to speak about the elements,¹⁷¹ concerning which, as is well known, he was in his very element. Probably he endeavored at this point to give a philosophical interpretation of the entire creation. In his introduction to the commentary to the Pentateuch,¹⁷² Ibn Ezra speaks about one Rabbi Isaac who had written two books (or two volumes¹⁷³) on the first chapter of Genesis.¹⁷⁴ There is no sufficient reason to doubt the authenticity of that title, which was also known to Jedaiah Bedersi (ha-Penini)¹⁷⁵ and, therefore, to consider this treatise |394| to be part of a commentary to the whole of Genesis,¹⁷⁶ or to identify this monograph with the work on metaphysical questions.¹⁷⁷ On the other hand, the little treatise – no. 4 below – may have been part of the one under discussion. [For further discussion of the treatise see Stern 1958 , 106–7.]

We finally call attention to a passage in which the author maintains that Jesus wanted to be crucified in order to confirm (or uphold) the erroneous opinion of his divinity. [This last sentence may be restated more clearly. Israeli cites a Christian view that Jesus wished to be crucified in order to establish his divinity, as an example of an interpretation that has taken hold on account of consensus, rather than having been established by investigation. As a Jew, Israeli brands this Christian claim “erroneous”.]

Treatise on “Let the Waters Increase”

Munich, Bayerische Staatsbibliothek Cod. hebr. 43 (IMHM F 1611), 325b–26b.

§226^b. 4. Munich BS Cod. Heb. 307 contains a small treatise bearing the inexact title, ספר הרוח והנפש… “Book of the Spirit and the Soul and of the Difference Between Them.” I have published it in the journal Ha-Karmel 1872a and presented the parallel passages from the Book of Definitions and the Book of the Elements, which were omitted by the Editor [of the Hebrew article], in Hebraeische Bibliographie 1872b.¹⁷⁸ This treatise, four pages long, is most probably a fragment of a book or a longer treatise, perhaps the final part of the treatise on Genesis 1 (see §226), although the multiple concluding formulas seem to be due to the copyist. The treatise contains an interpretation, garbled in the manuscripts, of Job 42:15 which Abraham Ibn Ezra, attributing it to Isaac “המהביל”, seems to rebuff.¹⁷⁹ In another place (n. 134) Abraham says expressly that Isaac used the foreign sciences in order to explain “living souls” (the plural form is for the sake of rhyming with נפש חיה Gen.1:30, 2:7). [A preliminary note, translation, and comments to this treatise by S. M. Stern are available in 1958 , pp. 106–17.]

§227. 5. (Commentary to Sefer Yeẓirah) [of Abū Sahl Dūnash Ibn Tāmīm]

[Largely as the result of a series of publications by the late Georges Vajda, the texts discussed here in the entry under Isaac Israeli are now considered definitively to be the work of his student, Dūnash Ibn Tāmīm. Vajda’s Le Commentaire sur le Livre de la Création de Dunaš ben Tamim de Kairouan (X ^e siècle) was recently revised and augmented by Paul Fenton, and we shall refer to this new edition as either Vajda 2002 or Fenton 2002, depending upon the sections referred to (although even the sections originally written by Vajda have been revised by Fenton.) The studies of Luzzatto, Dukes, Sachs, Munk, and Kaufmann, all furnishing material for Steinschneider’s own investigation, are briefly reviewed on 15–16 (Fenton). Although attributing the “paternité” of the commentary to Israeli in his earlier writings, Steinschneider came to the conclusion that, in its various versions, it is the work of Dūnash. But because he considered the work to contain “at least parts of a commentary composed by Isaac Israeli,” Steinschneider lists it here among Isaac’s work; he also seems to have left the question of the authorship of certain passages unresolved (see, e.g., the end of n. 203 below.)

Paul Fenton has also discovered a large portion of the original Judeaeo-Arabic in the second Firkovich collection in St. Petersburg and has announced its publication.]

This text poses some very difficult literary problems for us. The difficulties are complicated by the nature of the extant documents and multifarious because they impinge upon a slew of biographical and bibliographical questions. A number of scholars have touched upon the problem,¹⁸⁰ but it requires a thorough investigation; specifically, it demands a careful examination of the manuscripts to be listed presently. Here we have to confine ourselves to a succinct exposition of the present state of the problem.

The peculiar Hebrew treatise, called the Book of the Creation (ס' יצירה), itself presents a problem for the history of literature. It is preserved in two major recensions, both, however, replete with additions and alterations. The book has been printed about one hundred times, as it has been included in books intended for meditation and the edification of the mystics, |395| but as of now there exists no critical edition that makes use of the oldest sources, about which we shall speak. [Though a full critical edition remains a desideratum, scholars as a rule now rely upon Ithamar Gruenwald 1971 ].

The Book of the Creation adopts the theories of the Neo-Pythagoreans who take the numbers and letters as principle of all beings and combines them with monotheism. That is the basic idea of the book, presented in an imaginative manner that may be called mystical. But the book does not expound a “speculative” Kabbala in the modern sense of the word, nor is there a reason to date it to the Talmudic era.¹⁸¹ [The recently published study of Yehuda Liebes 2000 rejects the Pythagorean stamp that Steinschneider and others have placed on the book; moreover, Liebes moves the date of the Book of Creation backwards to the pre-Talmudic era. For a critique of Liebes on both points, see Y. Tzvi Langermann 2002.]

The book differs from the Talmudic writings, not only with respect to subject matter, but in methodological approach as well. May it be of foreign origin? There is no sure sign of the tract’s existence earlier than the end of the ninth century. Some tenth-century Jewish scholars translated and commented upon it, principally Saadia (§258), though he was not the first to do so. They looked upon it as a source of the ancient philosophy of the Hebrews. According to them, however, it could not be understood without a thorough study of the foreign sciences, and, therefore, they strove to harmonize it with their own (contemporary) philosophical views. These commentaries serve for us as important sources for the text of the book itself, but, at the same time, they are the oldest sources for the history of the foreign sciences among the Jews. It is therefore pertinent to identify the authors of these works. We possess the commentary of Saadia in the Arabic original and in Hebrew translation, but apart from that there are only translations that are attributed to various authors,

among whom Isaac is the oldest. This is reason for us to collect here all the information we have on these writings, but, even before that, on the various manuscripts.

As is known, the edition of Anonymous 1562 has two textual recensions in six chapters which, however, are not congruent. The paragraphs of the second recension are not numbered; hence, I quote the paragraph numbers according to the first one.

[An up-to-date account of all the manuscripts connected with Dūnash’s commentary, covering all those discussed below and others as well, may be found in Vajda 2002 , 24–27. Vajda accepts Steinschneider’s analysis in its general contours, but proposes on p. 27 some modifications with regard to the groupings of the manuscripts.]

§228. The manuscripts must be grouped as follows:

• (a) Berlin, SPK Or. oct. 243/4 (formerly Luzzatto no. 1), fols. 55–88, attributed, in the beginning and the end, to “Dūnash (דונש) b. Tamim, called IsḤāq al-Isrā’īlī,” translated according to the wish (בפיוס) of Samuel Franco b. Yequtiel by NaḤum¹⁸² [ha-Ma‘aravi, around 1240?],¹⁸³ who introduces the book with a short poem that was published, along with the notes of Luzzatto, by Kaufmann 1885a. The manuscript is defective; the foreword may have already been missing in the original used by NaḤum, but between 60 (where the custos has been cut away, perhaps intentionally) and 61 several leaves are missing, precisely those that contain the quotations from his [Dūnash’s] own mathematical writings. Two |396| passages from this manuscript were published by Dukes.¹⁸⁴ Paris, BN héb 1048/2 , fols. 95–107, beginning missing, has the same commentary.¹⁸⁵ According to Munk¹⁸⁶ the author is perhaps Jacob b. Nissim. However, it is only a different translation of the same work.¹⁸⁷ [According to Vajda, NaḤum ha-Ma‘aravi is the author.]

• (b) Munich, BS Cod. hebr. 92/22 (some lacunae, f. 103–104^b) and Parma, BP Cod. Parm. 3018/3 , purportedly by Jacob b. Nissim (of Qayrawan, end of tenth century), translated into Hebrew by Moses b. Joseph b. Moses [of Lucerne] etc. (for more information on his dates, see under Saadia).¹⁸⁸ A copy from Munich by S. Werbluner, now in the Breslau seminary,¹⁸⁹ was in turn copied by S. Sachs. Some extracts from the Munich manuscript were published, without indication of the passages and on the basis of “excerpts done by Landauer” by J. Fürst,¹⁹⁰ and on the basis of other copies, by Jellinek and Dukes.¹⁹¹ We arrange these quotations according to the Munich ms.¹⁹² [The “arrangement” of which Steinschneider speaks here is found in note 192, which discusses the order of passages in Munich BS Heb. 92. Fenton provides a critical edition of this translation in 2002 , 214–48.]

• (c) Paris, BN héb 1048/2 , fols. 65 ff., purportedly by Abū Sahl Ibn Tāmīm, without foreword, and ending defective. Halberstam^192b has a copy of it. According to Munk 1851, 46, it is a different translation of b, composed by Dūnash (955–56). Dukes has published two passages thereof, without giving exact references,¹⁹³ and without realizing that only one page belongs to the commentary.¹⁹⁴

  The author’s quotations from his own works, discussed by Munk,¹⁹⁵ are of some importance for [the history of] Arabic literature; |397| for example, the treatise on computation, called Gobar, is relevant for the history of the Arabic numerals.¹⁹⁶ But the most important quotation, decisive for the problem of the actual, or first, author of these explications [of Sefer Yeẓira], seems to be missing from the Paris manuscripts; otherwise it would not have escaped Munk’s attention. I have in mind the repeated citation of his work on urine (§479). Munk (50) arrived at an inadmissible result in connection with a hypothesis concerning the Munich manuscript which has proven not to be true.

• (d) I own a copy of Schorr’s copy of a ms. in Odessa, now Oxford, Bodl. Ms. Reggio 51 . [Steinschneider’s copy is now in New York, JTS ms. 1912.] In the beginning of the foreword the author calls himself Abū Sahl Dūnash b. Tāmīm. It is a shorter redaction and probably the translation of an Arabic compendium of an unknown scholar. Two of the various dates given (I, 5: 431 of the Arabs, 4800 after the Creation, 6547 of the Romans [Christians]), namely the first and second (= 1040 A.D. which could be the date of composition of the Arabic compendium), are correct. One passage from it (V, 1) was published twice.¹⁹⁷

• (e) Paris, BN héb 680/10 and Paris, BN héb 763/2 ¹⁹⁸ contain a compendium that may be only slightly different from the one mentioned before; its date is changed to 4852 (1092)¹⁹⁹; such changes, however, often are merely due to the copyist; most probably it is not the date of the translation. So far, no study has been made of the relationship of this compendium to the one presented under d).²⁰⁰

Manuscripts of the Commentary on Sefer Yeẓirah, grouped according to Steinschneider, as modified by Vajda-Fenton:

• (a) Trans. NaḤum ha-Ma‘aravi

  Berlin, Staatsbibliothek (Preussischer Kulturbesitz) Or. oct. 243 (SD Luzzatto 1) (Steinschneider 78) (IMHM F 2076), 55–87.

  Paris, Bibliothèque Nationale héb 1048 (IMHM F 31659), 65–96, 95–107.

  Rome, Biblioteca Casanatense heb. 3105 (Sacerdote 190) (IMHM F 75), 1–31.

• (b) Trans. Moses b. Joseph b. Moses [of Lucerne]

  Munich, Bayerische Staatsbibliothek Cod. hebr. 92 (IMHM F 23120, PH 2495), 99a–116a.

  Parma, Biblioteca Palatina Parm. 3018 (De Rossi 769) (IMHM F 13747), [90]a–[100]b.

• (c) Another translation

  Oxford, Bodleian Library Ms. Reggio 51 (Neubauer 2250) (IMHM F 20533), 8a–19b.

  London, Sassoon 959, 60–94.

• (d) A translation of an Arabic version

  Jerusalem, Jewish National Library 8 ⁰ 330/29 (IMHM B 277), 269a–267b.

  London, British Library Add. 15299 (Margoliouth 752/4) (IMHM F 4935), 11b–16a.

  Milan, Biblioteca Ambrosiana & 103 Sup. (Bernheimer 56) (IMHM F 14621), 28°–35°.

  Milan, Biblioteca Braidanza AD. X 52/5 (IMHM F 27757), 69°–98b.

  Paris, Bibliothèque Nationale héb 680 (Paris BN ancien fonds 222) (IMHM F 11558), 189b–201a.

  Paris, Bibliothèque Nationale héb 763 (IMHM F 12254), 7a–12a.

  Paris, Alliance Israelite Universelle H 55 A (IMHM F 3149), 83°–78b.

  Milan, Biblioteca Ambrosiana P 13 Sup. (Bernheimer 113) (IMHM F 14596), 319a–21b.

  Additional mss.

  Cambridge (Mass.), Harvard 38/7 (IMHM F 34447), 77a–89a.

  Jerusalem, Jewish National Library 28 ⁰ 553 (IMHM B 296 (553 = 28)), 18a–20b.

  London, British Library Or. 6307 (Margoliouth 734) (IMHM F 6538), 3a–106a. (19th cent. copy)

  Oxford, Bodleian Library ms. Opp. Add. Qu. 89 (Neubauer 1532) (IMHM F 16900), 37a–41a. (19th cent. copy)

  Paris, Alliance Israelite Universelle H 370 A (IMHM F 3319), 1a–43a, 101a–116b. (19th cent. copy)

  Paris, Alliance Israelite Universelle H 379 A (IMHM F 3407), 20 fols. (19th cent. copy)

  Parma, Biblioteca Palatina Parm. 2784/18 (De Rossi 1390) (IMHM F 13747), 83a–88a.

  New York, Jewish Theological Seminary of America Ms. 1908–1909 (IMHM F 11006–7), 107 fols. (19th cent. copy)

  New York, Jewish Theological Seminary of America Ms. 1912/2 (IMHM F 11010), 54a–84b, 86a–89b. (19th cent. copy)

  Prague, Jewish Museum in Prague, Ms. 47 (F 46440). 43 fols. (19th cent. copy)

  Prague, Jewish Museum in Prague, Ms. 119/1 (F 46700). 43 fols. (19th cent. copy)

  Warsaw, Zydowski Instytut Historyczny 616 (IMHM F 31050). 12 fols. (19th cent. copy)

§229. I pass over the historical problems linked to the authors who are mentioned here and there, e.g., the “Danites” – which were brought into connection with the well-known traveler Eldad [See Vajda 2002 , 39 n. 126] – and Abū Yūsuf Ḥisdai b. Isaac, doubtlessly the renowned Cordovan translator of the Dioscorides text.²⁰¹ I call attention here to one passage only, |398| one which may be of significance for the provenance of the author.²⁰² All manuscripts derive directly or indirectly from one tenth-century Arabic work which was probably translated three times (a, b, c) into Hebrew and whose Arabic compendium was perhaps translated twice (d, e) into Hebrew.

The Arabic work represented by the more complete translations contains at least parts of a commentary composed by Isaac Israeli. This follows from two passages where the author refers the reader to his book on urine,²⁰³ and perhaps also to his interpretation of Genesis I.²⁰⁴ Abraham Ibn Ḥasdai and Jedaiah ha-Penini refer to Israeli’s commentary as something that is extant and well-known. Gershon ben Solomon quotes one passage from it directly, giving the title, and another, without the title (based on NaḤum’s translation;²⁰⁵ this furnishes additional evidence that this translator flourished no later than the thirteenth century).

However, the Arabic work, at least as it appears in translation, is not simply the commentary of Isaac. In the preface the author tells us that “our Sheikh Isaac b. Solomon the physician” (cf. infra) corresponded with Saadia Gaon, before he – Saadia – betook himself to Babylon (928), at a time when the author was only 20 years of age. The same Isaac is quoted in the commentary itself,²⁰⁶ and this most probably refers to Israeli himself, rather than to a grandfather of the same name (unknown), or, even less likely, to a grandson of his, since Isaac remained unmarried. He also is not expressly named as a teacher of the author, but rather as “one of the scholars among our sheikhs.”²⁰⁷ This author, born before 908, cannot be Jacob b. Nissim, who corresponded with Shrira Gaon in 988, as Munk |399| rightly remarks.²⁰⁸ One would consequently have to assume that Dūnash b. Tamim prepared a redaction based on the commentary of Isaac, but including literal citations from Isaac in some places and inserting the exact words of the first author in some places, so that it is difficult now to decide to whom all of the citations are to be traced, if no other criterion can be found. This solution is not wholly satisfactory, although there is no satisfactory reason why it could not be so. A redaction of this sort is only a preliminary and expedient solution. [Vajda 2002 , 34, suggests that Dūnash’s work did not borrow from a completed commentary by Israeli, but rather utilized notes taken when the young philosopher was under Israeli’s supervision.]

The author, or redactor, had earlier composed an extensive commentary to this work.²⁰⁹ It later became his exclusive occupation, after he saw the commentary by Saadia Gaon, which he found to be insufficient from a scholarly point of view. He in fact attacks him several times openly, albeit with the respect due to such a venerable scholar (Foreword).²¹⁰

§230. The outward form of our commentary is in fact identical with that of Saadia: The text of the book is in Hebrew, the paraphrase (not throughout) and interpretation in Arabic.²¹¹ These formal components of the book can most easily be isolated in the translation of Moses (b), but less so in NaḤum’s (a), which starts the commentary with פירש. Translation and commentary blend together in the compendium (d). Like Saadia (his practice in the Biblical commentaries as well), our author joins together a number of passages belonging to a paragraph that he leaves unnumbered. In a very few cases he splits up a unit of text that appears as a paragraph in our editions, e.g., Munich, BS Cod. hebr. 92/22 , f. 108^b [ = Fenton, 233] (II: 5), where וזה סימן in the editions seems to be an error. Instead of צופה in the editions and in Berlin, SPK Or. oct. 243/4 , 70, the compendium has צורף, p. 34.

The continuation was doubtlessly indicated by Arabic יתלו “follows”, for which the compendium usually exhibits the formula תלה בזה ואמר, stemming from the same root, but not customarily used in this context.^211b

Moses b. Joseph calls the paragraphs of a given chapter halakhah (“rule”, as does Saadia). Paragraph 2 of chapter 1²¹² corresponds to I, 4, 5; in the compendium (St. 11) it is called “second discourse” (מאמר), but the next paragraphs bear the title “chapter” (פרק), as chapter 3 (f. 19 = |400| 6), 4 (25 = I, 11), 5 (26 = I, 11 bis, f. 56), 6 (28 = II, 2). Page 30 (II, 4) has: : 'תלה בהולדה [בהלכה] הג (for which a 66 וסמך אל זה הענין ההלכה השלישית). After that the numbering ceases, but under VI, 4, a, b, and d refer to the third “rule” in the beginning of the book.²¹³ Here and there the interpretation repeats part of the text to be explained, probably on the basis of old manuscripts, which were abbreviated by more recent copyists.

§231. A linguistic comparison of the translation calls for a careful examination of all manuscripts – most of them unfortunately incomplete – including a comparison with the quotations in Judah b. Barzillai (Commentary on Sefer Yeẓirah, Berlin 1885), who neither names an author nor indicates whether he makes use of a translation available to him which, in any case, is different from a–c.^213b Perhaps I shall give some results based on the available excerpts in an endnote. [No such endnote was supplied in this book.] I shall confine myself here mainly to the Arabic words that the translators have taken from the text with or without a Hebrew translation. It is worth remarking that the unknown author himself, when giving a linguistic description of the letters of the two alphabets, Hebrew and Arabic (I, 1 and II, 2) had occasion to use some technical terms or to present Arabic words as examples for the pronunciation of certain letters. He intended to compose a book on how the two languages are related.²¹⁴

In our presentation we follow the order of the book, designating the translations of NaḤum, Moses, and the compendium by a, b, and d, respectively. Of the translation c we know next to nothing, other than this: ראבגלא באסח is translated here by רפעה רפסמ ²¹⁵; passage I, 1 is lacking in a; in b (Ben Asher, Aaron b. Moses and Jacob ben Nissim 1846, 77) we find only ודוה ינב ןובשח, in d 10 קבאה ןובשחב עדונה ודוה (?רפסמ) חתפר ארקנה.

Preface b 99 [ = Fenton, 214] (Fürst 1845, 563) חכמת החבור כלומר מוציקי; also I, 11: a (65, last line) מוסיקא; b 105 [ = Fenton, 226]) מוציקי, d 29 חכמת הניגון

I, 1, b 100b [ = Fenton, 216] of the Arabic words (Ben Asher, Aaron b. Moses and Jacob ben Nissim 1846, 72, 73) קציב and גבהה are translated; a 58, 59); Ben Asher, Aaron b. Moses and Jacob ben Nissim 1846, 8 omits the examples; d 6) omits גבהה and does not give a translation.

I, 1, b ובעל הלשון אומר בספר הראשון מס' הלשון הנקרא מנטקיאת ובל' יון נקרא רטוריא indubitably for האירוגאטק, Categories; a 59 ןויגהה 'סב רמא ןויגהה 'ס לעבו; d ובעל ההגיון בס' הא' מחכמת המבטא; cf. 16 (I, 5) וזה מן ההגיון בס' הא' בשער הא' בשער הקנין (a is corrupt here; b I cannot investigate); cf. III, 1, b 109 [ = Fenton 234]) (!) ןאירוטיר ןושלה ילעב (see epilogue §232); a 71 ןויגהה ילעב; d 35 רמאנ ןויגהב. Worth noticing is ןושל for קטנמ.

I, 3, a 60 קפיז (a measure); b 75 כורי; d 9 משא – a 61 סבת הרעש הנקרא זלזלה.²¹⁶

II, 3 (Ibn Gabirol 1858, III) העציות והעץ נקרא בלשון ערב הפה; Berlin Or. 243 (fol 65) מוצא קפא at any rate one word, perhaps מצאפרה from צפר sibilavit [(= he produced a sibilant)]; |401| then אסליה and חלקיה (Berlin Or. 243: דולוקיה!) gutturalia; Munich BS 92 105 [ = Fenton, 227]) חלוקות (Kaufmann 1884c, 172).

V, 1, a 81 and d f. 46) the star names פרקדין and סהיל

VI, 1, a 83 and b f. 113b [ = Fenton, 244] אפריגיון (not in d f. 49)

Ibidem d 50 מרכז center (current usage already then?), a 85 and b 114 [ = Fenton, 244]) מוצק.²¹⁷

VI, 3 a 86 and d 54 מאסריקא מאסריקי; b f. 115 [ = Fenton, 247] גונבים!²¹⁸

Among Hebrew words and forms we mention in particular: וחיצנהל (b Ben Asher, Aaron b. Moses and Jacob ben Nissim 1846, 78, one line before the last), בכוכמ (87, line 4, so should be read for סכובמ, 80 line 5 from bottom, ditto Arabic),²¹⁹ – a 63 הלפטה ןימ תחת (Ben Asher, Aaron b. Moses and Jacob ben Nissim 1846, 42 הליפטה; Kaufmann 1884c, 141); b 104b [ = Fenton, 224] ךמסומה סרוגיטאק; d 27 תופרטצהה.

Ibidem a מדותיו; b ספוריו; c תואריו.

II, 2, a 64, c Ibn Gabirol 1858, iii, line 10: מתלכד, should be read מתלבד from לבד.

II, 2, b, i.e., Munich 92 105 [ = Fenton, 227]) (Ibn Gabirol 1858, xiii) גושמניות; a (ibid. III) גשמיות according to Berlin heb. 92, 64; d 28 גופניות.

II, 4, a for “to multiply” תשתכה) הכה, Arabic צ'רב), “product” ממון (Arabic מאל); b 106 [ = Fenton, 229] צרוף and קרן הממון, d ערך and היוצא.

III, 5, a 77 ומזה כלו אצלי נכנס על בעל ס' יצירה; b 111 = [Fenton, 238] כנוס (Jellinek 1852 I, 14, Arabic מדכ'ל or מדכ'כ'ל “interpolated”).

VI, 1, a 84 הנפש המבטאית; d 49 המדברת.

NaḤum writes רוחאני with א consistently, as does the translator of the Microcosm (§238).^219b

§232. We conclude this article with a word on the peculiar published epilogue which is extant only in the translation of Moses.²²⁰ There the author names three very prolific authors – known to him only through their works – who are very prolix: among the Greeks there is Galen, who was a Jew by the name of Gamaliel, living at the time of the Second Temple, since he mentions in his book on nutrition (ספר במאכלות) unleavened bread that one should eat for 7 days in such and such a manner.²²¹ The author had seen a work on medicine, translated from Hebrew into Latin, with the title (בהקדמתו) “Book of Gamaliel, the Nasi, called Galenus among the Greeks” (הידוע = Arabic אלמערוף). There is no other writer, our author says, as prolix as Galen, so much so that his prolixity often vitiates the effect of what he wanted to say. This reminds us of Isaac Israeli, who also speaks of Galen’s prolixity, “Galen, the master physician.”²²²

Galen’s counterpart among the Arabs is עמר בן גיהאט (sic in the manuscript, printed as גיהאני); perhaps ‘Amr al-JāḤiẓ.²²³ The third,²²⁴ Saadia Gaon, is prolix, yet he adorns our book (Yeẓira – or his own books?) with philosophy and with rhetorical arguments (?טענות רטוריאה), etc.

Finally we add to the |402| authors quoted in our work in Munk 1851, 54 Euclid (a f. 69, d f. 33; b f. 108 b): Euclid “the philosopher”.

§233. Judah ha-Levi b. Samuel,²²⁵ in Arabic, Abū l-Ḥasan²²⁶ from Toledo, lived in Cordova, but a somewhat mystical national sentiment²²⁷ drove him, in his 50th year (1140–50), to emigrate and betake himself to his ancestral fatherland. We can follow his itinerary up to his departure from Egypt. [The details of ha-Levi’s departures for the Holy Land were finally clarified in an exhaustive study by Moshe Gil and Ezra Fleischer 2001 ; Judah ha-Levi 1663 . In 1129–30, ha-Levi dramatically announced his plans, and he did leave Spain for North Africa, the first leg of his planned journey, along with Abraham Ibn Ezra. However, while there, ha-Levi received word that Ḥalfon ben Netanel, the Egyptian entrepreneur who was to be his host, was in difficulty and could not receive him. Ha-Levi returned to Spain; only some 10 years later was he able to carry out his plans. Yahalom 2009 and Scheindlin 2008 have recently studied the pilgrimage; while disagreeing on several details, they both agree that ha-Levi reached the Holy Land.] Legend, which also has made him the father-in-law of Abraham Ibn Ezra,²²⁸ tells us that the horse of an Arab crushed him when he recited his famous hymn, the Zionide [ציון הלא תשאלי], at the gates of Jerusalem.²²⁹

Judah ha-Levi was a physician, but this art did not satisfy his mind, nor did it fulfill the demands of his imagination.²³⁰ He is justly considered the most excellent Hebrew poet of the Middle Ages. His poems were admired, especially his hymns, of which more than 300 are preserved in almost all Jewish prayer books from Eastern and Northern Europe to India and China,²³¹ as well as in one of the two collections (diwans) of his poems. In more recent times this Jewish poet has, through Heine, nearly become a popular poet. Jewish theologians appreciated him also for a book that is relevant to the present context, although it is not strictly philosophical.

The Arabic title is actually כתאב אלחג'ה ואלדליל פי אלדין אלד'ליל The Book of Refutation and Proof of the Humiliated Faith. It is an apology for the revelation and the Jewish tradition, directed against Islam, Christianity, philosophy and the sect of the Karaites whose theology follows the methods of kalam. The author exploits the tradition concerning the conversion of the Khazars to the Jewish religion.

[Khazarite studies have been a lively field in the last few decades, attracting scholars and enthusiasts alike. For a recent and revised overview in English see Brook 2004. Cf. Golb 1988, 1982 , and Golden 1980 , an expanded version of the author’s Columbia doctorate. The International Center of Khazar Studies in Kharkiv, Ukraine, publishes a journal, Khazarskiy al’manax . On the web see http://www.khazaria.com/ .]

He introduces into his book the king of the Khazars who invites a Muslim, a Christian and a Jewish scholar (called Isaac Sangari in a report on the conversion).²³² |403| In the course of his replies to the king’s questions and objections, the Jew expounds a theory which, according to Geiger,²³³ is closer to Christianity than to Islam. The book is not known under its original title, but it became famous under the title of a Hebrew translation.

Only one manuscript of the Arabic original is known: Oxford, Bodl. Ms. Poc. 284 , of which there exists a defective and incorrect copy by Goldberg, ms. Quatremère, now Munich, BS Arab. 936 [Aumer Cat. 421].²³⁴ [The critical edition of D. H. Baneth and H. Ben-Shammai 1977 makes use of many fragments of the original. Nonetheless, the Oxford manuscript listed by Steinschneider remains the only complete copy; it too suffers from a few minor lacunae. There are in addition some indirect textual witnesses. The Ibn Tibbon Hebrew translation is still necessary in order to establish the correct text. The manuscripts and other sources are discussed in the “Introduction” to the edition of Baneth–Ben-Shammai.] Goldberg has published some long passages²³⁵; D. Kaufmann (1877a, 119) exhibits a greater number of passages from this copy, partly collated with the Bodleian manuscript. Hartwig Hirschfeld published a complete edition of the text along with a Hebrew translation (Leipzig 1887).^235b In his work on the three religions Sa‛d b. ManṢūr [Ibn Kammūna] adopted large sections from it; others are to be found in an appendix ascribed to him on the differences between the views of Rabbanites and the Karaite views. [Ibn Kammuna’s use of the work is discussed by M. Perlmann in 1971; Perlmann also notes parallel passages in his edition of the Arabic text 1967. Midrash ha-Nagid , attributed to the grandson of Maimonides, is another Judaeo-Arabic writing to make use of ha-Levi’s work; see Langermann 1996d , 293–96.] ²³⁶ The fifth treatise of this work presents a concise system of scholastic philosophy of the Arabs, the so-called kalam, which is relevant for the history of philosophy.²³⁷

§234. This work was translated twice. Judah Ibn Tibbon’s translation under the title ס' הכוזרי Kuzari (Cosari, Cosri, Cusri, Kusari, etc.; in Hirschfeld, Chazari) displays the date Lunel, 1167, in a number of manuscripts, and in the editions, beginning with that of G. Soncino (Judah ha-Levi 1506).²³⁸ Some are accompanied by commentaries.²³⁹ The Hamburg edition (Judah ha-Levi 1838, printed in Hannover, furnished with variant readings) lacks the paragraph numbering necessary for tracing quotations; I was unable to consult the Warsaw edition (Judah ha-Levi 1866). We have a Latin translation by Buxtorf (Judah ha-Levi 1660), a Spanish translation of Jakob Abendana (Judah ha-Levi 1663),²⁴⁰ a German translation with text and notes by H. Jolowicz and D. Cassel and an introduction by the latter (Judah ha-Levi 1853), whose revised version in the edition of 1869 is greatly abbreviated, and finally a German translation from the Arabic (with an introduction on the Khazars) by Hartwig Hirschfeld (Judah ha-Levi 1887). [There is a complete English translation from the Arabic by Hirschfeld (Judah ha-Levi 1905 ) and a partial one by I. Heinemann (Judah ha-Levi 1947 ), neither entirely satisfactory. A new English translation, begun by the late Lawrence Berman and completely revised by Barry Kogan is forthcoming. Charles Touati published a French translation (Judah ha-Levi 1994 ). Three modern Hebrew editions are available: those of Yehudah Even Shmu’el (Judah ha-Levi 1972 ), accompanied by a very extensive discussion of textual history (including printings and translations), copious notes, and indices; of Yosef Qafih (Judah ha-Levi 1996 ), which also has a facing re-edition of the Judaeo-Arabic; ; and of Rabbi I. Shailat (Judah ha-Levi 2010 ).]

Around the year 1420 a Provençal author, Solomon b. Menachem, vulgo Prat (Comprat?) Maimon,²⁴¹ read the book with some young students. Three of them went on to compose commentaries that resemble each other not only in their interpretations but also in their very frequent, occasionally important, quotations from philosophical literature. The commentary by Jacob b. |404| Ḥayyim, called Vidal Farissol²⁴² or קמראט (Comprat), composed in his 17th year (1422) under the title בית יעקב, exists in manuscripts Berlin, SPK Or. Qu. 653 (formerly Kayserling); London, Mon. 268 , and New York, JTS Ms. 2287 (formerly Heidenheim Cat., Rodelheim 1833 39, no. 4).

The commentary of Solomon b. Judah, called Solomon Vivas (or Vives) from Lünel, written in his thirteenth year (1424) under the title חשק שלמה, is extant in manuscripts Oxford, Bodl. Ms. Opp. Add. Qu. 114b ²⁴³ (formerly Mortara and Asher 16), Turin, BN A.VI.30 (some errors in [the catalogue of] B. Peyron 204, n. 193), with some extracts also in Cambridge, UL Add. 539/22 . [The Turin ms. was badly burned in 1904. The work was edited recently by Dov Schwartz in Solomon ben Judah 2007 .]

The commentary of Natanel Kaspi, called Bonsenior Macif of Argentierre,²⁴⁴ erroneously called עדות לישראל (that is the title of a work of the master which is quoted therein) and composed around the end of 1424, exists in manuscripts Oxford, Bodl. Ms. Mich. Add. 11 ²⁴⁵ (formerly Bislichis 18, a copy of which is owned by S. Sachs); Paris, BN héb 677 , an autograph²⁴⁶; Paris, BN héb 678/1 ; London, Mon. 269 (formerly Halberstam 1 and Luzz. 22); and Hamburg, SUB Ms. Levy 144 (formerly Asher (1868 cat. 17); Berlin, SPK Or. Qu. 822 , ‹Parma, BP Cod. Parm. 2255/4 › [Budapest, MTA Ms. Kaufmann A 270 ; Jerusalem, Benayahu V 10 ; Mantua, CI 19 ; Milan, BA X 122 Sup. ; Moscow, RSL Günz. 263/18 ; Moscow, RSL Günz. 1443 .]

Steinschneider 1879b, 111–115, has an analysis of these three commentaries and supplies references for almost all quotations. As to the variant readings, see §235. [These commentaries have been studied recently by Dov Schwartz 2000 ; see there for bibliography.]

Prat Maimon’s Commentary

Berlin, Staatsbibliothek (Preussischer Kulturbesitz) Or. Qu. 653 (Steinschneider 124) (IMHM F 1777), 157 fols.

London, Montefiore 268 (Halb. 214) (IMHM F 5232), 179 fols.

New York, Jewish Theological Seminary Ms. 2287 (Acc. 95) (IMHM F 28540).

Solomon b. Judah’s Commentary

Cambridge, University Library Add. 539 (SCR 268) (IMHM F 15877), 168b–71b.

Oxford, Bodleian Library Ms. Opp. Add. Qu. 114 (Neubauer 2383) (IMHM F 21663), 181 fols.

Netanel Kaspi’s Commentary

Berlin, Staatsbibliothek (Preussischer Kulturbesitz) Or. Qu. 822 (Steinschneider 203) (IMHM F 1750), 1a–216a.

Budapest, Magyar tudományos akadémia Kaufmann A. 270 (IMHM F 15124), 160 fols.

Jerusalem, Hamburg MS. Levy 144 (Hamburg Acc. 1906/11233) (Levy 157) (IMHM F 1584), 1a–172a.

Jerusalem, Benayahu V 10 V 10 (IMHM F 44467), 7 fols.

London, Montefiore 269 (Halb. 1) (IMHM F 5233), 85 fols.

Mantua, Comunita Israelitica Ms. ebr. 81 (IMHM F 799), 220 fols.

Milan, Biblioteca Ambrosiana X 122 Sup. (Luzzatto 37) (Bernheimer) (IMHM F 12344), 209 fols.

Moscow, Russian State Library Ms. Guenzburg 263 (IMHM F 457198), 288b–92a.

Moscow, Russian State Library Ms. Guenzburg 1443 (IMHM F 48505), 75 fols.

Oxford, Bodleian Ms. Mich. Add. 11 (Neubauer 1229) (IMHM F 22043), 249 fols.

Paris, Bibliothèque Nationale héb 677 (Oratoire 59) (IMHM F 11555), 1–204.

Paris, Bibliothèque Nationale héb 678/1 (Paris BN ancien fonds 214) (IMHM F 11556), 1b–176a.

Parma, Biblioteca Palatina Parm. 2255 (De Rossi 395) (IMHM F 13419), 17a–172b.

One or the other of these commentaries was known to Judah Moscato, whose extensive commentary קול יהודה, begun in 1573 and printed in Venice after his death (between 1590 and 1594),²⁴⁷ abounds with quotations of Jewish and other authors.²⁴⁸ These appear to be taken from one of these commentaries, and so also, perhaps, some of the variant readings.²⁴⁹

All of the commentaries printed together with the text, namely those by Isaac Satanow (Judah ha-Levi 1796), Israel Zamosc (Judah ha-Levi 1796)²⁵⁰ – but not the one by G. Brecher (Judah ha-Levi 1837) – have profited from this learned commentary.

Judah Ibn Tibbon did not write a foreword to his translation, which is generally very reliable. It would require a special monograph to fully characterize this translation.²⁵¹ The notes of Jolowicz and Cassel have frequently pointed out and explained its Arabisms. Under the pseudonym M. Sider, I published some extensive articles on the first installments of that edition in Steinschneider 1841, 1842, discussing in particular the Arabisms (1842, 225–34). Here I only point out the Arabic words which can be found among the Hebrew words in the “Index philologicus”, without any differentiation between the languages (Judah ha-Levi 1853, 438): לוחא, הידאשנא תאסקתסא, םילקא רישדרא (read for רישדשא),²⁵² ריתא (read for רותא p. 250), אינבב, היכרב, םאסרב, רהזוג, קינארג, ילויה, סורדנכ, זכרמ, םייאשמ (hebraized, on 406, missing from the index), שטרנג,עמש (p. 424).

§235. Judah Ibn Cardinal (קרדינאל, קרדנאל) ben Isaac translated the same book – we do not know under which title – from the Arabic for Joseph b. Baruch, perhaps that scholar who went to Palestine in 1211.²⁵³ In a very brief foreword the translator recounts |405| that the first treatises of his translation were originally brought by Joseph to England. However, they were not returned by the persons who had taken possession of them. Thus, in order to have the complete book, he had to translate it again. He assures us that he has not changed anything substantial in his translation; even if he added or omitted words, he strove never to change the intent. [The citations from the Hebrew Kuzari in Ḥesheq Shlomo (Solomon ben Judah 2007 ) do not always match Ibn Tibbon (whose text is conveniently reprinted by Schwartz, but Schwartz makes no comments about these divergences); Ḥesheq Shlomo may have made use, at least in some places, of Cardinal; or the variants may have already infiltrated mss of Ibn Tibbon, as Steinschneider suggests.]

Apart from this foreword, as well as a few lines from the beginning of the book that are in some manuscripts of the translation just mentioned,²⁵⁴ Munich, BS Cod. hebr. 47/5 contains a fragment (IV, 25) which is doubtlessly the same which Moscato found in the margin of some copies of the printed translation by Tibbon. This fragment is printed on the basis of a copy by Werbluner (a man of little scholarship) in Cassel’s edition (Judah ha-Levi 1853), 344–61, below Tibbon’s translation. [The fragments of Cardinal’s translation, culled from various sources, are displayed in the notes to Y. Even-Shemuel’s translation.]

Some variant readings in the edition of Ibn Tibbon probably ought to be traced back to this translation, perhaps by way of the three commentaries (§234), which know at least the printed portion of Cardinal’s translation (cf. Steinschneider 1879b, 115). In order to characterize this translation we point out the following words, in the order of their appearance in Cassel’s edition: וצבתשי (p. 346), םימלג (357 and elsewhere = םימשג, cf. תומלג 357, ימלגה, 361; on the other hand, םימרג 354), המצע and המצוע (346 and elsewhere = חכ, Arabic quwwa), םש ול בצוח, (347) הכירע (ibid. = ךרע) לידבמה רתס,(352 ff., םורק in Tibbon), ןוכיתה קודקד [המודה] האורה (353, זכרמ Tibbon), תוברקה (ibid.), העבט (355 ff., תויבג more commonly עבט) םישפנ (356), שילפתהל (l. שלפתהל 357) and תשלפתמ (358),²⁵⁵ (360 “living bodies,” the word has been added)²⁵⁶ רתום (“connected,” 360 twice). [For the impact of the Kuzari on Jewish intellectual history until 1900 see Shear 2008 .]

§236. Judah b. Nissim Ibn Malka, or Melka (מלכה),²⁵⁷ probably from Spain or North Africa (ca. 1365),²⁵⁸ [The consensus now is that Ibn Malka’s floruit must be moved up to the middle of the thirteenth century; see M. Idel 1990 . A book-length study of Judah b. Nissim is available in Georges Vajda 1954 .] and inspired by Neoplatonic teachings,²⁵⁹ composed a work in three parts: (1) אנס אלגריב “Companion to the Stranger,” an introduction to the Book of Creation (יצירה, see §227) which consists of two dialogues, the one between the author²⁶⁰ and his own soul, the other between a |406| student (טאלב) and his teacher; and ten chapters on the human attainment of the perfect science; (2) תפסיר ספר יצירה, a commentary on the Book of Creation, according to philosophical principles; (3) תפסיר פרקי ר' אליעזר, commentary on the Chapters of R. Eliezer, up to chapter 52 (the author had no more at his disposal), completed in 1365. The Arabic original is extant in Paris, BN Ms. héb. 764 and Oxford, Bodl. Ms. Or. 661c; Oxford, Bodl. Ms. Or. 661a; Oxford, Bodl. Ms. Or. 661b [all three sections of the work]; Oxford, Bodl. Ms. Opp. Add. Qu. 45 (Neub. 1536; bought by Zunz who had received it from the heirs of N. Krochmal; the end is defective); the original is probably quoted by Samuel Motot (ca. 1370).²⁶¹

Oxford, Bodl. Opp. Add. Qu. 45 also contains a Hebrew translation²⁶² of the commentary to the Book of Creation. [This was published by Georges Vajda 1974 . According to Vajda, one main objective of the abridgment was to remove passages that were deemed theologically obnoxious.] Perhaps Moses Botarel (1409) still knew of it.²⁶³

[The kabbalist Isaac of Acre prepared a Hebrew translation of the first part of the commentary on the “Chapters of R. Eliezer”; this text, which is accompanied by glosses of the translator, is found in two manuscripts: New York, JTS Rab. 1723, fols. 1a–6a and New York, NYPL Jewish Items XXP/32 (formerly Sassoon 919). This text was edited by Paul Fenton 1993 . A different translation of the commentary is found in Moscow, RGB Günz., fols. 258, 286a–298b; it is being studied now by Y. Tzvi Langermann, towards eventual publication. This text is in the same hand and most likely accomplished by the same individual who wrote the shortened translation of the commentary to Sefer Yeẓirah published by Vajda (concerning which see above). Indeed, the Moscow manuscript is not a true codex but rather a collection of fragments from diverse sources that have been bound together. It seems most likely that they simply fell out of the same manuscript that eventually found its way to the Bodleian Library.]

§237. (Joseph Ibn ‘Aqnin. [Joseph Ibn Simeon of Ceuta, see below]). Joseph b. Judah etc. b. Simeon, or Arabic: Abūl-Ḥajjāj Yūsuf b. YaḤyā b. Sham‛ūn al-Maghribī al-Fāsī,²⁶⁴ is certainly Joseph Ibn ‘Aqnin (עקנין or עכנין).²⁶⁵ He was a student of Maimonides, later on a very significant physician in Aleppo, where he died in 1226, and a friend of the vizier al-Qifִtī, who dedicated an article (published by Munk) to him and quoted in part by Ibn abī UṢaybi‛a (II, 213).²⁶⁶ [ S. Munk 1842 argued that Joseph b. Judah Ibn Simeon, whom he identified with the famous pupil of Maimonides, was not the same person as the philosopher Joseph b. Judah Ibn ‘Aqnin. This led to a sharp response by Steinschneider 3 years later, and to a scholarly controversy that continued for more than a century, with Steinschneider’s view gradually losing support. The scholarly consensus is now with Munk, i.e., against identifying the two and for viewing Joseph Ibn Simeon, and not Joseph Ibn Aknin, as Maimonides’ pupil. (See Baneth 1957, 1964 ; for other views, published before Baneth’s articles, see B. Lewis 1945 and A. Halkin 1950 .) Baneth also held that Joseph Ibn Simeon was the author of the metaphysical treatise that Steinschneider describes above, as well as another work unknown to Steinschneider at the time of writing HUe, described below (Baneth 1946, 2). (Ibn ‘Aqnin is indeed the author of the ethical work טב אלנפוס that is discussed in §10.)]

We have already dealt with Ibn ‘Aqnin’s ethical work טב אלנפוס, two parts of which were translated into Hebrew (§10).

Ibn ‘Aqnin is the author of a little metaphysical treatise, actually untitled, called in the edition after its contents²⁶⁷ מאמר במחויב המציאות ואיכות סדור הדברים ממנו וחדוש העולם “Treatise on the Necessary Existent, on How Things Emanate From It, and on the Creation of the World,”²⁶⁸ which is known only through the Hebrew translation of an anonymous author (probably fourteenth century; M. Löwy²⁶⁹ surmises that it is Isaac ben Nathan). For the most part it is found together with the questions and treatises of |407| Averroes, along with a treatise of al-Ghazālī (§192), in Leiden, BR Cod. Or. 4753/4 , Leipzig, UBL B. H. 14 .²⁷⁰ Munich, BS Cod. hebr. 36/22 ; Paris, BN héb 959/8 ‹cf. Isaac b. יושט “Just ou Jost” (!) 1225, Bacher 1890, 128 ms. Epstein [now Jerusalem, JNUL Yah. Ms. Heb. 1 cf. Weiser 1992 , 52]›; Moscow, RSL Günz. 305/12 .

Moritz Löwy began an edition of these translations on the basis of the Leiden and Leipzig manuscripts, along with a German translation and philological and philosophical notes.²⁷¹ [The treatise was edited in its entirety and translated into English in Judah Magnes’ doctoral thesis 1904; for criticisms and emendations of this edition, see Baneth 1957 .]

In this treatise, probably composed in 1187 in Aleppo, the author addresses his teacher, without doubt Maimonides, whom he had probably queried beforehand, concerning three theses, listed at the beginning of the treatise, and considered from a rational and (theologically) positive point of view.²⁷² Not very satisfied with the response of his teacher, Yosef attempts at a solution, which he then offers to him.

[Baneth argued that it is most unlikely that Maimonides was the teacher whose responses were deemed by Ibn Simeon to be unsatisfactory, and so the work, according to Baneth, was written before Joseph became Maimonides’ pupil in the early 1180s, i.e., while he was still in the Maghreb.]

According to Munk 1842, 47, the translation is done in “a very unclear style.” As M. Löwy very aptly remarks, its character is that of the translations of the second period. These are executed in closer dependence upon the Arabic texts. The transnational themes, so to speak, seemed to justify a literal rendering, too slavishly close to the text, but, on the other hand, too little governed by the laws and usage of the language. The translation contains less Arabic words, because more and more Hebrew synonyms had been introduced and missing elements were continuously replaced. In addition, these translations were meant for readers for whom Arabic had become so foreign that the Arabic word was completely useless. Moreover, Arabic was hardly a living language for the translators; they had learnt it by reading, like an extinct language.

There exists a small number of philosophical works, translated in a similar fashion by unknown translators. Most likely it is the copyists that have dealt with their writings negligently, for it is not be expected that the translator would have omitted his own name, against customary practice and his own vanity, or let us say, ambition. Stylistic similarity, even the usage of certain words, e.g., האותות,²⁷³ and of idiomatic phrases are not enough to attribute all such translations to one and the same person, as long as no particular circumstances support this hypothesis. For our treatise one could adduce as an argument its outward connection to the treatise of al-Ghazālī, translated by Isaac Nathan.

[2. The Silencing Epistle, Concerning the Resurrection of the Dead ( Risālat al-iskāt fī Ḥashr al-amwāt or Risālat al-iskāt fī tabyīn Ḥashr al-amwāt ). This monograph was written in order “to silence” criticism of Maimonides’ purported position on the question of the resurrection of the dead, particularly that circulated by Maimonides’ arch-rival, Shmu’el ben ‘Ali ?, Gaon of Baghdad. Portions of a Judaeo-Arabic treatise defending Maimonides on this issue were published by A. Harkavy in 1897. Baneth surmised that the author was Ibn Simeon 1964, 2. The Hebrew translation, which is complete and bears the title and the name of the author, was discovered by Y. Tzvi Langermann in a codex of philosophical treatises, Moscow, RSL Schneerson Ms. Yevr. 209 . For this ms. see Langermann 1996b . The full Hebrew text, as well as the surviving portions of the original and a close analysis (including full bibliography), with an introduction in English, was published by Sarah Stroumsa 1999 . The translator is Ḥayyim ben Yehudah Ibn Bibas (or Vivas); the translation was done in Valencia in 1343.]

§238. Joseph ben Jacob (Ibn) ẓaddiq (‘ẓiddiq’?),²⁷⁴ in Arabic Abū ‛Umar,²⁷⁵ judge in Cordova (died 1149),²⁷⁶ composed at the |408| request of a pupil a small dogmatical work according to the opinions of the philosophers (the Philosopher = Aristotle; De caelo et mundo is quoted on p. 10), which he called “Microcosmos” (אלעאלם אלצגיר). Man is a microcosm and, therefore, philosophy means his – man’s – self-knowledge,²⁷⁷ by means of which he will come to understand the macrocosm and its Creator. The fourth and last section is more dogmatic. It treats of the deeds of man and their recompense, ending with the resurrection. Eisler presents an analysis of the work in the Jüdisches Centralblatt, ed. by Grünwald, 1886, 1887. This work was not studied much;²⁷⁸ Maimonides knew it only by name and did not think much of it.²⁷⁹ [Sarah Stroumsa 1990 reinforces Steinschneider’s claim that Maimonides’ attitude towards Ibn ẓaddiq was negative. For a recent examination and summary of Ibn ẓaddiq’s life and works, including a review of studies and editions of the Microcosm, see Habermann 2002, 17–51 (Introduction to his English translation of the text). Haberman, 43 n. 3, cites Gershon Cohen 1967, 139f, who argues against the patronym Jacob, noting that Steinschneider himself may have had doubts.]

The Hebrew translation, under the title העולם הקטן is extant in: Oxford, Bodl. Ms. Poc. 280 ; Oxford, Bodl. Ms. Opp. 583 ; Oxford, Bodl. Ms. Mich. 491a ; Hamburg, SUB Ms. Cod. hebr. 92/7 ; Munich, BS Cod. hebr. 65/4 ; Parma, BP Cod. Parm. 2450 ; [Rome, BC 3089/3 ; Budapest, MTA Ms. Kaufmann 589 ].

The Parma ms. does not name the translator. The acrostic Moshe at the end of the Hamburg manuscript led me (in 1847) to surmise that Moses Ibn Tibbon is the translator, but by the time I wrote the article on Ibn ẓaddiq in the Ersch and Gruber encyclopedia, I realized that there exists a treatise with the same title by the same Moses which seems to be part of a larger work.²⁸⁰ [A treatise which bears the same title is attributed to Moses Ibn Tibbon; it is found in thirteen mss., including three cited by Steinschneider in n. 280: Paris, BN héb 185/2 ; Paris, BN héb 893/4 ; and Vatican BA Cod. ebr. 292/2 .] Jellinek put this name on the title page of his edition (Ibn ẓaddiq 1854), basing himself on the faulty copy in Hamburg.²⁸¹ [An edition and German translation was published by S. Horovitz 1903; it has been reprinted in J. Haberman’s English translation 2002.] In the foreword (XVII–XXII) Jellinek displays some variant readings from the Munich manuscript. D. Kaufmann’s German translation of part III 1877a, with the presentation of the most important Hebrew passages, demonstrates just how many better readings can be gained from the manuscripts. There are various reasons for dating this translation before Moses Ibn Tibbon. David KimḤi²⁸² seems to quote it, and its style corresponds more to that of NaḤum.²⁸³ Like him, our translator uses the Arabic |409| term נוע (species) in the Hebrew plural; both write the word רוחאני with an א, as in Arabic; the editor substitutes the common Hebrew form.²⁸⁴ We find the following Arabic words, almost all of them indicated as such: : אבדאע (54, Kaufmann 317), אהתלא (? 45, Kaufmann 271), 'גנב (35), רהו'ג (6), חמה (xv, xxi, should read ֞הימח?),דימ (45), לס (35), ןיתפצ (45), קמר (4). The names of the sciences are also Arabic: (2 מתמאתיקי lege מוסיקי,הנדסה ?ארת'מאתיקי).

There are [grammatical] forms, phrases that betray Arabic influence, and other peculiarities which will, perhaps, serve to confirm the assumption that NaḤum is the translator, or point to someone else; the following remarks, arranged according to the order of their appearance in the book, do not pretend to exhaust the matter. The bad state of the text, in which the copyists seem to have blurred some peculiarities, makes our attempt all the more difficult.

[The following textual comments refer to Jellinek, ed. Ibn ẓaddiq 1854 ] 1 השכילו (God) has made him intelligible; 5 תוילכ= תוילוכ 25 (תאילכ, also in Hillel b. Samuel²⁸⁵ as a title of the book by Averroes,²⁸⁶ cf. תילוכה שפנה 39, ch. 3); 5 and 49 טושפ odd number,²⁸⁷ commonly דרפנ; 6 תינרקח שפנ, which should be read, according to Brüll 1879, 138 n., as תינרבד (as in Ibn Gabirol’s Ethics 1807. 7, line 1 = Arabic הקטאנ); 7, ch. 2, line 3 הפרקים 8;המעשיים והטבעיים the differences? (cf. 53, Kaufmann 315);²⁸⁸ 8, last line (following XVIII) and 46 (according to Kaufmann, 276) תויד (Arabic היאפכ?);²⁸⁹ 9, last line and 10, line 9 םימוצע cf. םמוצע11, line 13 (cf. §240); 10 בוטר for חל, an Arabism; 11 הדוקנ for center (Kaufmann, 311 n. 158), cf. 15 רובט;²⁹⁰ 11 רבד וילע קפתסי; 14, line 2 םיכתחנ םיתפומ ; ²⁹¹14, 18 ררוגתמ as the opposite of גזמנ; ריקם 16, as Ḥarizi, for ריקות; ibid. מתקומם filling a space; 23, 24, 26, 30 צמחון and plural, besides םיחמצ; 26 ןוערזה; 27 last line ץובקהו טושפה; 28, line 5 שלפלמ²⁹²; 36, 37 52 רשם for definition;²⁹³ 36 מתמים and foemin.²⁹⁴ |410| 37, 38, 44, 51, 52 אפיסה for Arabic עדם (Kaufmann, 262), as Ḥarizi (see §248 below, n. 425b); 37 הדברים הנקפאים (Arabic אלג'מאד, see §124); p. 39 האישות, the individuality; 41 מאיסה 44 ;מזרח andאימות 47 ;חלישות (read אמות, Arabic ֞הליה, Kaufmann, 279); 49 כפל for multiply, as in NaḤum (§228); 50 חטיטה (Arabic תחטיט, Kaufmann, 299); 53 למות, as in Nissim b. Moses b. Solomon (§224); 56 ישנות (according to XXI (Kaufmann), 276 ישות?); p. 68 םעונ, usually תומיענ.^294b

I have discovered a small fragment of the introduction to another translation, printed 1586. [According to Steinschneider 1852, 1542, the other translation is cited in Jonah Girondi, Iggeret ha-teshuvah]; perhaps this was merely begun, and not pursued, or it is a quotation.²⁹⁵

Joseph ben Jacob (Ibn) ẓaddiq, Microcosmos

Budapest, Magyar tudományos akadémia 589 (IMHM F 15015), 1–58.

Hamburg, Staats- und Universitätsbibliothek Cod. hebr. 92 (Hamburg Acc. 1906/11233) (Hamburg 53) (IMHM F 26309), 278b–35a.

Munich, Bayerische Staatsbibliothek Cod. hebr. 65 (IMHM F 1130), 21b–87a.

Oxford, Bodleian Ms. Mich. 491 (Ms. Mich. 575) (Neubauer 1317/1) (IMHM F 22131), 1a–22b.

Oxford, Bodleian Ms. Opp. 583 (Ms. Opp. 1170) (Neubauer 133105) (IMHM F 22145), 97a–140a.

Oxford, Bodleian Ms. Poc. 280 (Ms. Uri 78) (Neubauer 1270/4) (IMHM F 22084), 29a–32a.

Parma, Biblioteca Palatina Cod. Parm. 2450/1 (De Rossi 1174) (IMHM F 13454), 2a–42b.

Rome, Biblioteca Casanatense 3089 (Rome Cas I.I.12) (Sacerdote 161) (IMHM F 65), 31a–73b.

§239. Moses [Abulafia] b. Joseph “b. ha-Lāwi” composed a “theological ‹or metaphysical› treatise”, מאמר אלוהי, on the First Cause, the Prime Mover, etc., in which he discusses some issues of Arabic philosophy. [G. Vajda called attention to citations from the Arabic original of the theological treatise in a Vatican manuscript of a work by Joseph b. Abraham Ibn Waqar 1948. He published excerpts in 1955.]

Manuscripts Oxford, Bodl. Ms. Opp. Add. Qu. 10 (with some notes of the copyists); St. Petersburg, RNL Ms. Evr. I 471 ; New York, JTS Ms. 2341/10 (= Pamplona Estante 6, 21, mentioned in Baer 1929, XII).

[H. Wolfson used the Pamplona ms. for his article on Averroes’ lost treatise on the Prime Mover 1950–51, noting (684, n. 5) that it appears to differ in a number of places from the French translation published in Vajda 1948.]

This treatise has perhaps been translated from Arabic by an anonymous author. It must be the writing of Moses ha-Levi who was later attacked by Ḥasdai Crescas and others.²⁹⁶ Zunz connects this author with the physician Moses b. Meir ha-Levi of the famous Abulafia family in Toledo (died 1255); our Moses b. Joseph is perhaps a cousin of the latter and an elder brother of the famous Todros Abulafia.²⁹⁷ In any case our Moses is identical with Abū Amram (=Abū ‛Imrān) Mūsā al-Lāwī, author of a musical passage quoted by Shem Tov b. Isaac from Tortosa (around 1254–64) without the eulogy for the deceased.²⁹⁸ [The Judaeo-Arabic original of this passage was found by Y. T. Langermann in Solomon Ibn Ya‘īsh's commentary to Ibn Sīnā's Qanūn ; see Langermann 1996a .] The Arabic form of the name confirms my assumption to some extent that Moses wrote in Arabic, which was still alive in Toledo in the fourteenth century.²⁹⁹

Moses b. Joseph, Theological Treatise

New York, Jewish Theological Seminary Ms. 2341/10 (NY JTS Acc. 834/Pamplona Est. 6. 21) (IMHM F 28657), 170a–76a.

Oxford, Bodleian Library Opp. Add. Qu. 10 (Neubauer 1324/5) (IMHM F 22138), 115a–25a.

St. Petersburg, Russian National Library Evr. I 471 (IMHM F 51318), 52a–62b.

§240. Moses Ibn Ezra b. Jacob, in Arabic: Abū Hārūn Mūsā (b. Abī IṢhāq?), member of a famous family in Granada, celebrated as a poet, still alive in 1138,³⁰⁰ is the author of an unedited work (Kitāb al-muḤāḍara wa-l-mu dh ākara, ‘The book of conversations and commemorations’) on Hebrew poetics which was utilized by later |411| bibliographers.³⁰¹ [A critical edition of the work was prepared by N. Bar-On, which later was revised, augmented, and published in 1975 by A. Halkin 1975. For a short history of partial editions, translations, etc., see Fenton 1997 , 32–33. (In n. 103 Fenton attributes the copying of Berlin, SPK Or. oct. 464 to Steinschneider, but according to the Berlin Hebrew manuscripts catalogue, the copying was done by his wife, Auguste, which he then reviewed.)] In this work (f. 38b) the author quotes his work אלחדיקה פי מעני אלמגאז ואלחקיקה, “The Garden, on Metaphor and True Meaning.”³⁰² I take this to be the original of the Hebrew treatise ערוגת הבושם³⁰³ (“Bed of Balsam,’ not to be confused with a ritual work or a medical treatise of the same title).³⁰⁴ [Steinschneider’s speculation, contested by D. Kaufmann, was verified in 1895 when A. Harkavy announced the discovery of an original Arabic manuscript of the Ḥadīqa in the second Firkovitch collection in the Imperial Library in St. Petersburg. This manuscript shows that the title of the medieval Hebrew translation, Arugat ha-Bosem , was already given by Moses. A second manuscript from Aleppo was acquired by D. Sassoon in 1913 and at auction by the Jewish National Library (8 ⁰ 570) in 1975. A critical edition based on the first manuscript had been prepared but not published by the Russian scholar Paul Kokovzov; Fenton 1997 has announced an edition of the original based on both manuscripts and on several fragments found in various libraries. For a short history of the scholarship, see ibid., 36–40.]

Fragments of the Arugat ha-Bosem are extant in the following:

Manuscripts Oxford, Bodl. Ms. Seld. Sup. A 104 ; less complete in Oxford, Bodl. Ms. Mich. 146 , copy of Hamburg, SUB Ms. Cod. hebr. 310a ; ‹Vatican, Neofiti 11/27 has an introduction›.

Perhaps this book was not translated in its entirety.³⁰⁵ The Hamburg manuscript was edited by L. Dukes in the journal ןויצ 1843a with a supplement to it from Oxford, Bodl. Ms. Seld. Sup. A 104 in 1849, 748.

The fragment displays no development of ideas, but rather a compilation of the sayings of the philosophers. Some names are to be found: Aristotle (121, 159; his Metaphysics 135, De coelo, etc. 158); Empedocles (134, 158, 175);³⁰⁶ Hermes, identified with Henoch (123); Plato (121, 138); Pythagoras (in his Golden Treatise);³⁰⁷ Socrates (120 and 135, his Hymns and Prayers, 175);³⁰⁸ the Arab Abū NaṢr (translated as אבו ישע) al-Fārābī (in his book המנהג הטוב);³⁰⁹ [according to Fenton 1997 , this is the first time a Jewish philosopher referred to al-Fārābī by name.] ar-Rāzī (158); furthermore an Arabic poet (136);³¹⁰ החכמה הקדמונית (Ancient, or Oriental Wisdom? 120).³¹¹ Among Jewish authors only Saadia Gaon is mentioned (137, 158); but he seems to quote Solomon Ibn Gabirol, at least in some places, under the name of “the philosopher,” or “the recent (lit. last) philosophers” (121).³¹² [Indeed, the only extant portions of the Judaeo-Arabic original of Ibn Gabirol’s Fons Vitae consist of citations in al-Ḥadīqa ; see the discussion under the entry for Ibn Gabirol. For a more complete list of Jewish sources mentioned, see Fenton, 193–96.] Very rarely are these sayings followed by biblical passages; only in the chapter on the attributes and the names of God (cf. infra) does he quote many biblical verses.

If the paragraphs translated follow the arrangement of the original, the book shows in its development of ideas some similarity with the Microcosm of Josef Ibn Ẓaddiq. It even begins with the analogy between man and every creature because of which man is called the microcosm,³¹³ and the science whose object he is, philosophy (121).³¹⁴ This is followed by a chapter on the unity of God (to which belongs the supplement in Dukes 1849, 748), and other chapters against the ascription of attributes (תואר, כנוי, 122)³¹⁵ and on the names of God (134). |412| The author repeats here emphatically that the attributes (מדות), particularly those which imply materiality, are to be understood as metaphors על דרך העברה (Arabic ‛ען אלמגאז), not in their literal sense (ע“ד האמת) (end of 135). This prominent chapter is probably responsible for the Arabic title, mentioned above, of the treatise. The author rebuts “the nations” (the Muslims) and the sectarians (מינים, the Karaites) who reproach Rabbanite Jews for believing in the corporeality of God (137).³¹⁶ There is no obvious connection between these chapters and the final ones on movement (157), nature (158), and intellect (159).

Assuming that the Arabic title, הקידחלא, is that of the original, we find that the entire fragment was translated from the Arabic; but the Hebrew translator substituted “(flower-) bed” for “garden”, in order to make use of a biblical phrase. This identification notwithstanding, the only possible assumption would be that the author, more precisely the compilator, had translated all the quotations – these are in fact almost co-extensive with the book – from the Arabic. Thus, Ibn Ezra would have compiled this book and at the same time would have translated this mass of sayings – but for which readers? In his time Spanish Jews did not understand this philosophical language, neither those residing in the Islamic south, nor those living in the Christian north. After the ten categories (תוחכה 118 read תוחכה?)³¹⁷ have been named (119) by the author, he adds: “and all of them are to be found in Hebrew.” Furthermore, after explaining what “nature” (עבט) means, he finally says (159): Hebrew has neither a name nor a designation (ראת) for it.³¹⁸ – Does this fit with an author writing in Hebrew? However, a precise translator could have made these remarks.

[Conclusive arguments for the identification of the translator with Judah al-Ḥarizi have been presented by Fenton 1997 , 54–56. The first to suggest this was S. Abramson 1976 , contra M. Idel’s 1977 identification with Judah Ibn Tibbon.]

It is not appropriate to talk of the style of this book. Moses Ibn Ezra, famous for the elegance and dignity of his diction, “a serious thinker who never smiles or jests,”³¹⁹prefers sublime sayings that are almost mystical, probably under the influence of the “Brethren of Sincerity.” What remains of the compilator, apart from them, reveals nothing about the genius of the excellent Hebrew poet.³²⁰

Our treatise was far too little known to allow any conclusion, on the basis of citations, as to when it was translated. Of the quotations, I can identify three or four. David KimḤ̣i,³²¹ who does not mention the title, apparently quotes from a slightly divergent text. ‹Josef KimḤi 1887, 3, mentions the title› Isaac b. Judah, the Babylonian, author of a grammar (1250), names in his preface³²² Moses Ibn Ezra as author of the work ערוגת הבושם, which shows that he knew the Hebrew translation. Joseph b. David the |413| Greek (1300–1350)³²³ certainly quotes from our translation. One author (around the middle of the thirteenth century) seems to have taken two passages from our treatise.^323b This offers a range of a full century for the translation; but it should be dated not far beyond the middle of the thirteenth century, by which time the terminology of the Tibbonids dominated. The characteristic data which our text – though not entirely free from error – presents are not sufficient for an answer to this question. In the absence of something better, however, some terms, not quoted so far and presented in the order of the fragment, may be specified here.

Page 118 מוצא mineral, cf. מתכות on 119; p. 119 עיון, “substance”, perhaps עצום? (cf. §238), תויוג,^323c ילויה hyle, הלעמל ירבדמ (Arabic ילאעת; cf. תומואה ירבדמ 'תי, Dukes 1849, 747); p. 120 תוישות תומואה;³²⁴ p. 122, line 6 הלענ וניאש רבד (? From ללענ =ללועמ, Arabic לולעמ?); םידחאה תונש³²⁵ p. 123 לע רבחהל; p. 136 שופע ארקנ רשא (from Arabic עפש worthless thing, or 'בעוץ?);³²⁶ p. 157 טבעיים elements (but 158 יסודות);³²⁷ p. 158 תנועה מסובבת; p. 159 בגוף הנקרא פינא בלעז או קליינדרו, figa and coliandro (coriander); ibid. כח גופיית and עצם פניני (read p. 120 line 12 for הפנימי); ibid. מוקדמות (Arabic מקדמאת); ibid. תודלות ודלויו,…ב ראותי; םישרפמה ינקז (Arabic 'ךויש [?], 1849, 748).

§241. Maimonides. No Jewish author is more renowned than Maimonides as theologian or as philosopher; this holds true for Christian readers as well. (“R. Moyses”, plainly, is always our Maimonides.) The historians of medieval philosophy have, as a rule, treated his main work as representative of “Jewish philosophy” in general; moreover, they relied upon a Latin translation from Hebrew. We shall not discuss here Maimonides the physician (§481).

In the past half-century abundant sources have become available. More significantly, these have been exploited in a critical fashion by several scholars, most notably Derenbourg, Geiger, and Munk. Now, with regard to the sources, we face an embarras de richesse.³²⁸ This situation relieves us from the duty to give a full-length account of his life³²⁹; nor can we pursue here the full impact of the system of Maimonides |414| upon Jewish theology. However, we shall indicate, where appropriate, the stimulus [Maimonides’ work gave] to the translation of some other works.

[We direct the reader below to recent bibliographical surveys. The most significant discoveries concerning his life and milieu have come from the Cairo Geniza, as well as from a fresh examination of manuscript evidence. For a short survey see Kraemer 2001 , and his biographical monograph 2008 . A thorough analysis of the biographical data is found in Davidson 2005 .]

Maimonides calls himself in Hebrew Moses b. Maimon (which is the Arabic name of his father who had no other Hebrew name), and in Arabic Abū ‘Imrān (= Amram) Musā b. Maymūn (or Maimon) Ibn ‘Abd Allāh, or ‘Ubayd Allāh (translated עובד אלוהים, or עבד האל; servus dei). [The Arabic versions of Maimonides are reviewed at length by Steinschneider in his entry on Maimonides in 1852.] He was born in 1135 in Cordova and accompanied his father when he emigrated to Africa; they arrived in 1165 at St. Jean d’Acre. [On the basis of the colophon to Maimonides’ autograph commentary to the Mishnah, Rabbi Yosef QafiḤ and others such as S. Z. Havlin 1985 have argued forcibly that Maimonides was born in 1138; this appears now to be the scholarly consensus.] In Egypt (hence his name Moses Aegyptiacus) he became physician to the princes (not to Saladin personally, as was wrongly deduced from the word “sultan”) and to the courtiers, specifically to the wazīr al-Fāḍil, (§481, 3); and there he died on December 13, 1204.

Only his great work on Jewish laws and customs is written in Hebrew: the Repetitio legis (Mishneh Torah), often designated only by the name of the author, which Jews usually vocalize according to the acrostic formed by the initials RaMBaM. We shall deal here only with those theological works of his, colored by Arabic philosophy, that offer, with respect to their form and employment of technical terms, some remarkable perspectives for the history of philosophy and the translations. Ignoring their chronological order, we attend first to his most significant work.

§242. 1. (The Guide.)³³⁰ Around the year 1190 Maimonides finished a work in three parts, called דלאלה֞ אלחאירין (The Guide of the Perplexed), or briefly אלדלאלה֞ (The Guide), of which he probably had sent single parts beforehand to his cherished pupil, Judah b. Judah (Ibn Aknin; §237). [Maimonides’ pupil is now thought by most scholars to be Joseph b. Judah Ibn Simeon (Sham‘un). See the supplementary notes to §237.] The purpose of this work is to harmonize the Jewish religion with Peripatetic philosophy in its neoplatonic form as it was developed by Arabic philosophers in Spain. Only the latter, Maimonides says, deserve to be called philosophers, in contrast to the theologians (scholastics, מדברים,מתכלמון), who were followed by the Jewish authors of the East, particularly the Karaites (§263). [Maimonides admires Aristotle and his “early” commentators, which include Alexander Aphrodisias and al-Fārābī, but he distances himself from some of the (unnamed) “later” followers.] In Maimonides’ view, this philosophy was the esoteric doctrine of the Holy Scripture and thus the key to its only valid interpretation. At the same time it comprises the ultimate purpose of human life in its entirety, because this philosophy leads to the attachment to the Active Intellect, which is the sole form of immortality.

His predecessors had already explicated and elaborated upon the basic views of philosophy to some extent. Maimonides, however, did not want to write a new and wholly systematic work. Instead, he hints (Maimonides 1856, I, 291) that his book, like other esoteric writings, will deliberately employ self-contradiction in order to mask certain ideas from the masses.³³¹ Sometimes an idea is developed further only at a different place in his book, to which he refers without, however, indicating exactly where it is to be found. This task was taken up by his commentators. Thus an anonymous author compiled a register of such references in which he indicates the relevant chapters.³³² |415| Maimonides writes for those who, like his student, have studied, or are able to study, strictly philosophical works and who are in need of a harmonization of the philosophical views with the divine word that appears to contradict them. He feels, however, obliged to give a succinct summary of the basic views of the mutakallimūn (I, chapters 73 ff.). This excellent part of the book will always remain a principal source for the history of this type of philosophy.³³³ [For some time Maimonides’ account was a key source for kalam doctrines. However, with the publication in the past half century of so many texts and studies pertaining to the kalam , Maimonides’ survey has lost much of its importance in this respect. On the other hand, scholars have turned their attention to the identification of the specific kalam texts that Maimonides exploited; see Michael Schwarz 1992, 1995 .] This is followed by an exposition of the basic views of the true philosophers³³⁴ in the form of twenty-five propositions placed at the beginning of part II. These attracted, as we have seen (§207), a Muslim commentator. Nonetheless, Maimonides had good reasons to write his book in Hebrew characters.³³⁵ These chapters are full of technical terms for which Hebrew had not yet developed an equivalent which was generally accepted.

The book which was dedicated to a student in Asia, where many manuscripts are still extant,³³⁶ soon enjoyed circulation in the West, including the Provence. [Steinschneider refers in n. 336 to reports of manuscripts of the Dalala that were still being studied in the Yemen. Quite a few of these have since been filmed, and some were incorporated in editions and studies. For a full and annotated listing of manuscripts of the Dalala , see Sirat 2000 and Langermann 2000 .] It was read in the Muslim schools of Fez.^336b Elsewhere, the text was used to amend the Hebrew translations, down to the end of the thirteenth century (Falaquera, Joseph Gikatilia; §244).³³⁷

We are indebted to the scholarship and diligence of Munk for an excellent edition (in Hebrew letters) with a French translation, notes, etc.³³⁸ [A second edition based largely on Munk, but prepared also on the basis of scholarship since Munk, was published by I. Joel. The Munk-Joel edition is now the standard edition of the Judaeo-Arabic original. An Arabic edition, based largely on an Arabic manuscript, was published in Turkey by H. Atay (Maimonides 1974b ), and reprinted in Cairo (Maimonides 1974a.)]

§243. In the beginning of the ‹thirteenth› century Samuel Ibn Tibbon achieved fame, equal or even greater than that of his father Judah, by translating the Guide into Hebrew under the title מורה נבוכים (Moreh Nevukhim).^338b ‹The Moreh (and its author) was frequently called מורה צדק, for example, by Levi b. Abraham, in the preface to Battei ha-Nefesh, 19 (אוצר הספרות III) [For a more accessible and complete version see Davidson 1940 , 84l. 24)]› In a number of letters, some of them in Arabic, the translator addressed the author and asked him about some passages of his translation and others in the text which he found obscure or difficult.

Fragments of this correspondence are extant.³³⁹ I have discovered three unedited pieces in Oxford, Bodl. Ms. Poc. 280 B-b , viz., the end of Samuel’s letter, dated March 1199 and printed further, |416| with a postscript on the translation of the title of the book on Meteora (§61) [Oxford, Bodl. Ms. Poc. 280 B-b], an unedited note of Moses Ibn Tibbon on this problem [Oxford, Bodl. Ms. Poc. 280 B-b],³⁴⁰ and finally a fragment of the answer of Maimonides which talks about the translation of the Arabic passages in question and the doubts of Samuel [Oxford, Bodl. Ms. Poc. 280 B-b]; this piece was also seen by Munk.³⁴¹ [Edited by Diesendruck 1936 . These and other materials related to Maimonides’ correspondence with Samuel Ibn Tibbon are now available in Shailat, ed. Maimonides 1986 II, 511–54.]

Since the Moreh is an epoch-making work in the history of translations, and a document that still awaits precise analysis, it is appropriate to enter into details.

Samuel preceded his translation with a foreword which apparently was, in part, written after this correspondence (vide infra). In it, he recounts that the scholars of Lunel (בקעת יריחו), headed by Jonathan Kohen (a famous commentator of part of the Talmudic work mentioned before), had asked Maimonides to send them his Moreh, if possible in (Hebrew) translation, or else in the original. Upon its receipt they had Samuel translate it. Approaching this difficult task with his weak limited abilities, he adopted two methods: For every dubious word he used the translations of his late father Judah, “the father of translators,” and the works on the Arabic language³⁴² and the Arabic books which he had. Secondly, he consulted the author, writing letters to him, about many doubtful passages. In part, his doubts are due to mistakes in the uncorrected copy of the original. We learn from Samuel’s letters that he had returned this copy to have it corrected by a student once or twice until no mistake remained, and he asks Maimonides to certify the revision.³⁴³ With regard to part III, Samuel goes on to write, Maimonides will find some corrections on the basis of a better copy of a section of this part, because the first transcription was done from a copy in Arabic letters,³⁴⁴ as Samuel had told him already, or from a copy of such a copy. Samuel indicated some places where he suspected an error with ink or with a mark of his fingernail in the margin of the line in question, but not always. He requests that whoever will propose a correction not delete any letter, but rather indicate the amendment clearly in the margin. – What an amount of care it was that Samuel demanded in the interest of a good text, and hence a faithful translation! After the foreword he counted the chapters (see below) in order to facilitate the references of his queries to the author and those of the subsequent readers to others. These queries perhaps constituted the basis for the notes about which we will talk. |417|

Maimonides accordingly replied in Arabic as well, at least in part. This part of their correspondence exists in two translations (one of them perhaps by Samuel himself?). Samuel, Maimonides says, has hit upon the incorrect passages and put the right questions which Maimonides answers at the end of the letter in all details. He recognizes Samuel as an apt and adroit translator and tells him (in Hebrew) how he should go about the whole translation.³⁴⁵ He is surprised that someone born among “barbarians” (עג‘ם = עלגים; non-Arabs) would pursue the sciences and understand Arabic so well – but that language, after all, is only a somewhat corrupt form of Hebrew;³⁴⁶ Samuel is but a “root in barren soil.”³⁴⁷

Before entering into particulars, Maimonides offers a general rule for all translators. They should render one word by one word only and keep to the order of the sentence.³⁴⁸ But keeping the word order (in the translation according to the original) is very difficult and leads to an unclear and imprecise translation, so this should not be done. Instead, the translator should first grasp the meaning of the text and then render that meaning³⁴⁹ clearly in a manner suited to the target language. Often this is impossible without changing the word order and without rendering one word by several words, or, conversely, without omissions or additions of a word. This was the method of Ḥunayn and his son IsḤāq (§197), Maimonides says, and it should be followed by Samuel in translating for his patrons.

After this passage Maimonides enters into the details,³⁵⁰ presenting the Arabic passages with their Hebrew translations (which Samuel has also inserted).³⁵¹ This passage is still unedited. [As indicated above, the entire correspondence is now available in the edition of Shailat (Maimonides 1986 ).] It ends with an apology in which Maimonides remarks that a translation is an original composition of sorts; hence the translator is, in a way, a co-author.³⁵² He closes with the admonition not to stick to the translations (פירושים) that he has proposed should he, Samuel, find better ones, and says again that the translator should understand the text before he translates. After this he deals with the passages that Samuel had forwarded to him for explanation. This part has been edited in two copies (one by Munk, the other by Goldberg, without indication of source, perhaps from a different translation?).³⁵³ Maimonides next turns to address Samuel’s plan to visit him (no doubt in order to ask him about the translation |418|; as we shall see, Samuel in fact went to Egypt later on) and gives interesting details about his position and work there – this passage has been repeated by almost all biographers. In the editions this passage closes with a remark on the completion of the translation. In the other translation, Maimonides admits that a more precise translation of the title would be הוראת הנבוכים [Instruction of the Perplexed)] but the term מורה (Arabic: דליל) fits better nonetheless [Shailat (Maimonides 1986 ) II, 523]. Maimonides’ enemies distorted the name: the Arabs called it צ‘לאלה (that which leads to error), the Hebrew writers, נבוכת המורים (the confusion of the rebels).³⁵⁴ [The Arabic distortion is reported by ‘Abd al-Lat ̣īf al-Baghdādī, who met Maimonides in Egypt and has left us a very negative impression of their meeting. See Davidson 2005 , 426.] The rest of this letter has already been analyzed (§13).

Let us return now to the foreword of Samuel, which became the prototype or source for later translators and which was quoted by other authors when they did not want to repeat its contents.³⁵⁵ First, Samuel apologizes for having undertaken an arduous task which demands knowledge of both languages. Translation, he states, has four causes, which he compares with those of a building.³⁵⁶ He refers to the foreword of his father to the translation of BaḤya’s book, which discusses the difficulties of this task. Moreover, he confesses that his knowledge of Arabic is but scanty, since he has not been educated among Arabs and in their country. The difficulty of the Moreh lies in its profundity, which, in turn, is due to the sciences with which it deals and which are studied in this region (Provence) only to a small degree. Some of it he has read in Arabic books.

He would not have undertaken the translation, were it not for the wise men etc. (see above). He asks the reader to excuse his mistakes, be they grammatical (gender, number) and caused by the Arabic word, or syntactical, as, e.g., the singular form of the verb preceding a noun in the plural, for which there exists an analogy in Hebrew. This is correct, the philologists (בעלי הלשון) explain, because the verb refers to every single component of the plural, particularly regarding the words היה and הוא (copula). Furthermore, the Arabs do not distinguish between feminine singular and masculine plural predicates, and in the plural forms between masculine and feminine (referring to the pluralis fractus). Arabic verbs are connected (יקשר) by prepositions different from those employed in Hebrew; in Arabic the word שם) ת'ם) denotes existence.³⁵⁷ And then there are idiomatic expressions that are quite common in Arabic and very rare in Hebrew, e.g., על דעתי, which occurs only in Job 10:7.³⁵⁸ These peculiarities lead astray the translator who aims to reproduce the sense of the text. After having translated the work, Samuel wants to revise it. Maimonides himself uses the word שם, mentioned above, in its Arabic sense, whereupon Samuel’s audience who did not know Arabic read shem (name). So if the Arabic |419| author, writing in Hebrew, could not avoid arabisms, how could the translator? – A number of words, he says, of whose gender in Hebrew he is not sure, he will consider arbitrarily as masculine or feminine. Words ending in ת-, like תוכיא, תוהמ, תולעפה, תולכתשה, תומלש, תילכת, and תעד, can be treated as feminine; but since one finds the same form as masculine, e.g., פלצות (Job 21:6), or ראשית (ibid., 8, 7), he will construe them as masculine. Words that alternate between masculine and feminine forms in our texts, he will use alternatively, as they occur to him. Words that allow for two different translations will be translated differently; sometimes he will supply the second meaning in the margin only in one place. In all this he follows his father, just as he does in creating new forms (תונבב םינינב) which do not yet exist, e.g., ףסלפתנ [ףסלפתה, ףסלפתמ?] after the Arabic ףסלפתמ. After all, authors of scholarly works create new derivations and give to commonly known words different meanings, as long as there is some kind of similarity, even if it is not real, between the two meanings. All this is due to the inadequacy (רצוק, again an arabism) of every language to express the concepts of profound (abstract) sciences. Even the prophets were forced to use metaphors when talking about God, the angels, and other concealed things, as the author of our work himself says. Samuel does not want to change the rhetorical style in order to preserve the meaning intended by the author. Sometimes, in the course of translating, he may not recall the apt or more fitting term, or he may not know it at all. This, however, can happen to more learned and more proficient translators.

Our translator certainly has contemplated the exigencies of his craft thoroughly. Though modern critics³⁵⁹ find in his translation only a “poor imitation” of the original, its fidelity enables us, possessing but little knowledge of Arabic, to re-translate the book and understand it. On the other hand, it has to be kept in mind that we are talking about a language that was at least half-extinct, whose spirit was not compatible with abstract sciences. As we shall see, Samuel composed a glossary to help the reader of this new idiom. He classified the foreign terms used by him and tried to reduce their number.

Characterizing Samuel’s translation in detail would mean nothing less than to write a book on the philosophical, or more generally, the scientific style of Hebrew writing that readers of this book eventually developed.³⁶⁰ A number of Arabic words employed here became full citizens of the Hebrew language, e.g., the mathematical terms (קפוא) קפא, אנווטצא, זכרמ, הסדנה, בטק³⁶¹, רטק; also םסלט, and Hebrew words having the same meaning as Arabic ones: םשג for body, הבס, הלע for cause, רצק, רוצק for inability. Other words gained a meaning according to a |420| concatenation of concepts going back to the Greek, like הלילש (בלס) στέρησις, privatio (negatio); רבדמ, קטאנ, meaning “possessing the capacity to think”; and, in particular, some terms for the sciences, e.g., קטנמ,ןויגה, logic; םידומיל, םילעת mathesis; also םיישומש, םיילגרה (הי'צאיר), תרובשת (רב'גלא) algebra. Expressions like וילע אוהש המ לע (הילע וה אמ ילע) are arabisms (§244).

§244. (The Glossary, the Critique). According to the postscript, Samuel finished his translation in Arles on November 30, 1204, 14 days before the death of Maimonides.³⁶² Doubtlessly it was copied soon and frequently, and perhaps among the great number of extant manuscripts there are some copies of the first edition which may be identified by means of textual variants that will be discussed presently. The translator felt the necessity to compile a glossary of the foreign terms, to which he gave the title (itself an arabism)³⁶³ ביאור מהמלות (ה)זרות, more precisely פירוש מן המלים הזרות, commonly appearing together with the Moreh in the manuscripts, and only rarely separately. It appears in the editions of the Moreh from 1551 onwards, though the text is not correct. Geiger³⁶⁴ extracted some additions (and some later supplements) from a manuscript. I do not know of a more recent edition which has made use of this information. Isak Satanow enlarged the glossary with some additions. [The most recent edition by Y. Even Shmuel (Kaufmann) 1987, now included in his one-volume vocalized edition of the Ibn Tibbon translation of the Guide, is based on the printed editions and some manuscripts; see p. 8 for details.] Parma, BP Cod. Parm. 2730 and Turin, BN 32 A II 12 [, no longer extant,] display a postscript according to which it was completed in Tammuz 893 (1213)³⁶⁵ while returning from Alexandria, on a boat in the great port of אלקליבה or לאלקליבה, four mil from Tunis. According to the catalogue (Peyron 1880, 37), the Turin ms. has, among other materials, two supplements, otherwise unknown.³⁶⁶

The beginning of the introduction to the glossary is not unambiguous: As he was completing (בהשלימי) his translation, Samuel tells us, he sensed the need to compile a glossary,³⁶⁷ particularly so as the “poet” Judah al-Ḥarizi, who had translated the book (“following our translation”, Turin ms.), had done so. The latter had prefaced his translation with two gates (chapters), one of which explains the (difficult) words but contains much unfounded, erroneous and mistaken (מכשולים) material, while the other, which lists the theme (כונת) of each chapter, “is full of stumbling-blocks” (§247). Did Samuel begin his glossary in 1204, but finish it only in 1213? Or was he prevented from completing his plan until his stay on the boat offered him the necessary leisure? This question also touches upon the problem of the date of his rival’s translation (§247).

The general remarks with which the alphabetic glossary begins show us the devices, or means, employed by the translator in his classification. These are then illustrated in the foreword by means of selected examples, without strict arrangement.³⁶⁸ Here, Samuel enumerates the words that require clarification, grouping them into five categories: (1) wholly |421| novel ones (חדוש גמור), the work either of earlier translators or of himself, as, e.g., קוטר and קוטב, taken from Arabic or another language ‹cf. 899 below›; (2) words from the Mishna or the Talmud (the Gemara), which are known only to a few scholars, as סיטסא and גוס; (3) verbs or adjectives derived from known nouns, e.g., verbal forms derived from אמת, in analogy to Arabic, מלאכית (read) from מלאכה, as in Arabic, or new verbal conjugations, e.g., קתענ from קתע; (4) homonyms (תופתושמ, including metaphors, or strictly homonyms, מצד השאלה or מצד הספוק), taken in a specific sense, like איש – this word gives the author reason for a prolonged and severe attack on the “poet” and his table of contents, and, at the same time, a defense against Ḥarizi’s remark in his prologue, accusing Samuel of intentionally making the book unclear. “No,” he replies, “I have observed the restraint that Maimonides requires from his readers; Ḥarizi, however, who divulges the secrets to the multitude, errs and gives offense.” Samuel assures his readers that his pronouncements concerning this deceitful chapter were not instigated by a sense of rivalry – although, to us, the harshness of his tone indeed betrays the sense of rivalry and hurt. Ḥarizi may, according to Samuel, understand Hebrew and Arabic. He is capable of translating texts that are easy to understand (מובנים), such as poetical and linguistic (ספרי לשון) works and chronicles;³⁶⁹ that is his job and his profession. However, he has taken the gross liberty (הרס) of translating scientific books that he does not understand. Therefore, he errs in translating even very simple words, both in this book³⁷⁰ and in the introduction to the Mishna commentary of the “Teacher of Justice” (מורה צדק, later on a conventional designation of the author of the Moreh). Moreover, Judah makes mistakes concerning Hebrew words in both texts that children would not make; he even confuses the Hebrew ממר with the Arabic. Again, Samuel declares that he is not trying to promote himself by disparaging Ḥarizi (התכבד בקלונו), etc.; (5) The fifth category consists of words to which he has given a new meaning, again by analogy to the Arabic, as, e.g., מצייר, ציור, to designate the representation of a thing according to its reality (Arabic רוצת); (6) [The final category consists of] the rendering of sentences of which every single word is known but whose context is difficult to understand for someone who does not understand the (language of the) mathematicians (בעלי חכמת הלימודים). This category is again a field in which the “poet” has sinned.

After these remarks Samuel indicates the amendments that should be made, on the basis of his revised version, to the copies “scattered over the earth.” He has substituted על by לפי and כפי, particularly in the idiom על דעת, although it does occur in Job (10:7) (§243, end); some passages, however, may have escaped him. He has supplied the word עליו in the idiom על מה שהוא עליו which is not proper Hebrew. He had used מופת (strict proof) for ראיה (argument) |422| as well, but now he has introduced the latter so that the former always means the same thing. He puts נושא for משכן, and קסם,³⁷¹ which is found in Maimonides’ “Epistle to Yemen,” for כהון. – In the glossary he adds, s.v.: “For when our language (Hebrew) has a certain word, one must not use a foreign word in its place, if there is no particular reason to do so.” – He concludes this introduction, or, rather, he introduces the glossary proper, with the remark that he will observe the alphabetical order with regard to the second letter of the word as well. Moreover, he will arrange some (derived) words not according to their root, but rather by their first letter. In fact he takes no notice of the third letter. The first three entries are, איכות איכול,איש (read אכול).

This glossary gives not only a brief explanation of the word and its origin, e.g., a title מגסטי (by Ptolemy) and a people, צאבה, but very often also the scientific definition, giving examples and even going into arguments.

Already in the first entry, איכות, he remarks that it refers to one of the ten categories treated by Aristotle in the book known as ספר המאמרות and that “category” should actually be translated by a participle נאמרות or נאמרים, following the Arabic (we find the latter in Falaquera); but Samuel adopted the word מאמרות from the older translators and authors. Before he enumerates the categories he elucidates seven terms, regularly used by the dialecticians (חכמי המופת); in fact these are the quinque voces of Porphyry: מקרה,סגולה,מבדיל,מין,סוג, with the addition of two related terms, definition and description, גדר and חק, the first of which was coined (חדשום) by the authors and translators for the Arabic חק. The second, standing for רסם, was introduced, he supposes, for the first time by himself, from which he also derived a verb, as in Arabic. After the ten categories someone has erroneously put the title “Letter alef”, but all of the preceding material was, in fact, inserted in the entry איכות. [This is not in Even Shmuel’s edition.]

Samuel’s translation found its critic in Shem Tov Falaquera, who was superior to him, better versed in the literature of Arabic philosophy, and even more committed to [textual] fidelity. In 1280 he wrote a commentary on most of the Arabic Moreh in which he retranslated into Hebrew those passages which he chose to explain. Moreover, he collected parallel passages or explanatory material to Maimonides from Arabic philosophers and presented these in Hebrew translation. His main source is Averroes, for whom he employs the shorthand “the aforementioned philosopher.” This work was printed in Pressburg [Bratislava] in 1837 under the title Moreh ha-Moreh (Guide to the Guide) 1837.³⁷² [See now the edition of Shiffman 2001.] The wealth of quotations it contains is such as to give the semblance of great scholarship to those who took excerpts from it. Falaquera appends three chapters to his commentary: (1) Passages from philosophical works on the human attainment of perfection and the conjunction (with the Active Intellect), mentioned by Maimonides; (2) a resolution of the doubts raised by the |423| translator Samuel concerning Maimonides’ views on providence and miracles (Moreh III, 51; §243). After giving an abstract of Samuel’s treatment, he gives us his own views on the problem in question and, at the same time, corrects a grave mistake in the translation (147) which at first glance only seems to present a minor nuance. (3) Corrections of the Translator. – Fr. Delitzsch began to translate this chapter into German while correcting the Arabic passages without the help of a manuscript, but still adding useful notes.³⁷³ This chapter starts with the remark that a book that has been written with such care must be translated with no less precision in order to preserve the meaning. For many words have an inner (פנימי, mysterious) meaning which is known only to those versed in scholarly works. This is what Maimonides alludes to in his “recommendation” (Maimonides 1856, I, 22) [Pines 1963, 15, translates this “instruction”]. He concludes his remark with the observation that the translator would certainly have corrected this translation had he been aware of his errors.

Samuel’s translation lost much of its importance for the text with the appearance of the edition of the [original] text itself and a French translation. Munk did not neglect the help that the Hebrew translation could offer when correct manuscripts were utilized. In any event, the work of Samuel will remain not just one of the most important documents for the history of translations, in the course of which it has served so many for so long as a model. It is also the best guide for acquiring this artificial language. This feature has even been enhanced by the edition of the [Judaeo-Arabic] text.

§245. (Editions, translations, commentaries.) The Hebrew Moreh is one of the first products of the Jewish printing press. It was printed, without indication of place or year, shortly before 1480, probably in Italy,³⁷⁴ and continuously reprinted thereafter; but already the second edition (1551) is accompanied by three medieval commentaries.

Studied as it was by scholars, the book was at first not popular enough to be translated into modern languages. ‹There is an Italian translation by Jedidiah b. Moses in 1583, on which see Sacerdote 1892.› In 1829³⁷⁵ Mendel Levin (Maimonides 1829) published a partial Hebrew paraphrase. The Latin paraphrase of the younger Buxtorf (Maimonides 1629) has its merits, considering the time and his limited means.³⁷⁶ His translation of the title by “Doctor perplexorum” has gained currency; the old Latin translation (§250) exhibits other versions of the title. A German translation of the three parts has recently been supplied by three scholars: I by Fürstenthal (Maimonides 1839), III by Scheyer (Maimonides 1838), with the help of the original text, and II by M. E. Stern (Maimonides 1864), who was able already to make use of the French translation by Munk, as was done also in the Italian translation, begun by D. J. Maroni (Maimonides 1871), the Hungarian translation of Moritz Klein (Maimonides 1878–1890, 1977), and the complete English translation of M. Friedlaender (Maimonides 1881–85), which contains a thorough analysis of the whole work.

[Of the many other translations since Steinschneider, three should be singled out: There is the English version executed by Shlomo Pines (Maimonides 1963 ), accompanied by an extensive essay on the sources of the Guide as well as a proemium by Leo Strauss on how to read the Guide . There are notes but they are very sparse. In addition, the Guide has been translated twice into modern Hebrew. Rabbi Yosef Qafih published his own translation, with a facing edition of the text in the original (Maimonides 1972 ). Rabbi Qafih’s edition utilizes the Munk-Joel edition as well as a number of other manuscripts, all of them Yemenite. In the introduction, Rabbi Qafih lambasts the medieval translations of Ibn Tibbon and al-Ḥarizi. Significant divergences from those earlier translations are noted ad loc . The notes also include much other valuable material, including cross-references to other Maimonidean texts and the rabbi’s own Hebrew renderings of pertinent texts by al-Fārābī; perhaps their most useful and original contribution are the numerous cross-references to Saadia, whose writings (especially his biblical translations) seem to have exerted a strong influence upon Maimonides; this was not noticed previously, and its full significance remains to be clarified. Rabbi Qafih’s translation alone has been reprinted many times, but the version with the facing Judaeo-Arabic text was printed only once and is no longer available. Michael Schwarz published an elegant and accurate Hebrew translation (Maimonides 2002 ), accompanied by copious notes, which are particularly strong in all that concerns philosophical terminology and scholarly bibliography. Notice must also be taken of Y. Even-Shmuel’s proto-edition of Ibn Tibbon’s translation (Maimonides 1935, 1987 ), described on the title page as a “pointed and corrected publication, on the basis of the first printings, with variants from manuscripts, and compared to the Arabic original.” His text is accompanied by copious notes, most of which are based upon medieval commentaries. This achievement notwithstanding, an edition of the Ibn Tibbon version remains a desideratum.

Among the other languages that the Guide has been translated into are Latin (Maimonides 1520, 1629 ; on the Latin versions see below), Judaeo-German (Maimonides 1839), Italian (Maimonides 1861), Persian (Maimonides 2011), Spanish (Maimonides 1984, 1987, 1988 , Hungarian (Maimonides 1878–1890), and Chinese Maimonides 1998 .]

Commentary is generally one of the favorite genres of medieval literature and, for special reasons, dominates |424| Jewish scholarly literature.³⁷⁷ The Guide stimulated enterprising spirits to reveal what it sought to hide, and it was easy to find an excuse to trespass on the author’s solemn invocation – Samuel Tibbon had already criticized his rival on this very point.³⁷⁸

On the other hand, the new style, as well as the knowledge that Maimonides presupposed in his readers, required most of them to look for explanations. The controversy about the new system of theology attracted the attention of those who were not accustomed to an exclusively rational way of arguing. Certainly the book was read openly, and its ideas found their way into the sermons, e.g., those of the translator Jacob Anatoli (מלמד התלמידים), who was the target of Orthodoxy’s wrath. So one should not be surprised about the number of commentaries still extant, almost all of them to Samuel’s translation, which he himself had already accompanied with notes (unedited, because they are extant only in a few manuscripts).³⁷⁹

[The critical notes of Ibn Tibbon to the Guide have now been edited in the 1999 master’s dissertation of Carlos Fraenkel ; they were studied further in the same scholar’s 2001 doctoral thesis and monograph 2007. Fraenkel’s painstaking investigation reveals that, in fact, Ibn Tibbon’s notes are preserved in quite a few manuscripts, though most manuscripts contain only a few of them, and many are recorded anonymously. Fraenkel’s studies contain the most thorough survey yet undertaken on manuscripts of the Ibn Tibbon translation.

Steinschneider comments below that he is referring to only a few of the unpublished commentaries. In fact their number is significant; one should also include under this rubric marginalia to manuscripts, which are very dense in some cases, as well as the few but interesting commentaries and glosses to the Judaeo-Arabic Dalāla . The only survey is the one published by Steinschneider himself 10 years after HUe in the A. Berliner Festschrift (Steinschneider 1903 ). The only text of a Hebrew commentary to have been edited since then has been Y. Shiffman’s edition of Falaquera’s commentary 2001 .

Any discussion of commentaries to the Guide (and other works as well) must also take into account marginalia and other annotations. Indeed, many commentaries are in fact marginalia that were collected and lightly edited. The most heavily annotated copy of Ibn Tibbon’s translation is Sassoon 341 ; most of the glosses are listed in the catalogue 1932 1:417–419. Another heavily glossed copy which is, however, in a horrible state of disorder, is St. Petersburg RNL Ms. Evr. I 474 . Some copies of the Judaeo-Arabic original bear annotations as well; see Langermann 1995 . Finally, mention must be made of the extremely dense, often illegible marginalia, to the Arabic-letter copy of the Guide found in Istanbul, Carullah 1279; the sources named in the glosses (which too are in Arabic letters) were meticulously listed by the late Franz Rosenthal 1955 ; one must go through the indices and see if the folios cited fall between 189b–301a, which contain the Guide. The notes include citations from some Jewish works that are otherwise lost. A selection of marginalia from the Carullah manuscript, highly critical of Maimonides and of revealed religion and its tenets in general, with Hebrew translation and analysis by Almog Kasher and Y. Tzvi Langermann, appears in Langermann and Kasher 2013 .]

The first known commentary is from Italy and dates from the middle of the thirteenth century (§250). Towards the turn of that century (1290), Abraham Abulafia dared to compose a mixture of philosophy and mysticism, arranged according to the chapters of the More. It is, in part, an absurd concoction. Two recensions exist. [For a detailed account of recensions and manuscripts of Abulafia’s commentary, see Idel 1976 .] ³⁸⁰ We have seen that shortly beforehand Falaquera had explicated the Moreh. The main commentaries, however, belong to the fourteenth and fifteenth centuries. A critical assessment of this literature would divert us too far from our subject, although it offers interesting details which are highlighted in their appropriate places. We have to confine ourselves to a list of published commentaries, arranged according to date of composition and indicating only the first editions. From among the unpublished commentaries we name only a few and refer for details to the bibliographies.³⁸¹ [Once again, the reader is referred to Steinschneider 1903, which lists over sixty Hebrew commentaries to the Guide.]

Joseph Kaspi wrote (around 1330) a double commentary, dealing separately with the explanation of the “secrets.” See Kaspi 1848. The manuscripts: Munich, BS Cod. hebr. 263/1 and Oxford, Bodl. Ms. Hunt. 559 contain divergent recensions. [According to Neubauer, Oxford, Bodl. Ms. Hunt. 559 is Kaspi’s commentary, but this is unclear. Other manuscripts: New York, JTS Ms. 2341/3 ; Paris, BN héb 694 ; Paris, BN héb 700/8 ; Turin BN A VI 34, no longer extant; Vienna, ON 35 .]

Moses Narboni who, in his commentaries on the philosophical works of the Arabs, never loses sight of the Moreh, completed in 1362, after 7 years of uninterrupted work, an interpretation of “the book which has achieved fame among Jews, Christians, and Ismaelites,” addressed to an intimate circle of scholars and concentrating on revealing the “secrets”, in Soria, according to the epilogue (printed twice, 1880 and 1881) which is missing in the – very inaccurate and defective – edition (1852).³⁸² Falaquera, Kaspi, and Narboni are the most important basic commentators of the Moreh. Profiat |425| Duran (“Ephodaeus”, or Isaac b. Moses Levi, cf. section II and III) composed (1391–1403) a short and simple explanation (ed. 1551);³⁸³ the Paris catalogue (no. 705) did not recognize it. The absence of a foreword gives us no reason to suppose that it is truncated or mutilated.³⁸⁴ According to Delmedigo, Profiat furnishes the reader with correct answers, but spares the reader the questions, following the style of R. Solomon ben Isaac (Rashi), the famous commentator of the Talmud. [London, BL Or. 1388 ] ‹Isaac b. Shem Tov, ms. Parma R. 1388 [?]›

Asher b. Abraham, or Bonan Crescas (probably first half of the fifteenth century, ed. 1551),³⁸⁵ plans to interpret several parts of the book for adolescents,³⁸⁶ young people who have not yet acquired the sciences that are necessary for the correct understanding of it, but who seek instruction. In his high respect for Maimonides³⁸⁷ he does not expect to penetrate everywhere into the profound ideas of the book and always combines the admission of his own ignorance with his objections and doubts. He knows the commentaries of Falaquera and Kaspi.³⁸⁸

Shem Tov b. Joseph b. Shem Tov composed his commentary in 1488 (cf. I, 74, at the end) with an attitude of reconciliation between reason and law – ”the two lights” of which the former is the greater.³⁸⁹ This is all the more significant as his grandfather was a zealous Kabbalist, and his father objected to some of the basic ideas of Maimonides. In his foreword, he says that he plans to interpret some profound passages and to make use of everything that is correct in the commentaries. As a matter of fact, this foreword already contains a borrowing from Narboni.

Soon afterwards (1493)³⁹⁰ Don Isaac Abravanel commented upon the greater part of the Moreh, leaving, however, his work (ed. 1831–2)³⁹¹ without proemium and probably unfinished. He had undertaken the commentary with the intention of opposing some scarcely orthodox interpretations, particularly those of Kaspi, Narboni, and Pr. Duran.³⁹² The textual passages are often given in paraphrase, but following Tibbon’s translation.

After the medieval commentaries, we mention the notes of a famous Polish Talmudist, Mordechai Jafe (1594) to the printed commentaries; the commentary of a philosopher in the Kantian tradition, Solomon Maimon; and that of an industrious and skillful author, Isaac Satanow (1791ff.). Simon Scheyer wrote an interpretation of part II, chapter 45 under the title of מעלות הנבואה, Commentarius hebraicus etc. de prophetiae gradibus, Rödelheim 1848, 16.

Among the unedited commentaries we name that of Solomon |426| b. Judah ha-Nasi (1368) who, after having had one Jacob b. Samuel, otherwise unknown, as a student in Germany for 2 years (the name of this country occurs here for the first time), composed a commentary to the Moreh for him, in commemoration and gratitude for the honorable treatment that he received:

Manuscripts. London, BD and BM 52 ; Cambridge, UL Add. 393/2 .³⁹³ [Michael Z. Nahorai has prepared an edition of this commentary but has not yet published it.]

In the first decades of the sixteenth century, David b. Judah Messer Leon wrote a commentary under the title of עין הקורא (Eye of the Reader); it is found in Oxford, Bodl. Ms. Reggio 41 and Parma, BP Cod. Parm. 3546 . Günzburg has a fragment of it,³⁹⁴ rich in quotations (partly borrowed). In David’s eyes, Abravanel is an amateur who has read only one book of al-Ghāzālī,³⁹⁵ while Levi b. Gershom is a heretic whose book deserves to be burnt because he battles against Aristotle and Averroes with trifles.

Commentaries to single parts of the Moreh were also written, e.g., to the 25 (or 26) introductory propositions of part II. Authors of such commentaries are (end of the thirteenth century): Jedaia ha-Penini, whose commentary (מדבר קדמות) is known only from his own quotation;³⁹⁶ Hillel b. Samuel, whose commentary has been edited (1874) along with his philosophical work (Hillel ben Samuel 1874); (end of the fifteenth century:) David b. YaḤya b. Solomon from Lisbon, a preacher in Corfu;^396b (sixteenth century): Moses Provençal (Mantua, CI 39).

Perhaps the piece called “Short Explanation,” an analysis of the two kinds of lines (hyperbole and asymptote) mentioned in the Moreh (I, 73, 410 in Munk’s French translation), by Simon Motot (1446–50), Munich, BS Cod. hebr. 36/28 and Vienna, ÖN Vienna, ON 46/3 ³⁹⁷ figures in the commentary of Moses Provençal (1549), which was published together with the Moreh in Hebrew (Sabionetta, Maimonides 1553) and was translated into Latin by Baroccius (Barozzi 1586) under the name of Moses Narboni and on the basis of an Italian translation (1550) printed in Hebrew letters.³⁹⁸

[Maimonides’ description ( Guide I, 73) of the asymptote to the hyperbole engendered a number of elaborations in Hebrew; although similar constructions exist in Arabic and Latin, these were especially necessary for the Hebrew reader, since Apollonius’ Conics was never translated into Hebrew. Proofs that do not depend upon Apollonius have been collected and studied by Lévy 1989 . For a discussion of the matter see Freudenthal 1988 .]

§246. (Introduction, versification.) [With the exception of the very last sentence, this entire section is devoted to the work Ru’aḤ Ḥen . Steinschneider looks upon that writing as an “introduction” to the Guide ; it seems likely that a more precise reappraisal would classify it as synopsis of philosophy heavily dependent upon the Guide, rather than a tool to prepare one for the study of Maimonides’ book. The authorship of the tract has never been determined even though, as Steinschneider can already report, the first generation of modern scholars invested considerable effort in the study of the treatise. A Hebrew University dissertation of Ofer Elior 2010 on the matter is being prepared for publication.]

An anonymous author wrote, probably between 1200 and 1250, a small introductory treatise (numerous printings since 1544) which discusses the basic philosophical ideas that are necessary for understanding the Moreh. It is called, after the first two words, חן רוח, Spirit of Grace, and it is divided into eleven chapters; the supplement on equivocal, homonymous, and metaphorical³⁹⁹ terms is probably the work of a different author. This small treatise has been ascribed, without reason, to various authors: one of the three Tibbonids (Judah, Samuel, Moses), Jacob Anatoli, then again another Anatoli,⁴⁰⁰ or ZeraḤiah ha-Levi Anatoli, identifying him with an |427| author of hymns in Greece, because Anatoli (anatolê) is the translation of ZeraḤiah. This small treatise, however, has no connection to Greece.⁴⁰¹ It has not yet been noted that in chapter 2 and 3 the translation of al-Fārābī’s Book of Principles has been utilized.⁴⁰² The small treatise is already quoted by Abraham Abulafia and in an anonymous medical work (around the end of the thirteenth century);⁴⁰³ Gershon b. Solomon presents extended excerpts from it.⁴⁰⁴

It found many medieval readers;⁴⁰⁵ its manuscripts are almost countless. [The catalogue of the Institute for Microfilmed Hebrew Manuscripts now lists over a hundred copies, not including commentaries, abridgements, etc.] The first Firkovich collection has no less than ten manuscripts (nos. 491–500); some of them, however, are copies from printed books. In Ha-Karmel, Firkovich gives some information on no. 494, which D. Slutzki appropriated for the introduction to his edition (Warsaw, 1865). In his catalogue Firkovich assumes no. 488 to be an autograph!

The small treatise seemed to be important enough to attract commentaries, and not only in the Provence, where one of the students of Prat Maimon (around 1420) or he himself wrote a commentary, the beginning of which, down to the middle of chapter 4, is attributed to Natanel Kaspi in the Paris catalogue (107) on Paris, BN héb 678/1 .⁴⁰⁶ ‹ = Parma, BP Cod. Parm. 2255/2 ; this means that the commentary in the same ms. on Maimonides’ Eight Chapters is by Natanel›. The beginning tallies with the anonymous commentary in Paris, BN héb 1239/4 , whereas the anonymous commentary in the editions 1549, 1566, etc.⁴⁰⁷ is different; it contains interpolations from “another commentary” (chs. 3 and 7). But also in Germany a learned Talmudist, Zalman (or Seligman Zion Levi) from Bingen (around 1450–60), wrote a philosophical and mystical commentary⁴⁰⁸ which is probably lost, if it is not one of the two anonymous commentaries, published in 1594 and 1620. – From among the editions⁴⁰⁹ we mention that of the baptized Jew Joseph Isaac (Cologne 1555), accompanied by a poor translation, and published under the title Physica hebraea nunc primum edita, etc.; the vocalization is inexact, but the variant readings sometimes are superior.

The author of the רוח חן quotes from the Meteorology under the title of אותות השמים (chapter 7). He talks about the ten categories, and he follows, from among contemporary enumerations, most closely the logic of Maimonides in its translation by Moses Tibbon (1254), both in the order and in the terminology of the two last categories.⁴¹⁰ |428| Otherwise the style does not have any conspicuous characteristics. The author considers קוטר (chapter 3, no. 1; chapter 6, no. 4) to be a well-known word; היולי is Greek, for which the Hebrew is חומר ראשון (chapter 8); duality, or plurality (that is, in God), is שניות (chapter 6, no. 4); the mineral is דומם; the “higher elements” (היסודות העליונים, chapter 5) seem to be the “higher bodies” (הגרמים, or הגשמים, in other authors). – Perhaps we have spent more time with this little book than one would expect. We have done so because it sums up the most sublime philosophical problems within a very restricted framework, and it illustrates the character of the clear and simple philosophical style of the Tibbonids’ time.

[A number of other works may be cited here that are paraphrases of the Guide , or parts of it; their kinship to the Guide is far more intimate than that of Ru’ah Hen .

• (a) Nev’uat ha-Moreh , a summation of Guide II, 32–48, in which Maimonides’ theory of prophecy is set forth. (New York, JTS Ms. 2441/8 , 130a, 133b; 16th century).

• (b) A paraphrase of Guide I, 1, 3, and 4, done perhaps by a Karaite. (St. Petersburg IOS B 342a , 123a–b; 18th century).

• (c) An odd, incomplete work, parts of which are attributed to “Galen” but which is really compiled from extracts of Ibn Tibbon’s translation of Guide I, 8–30; II, 47–48; and III, 14. (St. Petersburg RNL Ms. Evr. I 476 ; 15th/16th centuries).]

The Moreh was versified by Mattithiah Ibn חרטון (Kartin).⁴¹¹ [In 1885 Steinschneider published a collection of Hebrew poems relating to “Maimonides and his Famous Books” 1885b ]

§247. Soon after Samuel (1205–13) the famous poet Judah al-Ḥarizi translated the Moreh. This translation was published on the basis of a single ms. (written in Rome 1234)⁴¹² by Leon Schlosberg, part 1 containing notes by Dr. Scheyer which compare the Arabic original and Tibbon’s translation (London, Maimonides 1851–79, title in Hebrew), parts 2 and 3 (Hebrew and Latin titles), 1876, 1879.⁴¹³ [Paris, BN héb 682 is still the only complete copy of al-Ḥarizi’s translation known to exist. It is written in a square Spanish hand of the thirteenth century; the copyist may be Yom Tov ben Shemaia, who also wrote a marginal note on f. 14a. Rome is mentioned but there is no clear indication that the manuscript was copied in that city. There are several long (but unfortunately faint) long marginalia that have not been studied. There is another fragment (five folios) of al-Ḥarizi’s translation in New York, M. Lehmann MA 13 , containing parts of Guide II, 32–33, 38–40. Interestingly enough, someone has copied nearly all of the Judaeo-Arabic text of II, 38 into the margin. Finally, Tübingen, UB Ma. IV 2 , a copy of Ibn Tibbon’s translation executed in 1343, records several times al-Ḥarizi’s translation of a particular phrase in the margins.]

According to the foreword the translation was commissioned by several Provençal scholars (Marseille, according to the manuscript cited by David Conforte, f. 12). But al-Ḥarizi says expressly in his Divan⁴¹⁴ that he had translated the Moreh in Spain (ספרד) for one of the noblemen (נסיכים), whom he calls Joseph in the dedicatory poem – we shall see, however, that he dedicated his Divan successively to four different persons in different countries.⁴¹⁵ – Al-Ḥarizi was asked to translate the book in simple, elegant, and easy to understand style. The intelligent scholar (Samuel) had in his translation “intentionally made its meaning obscure.” Judah was thus “forced” to translate it (again), promising, however, not to divulge any of the “secrets”, nor to intentionally explain anything. He prefaces the translation with two gates (chapters), one of them explaining every foreign word in alphabetical order, the other giving a table of contents. For the moment we shall dwell upon the second gate.

Judah’s table of contents was added to Samuel’s translation in the manuscripts (under Judah’s name) already in the first edition, because Samuel did not replace it with another one. It is, however, abbreviated in several places, e.g., in I, 7 (ילד) where Judah had expressly remarked that Maimonides does not quote two biblical verses (Deut. 32:18 and Ps. 2:7) which should be interpreted as metaphors.⁴¹⁶ This |429| enraged Samuel, as we have seen, and probably more so the orthodox enemies of the school of Maimonides. Nevertheless – and this is an instructive example of how pseudepigraphy works – Samuel’s name could still be put in the place of Judah’s in Leiden, BR Cod. Or. 4800/1 , which has מפתח (clavis, key) or פתיחה (introduction),⁴¹⁷ as in ‹Munich, BS Cod. hebr. 401/17 ›, and similarly הקדמה (proemium) and פתיחה in Berlin, SPK Or. Fol. 1057/13 . In fact its proper place is in the beginning, not the end, of the book, which is where it is in Schlossberg’s edition. This edition agrees with the manuscript mentioned above, also concerning the term פתיחה at the end.⁴¹⁸ Even Narboni (to I, 59) agrees with Samuel’s remarks concerning the register, which he calls בכוונותיו ובאוריו על פרקי זה המאמר. In the edition of 1553 it is also called שער בכוונות הפרקים. Incidentally, the register in Vatican, BA Cod. ebr. 262 and in the editions 1553 ff. name Judah as אלחפני. The origin of this manifest mistake is not known; perhaps it is a copyist’s error.⁴¹⁹

In al-Ḥarizi’s translation, chapters 26 and 27 of part I are conflated into one, as they were also in Samuel’s first version. The sum total therefore counts 177 [chapters], which is expressed by ג“ן עד“ן alphanumerically. Contrary to this, the propositions at the beginning of part III are counted as chapter 1 in Schlossberg’s edition. Some manuscripts combine chapters 9 and 10 of part II.⁴²⁰

§248. Al-Ḥarizi’s glossary drew a lot of criticism from his rivals. For us it provides very interesting documentation of the status of the Hebrew language at the time, and thus reveals the influence the scientific translations had on its development. A comparison of both glossaries will lead us to the salient points. First of all, we are taken aback by the awkward arrangement. The entries are arranged haphazardly, sometimes according to the [first letter of the] root, but also according to the first (and second) letter. Samuel seems to be influenced in this regard by his predecessor, but at least he says something about it in his introduction. His arrangement mostly follows the order of roots, while al-Ḥarizi places under the letter ה many words in which the initial ה is not part of the root. Al-Ḥarizi’s vocabulary has a greater number of purely Hebrew or Rabbinic (Talmudic) words which, one would suppose, should be sufficiently familiar to readers of a book of this sort (e.g., נגר) so as to lend themselves to the correct interpretation of their particular nuance in a given context. Al-Ḥarizi, “the poet,” however, designed his translation for the general public; his entries, most of them consisting of one or two lines, offer a short definition. Samuel, on the other hand, aimed at a limited circle of scholars. He confined his glossary to technical terms or neologisms, but he offers substantive discussions of them. Al-Ḥarizi is more of a purist and stylist, and thus the Arabic does not dominate his own writing.⁴²¹ The only Arabic words in his |430| glossary are: קוטר and קוטב, which he had already found in Samuel. He does not translate the Arabic word 'נילג, and has in I, 92 סופסטאנין and in III, 49 עצ‘לה (cf. III, 250, in the French translation, Maimonides 1856). He imitates the Arabic מקדמה in his מקודמת s. v. הקדמה, as well as three times in I, 85; hence one has to correct 85b, line 7 מוקדמות accordingly; but since the pu‘al of qdm was not used, other authors have the form מוקדמה, e.g., in Falaquera, beginning of I, 66, where al-Ḥarizi uses הקדמה. The word אנתחל, to falsely appropriate for oneself, or to plagiarize (II, 61, 62), is translated as יתנחל and מתנחל, which do not exist in Hebrew and are not listed in the glossary.^421b – Ḥarizi translates the name Abū NaṢr (al-Fārābī) אבי ישע (II, 25, 31, where we find בהשגת דרכים for his commentary to the Physics! III, 28; I, 89 אבי נצר is a gloss; I, 95 ן‘ יושע!).⁴²² With regard to Arabic personal names he sometimes remarks (e.g., II, 46 Ibn WaḤshiyya) that it is a ישמעאלי.

More important is an erroneous reading of the famous Book of Nabataean Agriculture by Ibn WaḤshiyya that can be accounted for by Arabic orthography⁴²³ and which leads to a number of literary discussions.⁴²⁴ Al-Ḥarizi read קטביה֞ for נבטיה֞, changing the Nabataeans to Egyptians. Another textual corruption is found in III, 44: for עבודת האכזרים המצריים, read האיכרים; III, 46 displays only ס‘ העבודה המצרית. NaḤmanides quotes this passage in two works, following the translation of al-Ḥarizi.⁴²⁵

Let us return to the glossary and discuss the following words: הסיפא, an antonym for ןינק (Tibbon הלילש and ןינק),^425b העצה, שגרה (שגרומ, Tibbon שוח and שחומ), ךתח, ןידה ךתח II, 5b, 11; רודכ = הטולפ (ballota?), read רודיכ (cf. II, 15 ff.), םירזואמ, אקיסומ, קצומ explained by דומע, as the following םיקוצמ, read םיקצומ? The Book of Conic Sections is named (I, 91) (תאטור'כמ) םיטורחה םיקוצמה, but Arabic זכרמ is translated as דומע (II, 17 chapter 12 = 13), םירמסמ (poles, cf. בטק), עלצ; תוינק (in the ethical sense). The following words and forms are not found in the glossary: םילשמתמ I, 84, םקיר (there is no word at all under the letter 'ר), רדסו המושת לעב and םירקיעה ילעב (ןיילוצא) I, 85, בבוסמ 89b, תבבתסמ העונת II, 16; ךתחנ תפומ II, 17 chapter 12;⁴²⁶ ץופי ('ץיפי) II, 18 for עפשי (cf. the Glossary s.v. עפש), יעבטה עדמ II, 19; – to the word תבאושה ןבא (II, 19) al-Ḥarizi adds טינגמ ארקנה. The word טאוטו (III, f. 64 Arabic) is translated by Tibbon as םירחא םיצרש; probably he did not know what to do with it; al-Ḥarizi (III, 44) explains it as טברקשא (read שיברקשא?) “scarab”, as Munk has it in his translation (230). A close study of this translation, which I could not undertake because of the fine print, might yield some more remarkable details. Here it may suffice to refer to the parallels |431| between al-Ḥarizi and NaḤum and the anonymous translator of Joseph Ibn Zaddik’s Microcosm (§238).

A final remark on the rivalry between the translators. In 1191 Maimonides wrote a treatise on the resurrection, under the title אגרת or מאמר תחיית המתים in Samuel b. Tibbon’s translation. It has been printed repeatedly after 1629.⁴²⁷ We do not know the date of this translation. A short while ago, two prooemia to it were discovered in a ms. (now Oxford, Bodl. Ms. Opp. Add. Qu. 163 )⁴²⁸ which are preceded by the verse מאתך השר sent by al-Ḥarizi (according to his Diwan) from Spain to Maimonides.⁴²⁹ Al-Ḥarizi begins his short foreword by saying that Maimonides had addressed the following letter (מגלה, for reasons of rhyme?) in Arabic to the (Jewish) communities and that the translator (whose name is not given here) has not rendered the contents clearly⁴³⁰ – again the same criticism as in the foreword to the Moreh. – This translation was retranslated into Arabic, and al-Ḥarizi translated this Arabic version for Meir b. Sheshet ha-Nasi back into Hebrew.⁴³¹ After this comes a proemium by the Arabic translator Joseph b. Joel, a friend of Samuel, who had furnished him with the Hebrew translation. Joseph remarks that Hebrew is not sufficient for rendering Arabic writings, and therefore the translation is inevitably unclear. Thus, he says, he was asked to retranslate it into Arabic. The main principle of translation, generally accepted for quite some time, consists in rendering the simple meaning of the text by the appropriate words of the target language; if one is able to put it word for word, then one has risen to the summit of the art. Meaning is the main thing; wording is of secondary importance. What comes next makes us wonder whether we are hearing the Arabic translator, and he is the author of the versified foreword in Hebrew, or whether al-Ḥarizi himself here as well has translated the Arabic foreword. The examples, for their part, are in Hebrew. The word ענין (probably Arabic מעני) must be translated by מעמד, כוונה,ענין and דבר; and the word ג'סם) גוף?) by גויה,גשם (n. 323^c), גרם, עצם and גוף. What perplexes us completely, however, is the fact that after this proemium there follows none other than the translation printed under Samuel’s name, which is closer to his other works than to al-Ḥarizi’s; thus the latter’s was removed from his translation (or found separately?) and placed in front of Samuel’s! A quotation in NaḤmanides could be taken from that translation, but it does not seem to be verbatim at all.⁴³²

[This confusion led Finkel 1939 to conclude that the story of the re-translation was a work of fiction; Baneth, however, argued that essentially it was correct, and that the passage by R. Joseph was itself translated from the Arabic by al-Ḥarizi as an introduction to his translation in Baneth 1940. In 1980 A. David discovered an unknown translation of the Treatise on Resurrection in Jerusalem, JNUL 3942 8 ^o, ff. 1–13, and speculated that it was that of al-Ḥarizi 1978–79. Halkin substantiated the speculation and published the translation on the basis of the Jerusalem and Oxford manuscripts 1980.]

§249. Al-Ḥarizi’s Moreh sometimes renders the meaning of certain passages more accurately than Samuel’s (even where the latter followed more correct readings). Nonetheless, it did not escape the criticism of the friends and enemies of the book,⁴³³ nor was it entirely supplanted by its rival. Apparently it was read more extensively in northern Spain |432| where it is perhaps quoted by NaḤmanides. (See n. 438.) Abraham Maimonides (around 1235) already considered al-Ḥarizi’s translation faulty and textually corrupt; we do not know whether this judgment was based on his own criticism or on hearsay.⁴³⁴ Shem Tov Falaquera⁴³⁵criticizes it in a letter dated 1290 and printed anonymously; he⁴³⁶ talks about both translations, without, however, naming either author. The first one (Tibbon’s), he says, has only a few mistakes, and had the learned translator had more leisure, he would have corrected them⁴³⁷; the second translation (al-Ḥarizi’s), however, contains numerous errors and even the correct parts do violence (to the text), often perverting the meaning into its opposite sense. The translator aimed to explain, but instead he “put a big stone on the bridge” (a play on words), “and I say that whoever receives this translation in his tent, harbors an injustice” (an allusion to Job 11:14).

A number of notes to the Moreh, up to part I, chapter 14, under the name of the mystic Joseph Gikatilia (read Chiquitilla) were edited in 1574. They correct, right at the beginning and in fol. 20, line 2, al-Ḥarizi’s translation, on the basis of the Arabic text⁴³⁸; fol. 22 has the following passage: “The poor man had no clear eye in the science, he thrust himself forward to a place flaming with fire which was not seemly for him.”

Al-Ḥarizi’s translation was still studied in the fifteenth century by Asher Crescas⁴³⁹; but the supposition that Isaac Abravanel took his paraphrased texts from this translation has proven to be erroneous.⁴⁴⁰

A critical assessment of the editions of both translations will arrive at the same judgment as that of Pococke, who summarizes his verdict in few words⁴⁴¹: “Versio (Harizii) illi ab Aben Tibbon factae postposita fuit, non quod illa Tibbonidae elegantior, sed materiae congruentior fuerit,” etc.

§250. (The old Latin translation) Al-Ḥarizi’s translation has a historical significance which escaped even Munk, who does not even mention it in the foreword to his Guide, although he quotes from it repeatedly in his notes.

Agostino Giustiniani edited (1520) a Latin translation under the title Dux neutrorum sive dubiorum (the title page has Dux seu director dubitantium aut perplexorum) 1520;⁴⁴² this work has been attributed to the editor himself or to Jacob Mantinus. Certain quotations from the Guide in some Christian authors of the thirteenth century show, however, that already in that time a Latin translation existed.⁴⁴³ I have demonstrated that traces can be found as early as the middle of the thirteenth century in southern Italy. |433| Moses b. Solomon of Salerno⁴⁴⁴ composed, probably between 1240 and 1250, a commentary to the Moreh which he reworked, but did not complete. [According to Caterina Rigo, Moses wrote the commentary in the 1270s.] Only a part of the second recension is known to exist, with notes by his son Isaiah⁴⁴⁵ in the manuscripts.⁴⁴⁶

Cambridge, UL Add. 672 ; Florence, BM-L. Ms. Plut.II.11 ; ‹London, BD and BM 40/5 ›; ‹Munich, BS Cod. hebr. 370 ›; ‹Munich, BS Cod. hebr. 60/1 ›; Oxford, Bodl. Ms. Opp. 576 ; Paris, BN héb 687 ; Parma, BP Cod. Parm. 2435 ; Parma, BP Cod. Parm. 2910 ; Parma, BP Cod. Parm. 3162 ; St. Petersburg RNL Ms. Evr. I 482.

The author of the commentary had received in the course of his work, perhaps between the first and the second recension, the glossary of the translator Samuel, which he labels “Introduction,” פתיחה or הקדמה. His commentary is in the main a paraphrase with explanations or translations of single words into the vernacular (i.e., Italian). However, he read the Latin translation together with Nicolao da Giovenazzo, probably Nicolo Paglia, Nobile di Giovenazzo, the founder of the Dominican monastery Santa Croce in Trani.⁴⁴⁷ [See Sermoneta 1969–70.] On the other hand, we have a report concerning a remark of the Emperor Frederick II on a passage in the Moreh.⁴⁴⁸ The person to whom this remark was addressed was, according to some manuscripts, none other than al-Ḥarizi. The place where this occurred is also named, perhaps Tropea in Calabria. I do not, however, believe that al-Ḥarizi visited this province in the course of his travels. Other sources name Samuel Ibn Tibbon, that is to say, one translator instead of the other. My surmise⁴⁴⁹ is that this translation was commissioned by this same Emperor, for whom Jacob Anatoli translated other works in Naples (§19); and Amari agrees that this is rather plausible.⁴⁵⁰ Finally, Perles who studied one manuscript of the Latin translation in Munich, showed that it follows the Hebrew translation of al-Ḥarizi, but in consultation, according to him, with the Arabic text and with the cooperation of a learned Jew.⁴⁵¹

Moses b. Solomon of Salerno, Commentary on the Guide of the Perplexed

Cambridge, University Library Add. 672 (SCR 701) (IMHM F 17001), 1–139.

Florence, Biblioteca Medicea Laurentiana Plut.II.11 (IMHM F 17658), 1–132.

London, Beth Din & Beth Hamidrash 40 (IMHM F 4708).

Munich, Bayerische Staatsbibliothek Cod. hebr. 60 (IMHM F 1140), 1–329.

Munich, Bayerische Staatsbibliothek Cod. hebr. 370 (IMHM F 1606), 1–296.

Oxford, Bodleian Ms. Opp. 576 (Ms. Opp. 1163) (Neubauer 1261/1) (IMHM F 22075G), 1a–313b.

Paris, Bibliothèque Nationale héb 687 (ancien fonds 234) (IMHM F 11565), 1–187.

Parma, Biblioteca Palatina Parm. 2435 (De Rossi 1369) (IMHM F 13439), 1–195.

Parma, Biblioteca Palatina Parm. 2910 (De Rossi 1071) (IMHM F 13803), 1–49.

Parma, Biblioteca Palatina Parm. 3162 (De Rossi 106) (IMHM F 13902), 1–222.

St. Petersburg, Russian National Library Evr. I 482 (IMHM F 51349), 1–181.

The extant manuscripts of the Latin Moreh have not been investigated sufficiently;⁴⁵² perhaps they contain information that is as yet unknown. The translator allowed himself to omit passages that he deemed unimportant, but he also added some comments of his own.⁴⁵³ For particulars and some extracts which are proof of the inadequacy of the edition, we must refer the reader to the useful article of Perles. Wolf (3rd ed., 782) has already compared the two Latin translations of the introduction. He has called attention (779) to the titles Directio and Director neutrorum, Directio Perplexorum, und Demonstrator errantium which we find in the writings of Raymund Martin (a contemporary of Albertus Magnus), and the baptized Jews Paulus Burgensis and Alfonsus de Spina. Perles remarks that Raymund partly follows the translation of al-Ḥarizi.⁴⁵⁴ Thus it was primarily through this translation that the learned Christian world |434| was first acquainted with the philosophy of “Moses Egyptius,” and not unprofitably, as M. Joel in his treatise on the connections between Maimonides and Albertus Magnus has shown (Joël 1863).⁴⁵⁵

[Scholarship on the Latin Guide has proceeded apace in the last century. The scholarly consensus is that there are three independent translations of at least parts of the Guide : De parabola ( Guide 3.29–30, 32–49) (completed 1222–23), De uno deo benedicto (Guide 2.1–2), considered by Kluxen 1954 to be completed around 1240, and Dux neutrorum sive dubiorum , also from around 1240. In addition, there apparently was a fourth translation, based on the Ibn Tibbon translation, to which Giles of Rome refers, but which is no longer extant. The Perles-Steinschneider-Sermoneta hypothesis, that the Latin translation of the Guide was made in southern Italy, perhaps as the result of Jewish and Christian scholars working in tandem, was challenged by Kluxen 1954 , who argued for a Provençal location and who connected the translation with the Dominican involvement in the Maimonidean controversy in the late 1230s. This hypothesis was argued against by Gad Freudenthal 1988 , 120–29. Schwartz 2002 , 46, speculates that the fourth translation originated from southern Italy.

Recently G. Hasselhof has questioned the attribution of De Parabola to Maimonides; rather he considers it an “adoption of Maimonides into a different context,” 2001 , 262. He does not appear inclined to accept the southern Italian provenance of the translation and concludes that an analysis of the manuscript tradition, of the sort first proposed by Steinschneider, leaves the question of provenance open. See also Hasselhof 2002 .]

We think that we should conclude this long entry on the most important Jewish-Arabic work by indicating a few recent publications (apart from those by Munk and Friedlaender already named) that analyze the Moreh, or explain the Maimonidean philosophy basing themselves mainly on the Moreh, or treat a particular problem in it.

[Steinschneider proceeds to list works on Maimonides and his influence by Scheyer 1845 , Joël 1859b , Foucher de Careil 1861 , Rubin 1868 , Eisler 1870–83, Kaufmann 1877a , Münz 1887, ⁴⁵⁶ Holub 1884 , the latter of which he knows only by the German title.

Needless to say, an enormous amount of literature on Maimonides has appeared in print since 1893. Specialized bibliographies on a number of topics have been prepared by Jacob Dienstag. For a partial listing of studies through 1964, one can consult sections on Maimonides in the English translation of Guttmann’s Die Philosophie des Judenthums 1964 , 405–7. In the second part of Vajda’s important annotated bibliography of studies in medieval Jewish philosophy from 1950 to, 1973 Vajda 1972, 1974 , Maimonidean studies are discussed on 206–22. Other bibliographies help to bring the field up-to-date: Lachterman 1990 (English studies from 1950–1986); Bitya Ben-Shammai 1991 (Hebrew studies from 1965 to 1990); and Kellner 2004 (English studies from 1991 to 2004). There is also a very extensive bibliography – as well as a detailed discussion of the translation and transmission – in Schwartz 2002 , available online at http://press.tau.ac.il/perplexed/ ].

§251. 2. Logical Terminology. Maimonides wrote, certainly in his youth, perhaps still in Spain (before 1160, earlier, that is, than his 15th year), an explication of the terms employed in logic, obliging the wish of a noble theologian who was well versed in Arabic. The treatise contained fourteen chapters, and at their end the explicated terms are enumerated.⁴⁵⁷ No complete manuscript of the Arabic original is known to exist. Paris, BN héb 1202/5 (in Hebrew letters) contains only chapters 1–7 under the title מקאלה פי צנאעה אלמנטק, Treatise on the Art of Logic. I discovered chapters 7 and 8 and a fragment of chapter 11 in a book, Bodl. Hunt. 593 (Hebrew letters). Some |435| passages of it are published.^457b [These fragments were taken from the binding of Bodl. Hunt. 593 and now are in Oxford, Bodl. Ms. Hunt. 632 . They were catalogued by Neubauer as Bodl. Opp. Add. Qu. 151. According to B. Richler 1987 , the Oxford pages were separated from the manuscript now at Paris. The fragments were published by I. Efros in 1938 as part of his edition of the work 1937–38. Later, M. Türker discovered two complete manuscripts of the work, written in Arabic letters, in Ankara and in Istanbul, and published the text twice Türker 1959–60b, 1961 . Efros 1937–38 then published a complete critical edition of the Arabic text, in Hebrew characters, based on Türker’s readings and the Paris and Oxford excerpts.

Recently, Herbert A. Davidson has questioned the attribution of the work to Maimonides in Davidson 2001, 118–25, and Davidson 2005, 313–322. As Davidson points out, the first to do so was J. Reifmann; see n. 447.]

The Hebrew translation under the title of מלות ההגיון or ביאורr(פירוש) is extant in many manuscripts; there are no less than ten in Paris alone. Some manuscripts bear the abbreviated title הגיון (logic), and even הגיון קצר (concise logic). The printed catalogues have not correctly identified the copies of our treatise that are found in Parma, BP Cod. Parm. 2445/2 and Paris, BN héb 1005/3 (הגיון אחר קצר) [a copy of which appears in Paris, AIU 432 ].⁴⁵⁸ Most manuscripts, as well as the numerous editions since 1552, do not name the translator. Some manuscripts end with an epigraph saying that Moses Ibn Tibbon translated the book in Kislev 5015 (Nov. 14–Dec. 13, 1254), from a faulty and deficient manuscript of the original, a statement which we know already to be merely an empty phrase.

[In addition to the translation by Moses Ibn Tibbon, there is one by AḤituv, a thirteenth-century Jewish physician in Palermo, and one by Joseph Ibn Vivas or Joseph Lorki (see below). The AḤituv translation was first edited by M. Chamizer 1912, but later corrected on the basis of more material by Efros, who published an edition of the Arabic text (then extant), the three medieval Hebrew translations, and an English translation in 1937–38. Langermann 1995 , 381, noted another copy of the AḤituv translation (incomplete) in Moscow, RSL Günz. 1020/4 . Efros based his edition of the Ibn Tibbon translation on the first editions and on eight mss; there are close to eighty manuscripts of this translation, most of which have not been studied, not to mention marginal glosses on other manuscripts. The version of Ibn Vivas is extant in Paris, BN héb 1201/4 ].

Mss. of Hebrew translations of Maimonides’ Treatise on the Art of Logic other than by Moses Ibn Tibbon

Trans. Ibn Vives

Paris, Bibliothèque Nationale héb 1201 (IMHM F 31369), 63a–75b.

Trans. AḤituv ha-Rofe

Moscow, Russian State Library Ms. Guenzburg 1020 (IMHM F 48110), 25b–8b.

New York, Jewish Theological Seminary of America JTSA 2278 (IMHM F 28531), 16 fols.

St. Petersburg, Russian National Library Evr I 419 (IMHM F 52721), 4 folios.

Vatican, Biblioteca Apostolica BAV 349 (IMHM F 380), 2a–17a.]

Sebastian Münster edited (Basle 1527) a bad text⁴⁵⁹ under the title ס‘ ההגיון (!) with a Latin translation which, according to Richard Simon (1638–1721), does not contain one single passage that is correct. Richard Simon falsely attributed the translation to Samuel Ibn Tibbon.

The first editions are already accompanied by two anonymous commentaries whose date I have not investigated. Munich, BS Cod. hebr. 289/1 . See above, §18 [where Steinschneider, following Neubauer, misattributes this commentary on Averroes’ Epitome of Logic to Narboni, rather than to Mordecai Nathan. For Moses Narboni’s commentary on the Millot ha-Higgayon , see Hayoun 1983 ], and the end of §36, where a commentary found in Paris, BN héb 1061/11 may be attributed to Albo. In the fifteenth century the learned Mordechai Khomatiano (or: Komatiano/Comtino) composed in Turkey a commentary according to the wish of his student Isaac ẓarfati whom he addresses with the same words as does Maimonides in the Moreh. This Isaac ẓarfati is probably the author of a letter, published in print.⁴⁶⁰ Khomatiano’s commentary is extant in several manuscripts,⁴⁶¹ among others in Paris BN héb 681 where in the end an alphabetic register of the terms that have been discussed is appended. The catalogue attributes this to Khomatiano; however, it does not figure in the manuscripts that have been studied nor in the edition by D. Slutzki (Warsaw 1865). Khomatiano, a famous mathematician, was well versed in logic.

Frankfurt, SUB oct. 55 ; Leiden, BR Cod. Or. 4794/10 ; Moscow, RSL Günz. 469/1 ; New York, JTS Ms. 2407/7 ; New York, JTS Ms. 2875/4 ; New York, JTS Ms. 3409/8 ; Oxford, Bodl. Ms. Mich. 214 ; Oxford, Bodl. Ms. Opp. 519 ; Paris, BN héb 681/4 ; Parma, BP Cod. Parm. 2446/4.

Since this small treatise had already become famous by virtue of its author and, in addition, it was adequately congruent with the principal rules of logic (which, among the Jewish scholars of the North, had been supplanted by the degenerate method of Talmudic studies), Moses Mendelssohn composed a Hebrew commentary. In his introduction he recommends the study of Aristotelian logic and justifies it by referring to Maimonides. The first two printings of this commentary (1761 and 1765) do not display the name of the author; it has been frequently reprinted ever since.⁴⁶² M. S. Neumann published a German translation. Another one, with a register of terms (actually by R. Fürstenthal?), was issued by Heilberg in 1828; the translation of W. Heidenheim remains unpublished (manuscript Bodl. Mich. 83). P. Heilprin published (1846) a text, allegedly based upon the rules of textual criticism, but which is in fact abysmal. |436| [The standard edition of Mendelssohn’s commentary was edited by H. Borodianski in volume 14 of the Jubilaeumsausgabe of the Gesammelte Schriften 1929 , [23]–119. J. Dienstag published a comprehensive bibliography of editions (28), translations (several into German, one into Italian, Russian, French, and English), commentaries, and studies on the Logical Terms in Dienstag 1960 .]

The terminology of this small treatise has come to dominate in the Hebrew literature. Thus it achieved in Hebrew letters the goal that had been the original purpose of the Arabic writings. It has often been confused with the logical compendium of Averroes (§17).

Paris, BN héb 1201/4 contains our treatise, translated from Arabic, according to the title, by Joseph b. Joshua Ibn Vives from Lorca, or Joseph Lorki.⁴⁶³ This translation is dedicated to Ezra b. Solomon Gatigno (or Gatinho), a well-known writer who lived in Saragossa and Agremont (1356–72)⁴⁶⁴ and refers elsewhere to his late teacher Joseph Ibn Vives, certainly our translator.⁴⁶⁵ According to Simonsen,⁴⁶⁶ however, Joseph’s work is not a new translation, but a transcript amended by him, if not simply a text plagiarized from Moses Tibbon – which is surprising since Tibbon’s translation should have been already well-known at that time. If used cautiously, the manuscript might serve to amend the existing editions. Is there any relation between this recension and the one published by Münster? [Efros 1937–38, 10, calls Simonsen’s judgment “unfair”; according to him, while the author did have the Ibn Tibbon text at his disposal he translated directly from the Arabic. There are close to a hundred mss. of the Hebrew translations of Maimonides’ treatise on logic, the vast majority of them by Moses Ibn Tibbon. As for the others,

Mss. of Mordecai Khomatiano’s commentary on Maimonides’ Treatise on Logic

Frankfurt, Stadt- und Universitätsbibliothek oct. 55 (Merzbacher 13) (Carmoly 227) (IMHM F 34036), 1–18.

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4794 (Warner 56) (IMHM F 27912), 251a–66b.

Russian State Library Ms. Günzberg 469 (IMHM F 43040), 1a–64a.

New York, Jewish Theological Seminary 2407 (IMHM F 28660), 85b–95b.

New York, Jewish Theological Seminary 2875/4 (IMHM F 31713), 192a–220b.

New York, Jewish Theological Seminary 3409/8 (ENA 2771) (IMHM F 32094), 192a–220b.

Oxford, Bodleian Ms. Mich. 214 (Ms. Mich. 81) (Neubauer 2187/3) (IMHM F 20469), 129a–68b.

Oxford, Bodleian Ms. Opp. 519 (Ms. Opp. 1026) (Neubauer 1911/8) (IMHM F 18844), 209a–28b.

Paris, Bibliothèque Nationale héb 681 (ancien fonds 223) (IMHM F 11559), 104b–49b.

Parma, Biblioteca Palatina Parm. 2446t (De Rossi 556) (IMHM F 13450), 90a–130a.

§252. 3. Treatise on Unity.⁴⁶⁷ No copy of the Arabic original of this small theological and ethical compendium in three chapters, followed by a chapter addressed to a friend,⁴⁶⁸ is known. [Langermann recently found a fragment of the Arabic original in New York, JTS Ms. 9069/4 , 19b, and published it as an appendix to 1996c .] But for those who make their judgment only after they have studied the matter there is not the least reason to doubt the authenticity of this treatise in which the author refers to his Moreh.⁴⁶⁹ It is already quoted by the Hebrew title of מאמר היחוד (perhaps it carried no Arabic title, which would have been something like מקאלה פי תוחיד) by the translator of the Mishna commentary (Ord. II [ = Seder Mo’ed], around 1287).⁴⁷⁰ It contains an outline of the fundamental philosophical ideas upon which Maimonides had expounded in Hebrew in the first two books of his great Talmudic work. The passages quoted from the Holy Scripture were paraphrased in Arabic in the original, but the Hebrew translator Isaac b. Nathan (middle of the fourteenth century) (§192) did not restore the [original] texts. Instead, he translated the Arabic quotations literally, a fact that was expressly noted by one copyist.⁴⁷¹ This title is signalled in:

Hamburg, SUB Ms. Cod. hebr. 310b ; Mantua, CI Ms. ebr. 78b ; Munich, BS Cod. hebr. 150/1 (with better readings)⁴⁷²; New York, JTS Ms. 2274/6 ; New York, JTS Ms. 2274/6; New York, JTS Ms. 2407/7 ; Oxford, Bodl. Ms. Mich. 491b (an 1840 copy of Hamburg, SUB Ms. Cod. hebr. 310b); St. Petersburg, RNL Ms. Evr. I 357 ; Vatican, BA Cod. ebr. 170/3 (without the Supplement); Vatican, BA Cod. ebr. 171/13 .

I have edited the small treatise, first on the basis of a copy in Hamburg made by Dukes (issued in Berlin in only a few copies in 1846), and again in Maimonides 1847, together with a small treatise of Abraham Ibn Ezra, under the title תורואמה ינש (The Two Lights). The text is accompanied by Hebrew notes which cast light upon the style and some other details as well, and by a study in German. It is preceded by a letter of S. L. Rapoport who (erroneously) |437| supposes this treatise to be identical with a pseudepigraphal work (אצמנה 'ס) ^472b

The translator has retained some Arabic words from the original: p. 37 אלכ‘יאר (ms. for םיאושיקה), עקפלאו ˝האמכלא and רמ’תלאו ירומלא (ms. for םירמתו המלא); he translates אלמלאء by המלוא (p. 36)⁴⁷³ and seems to have translated חכים twice (18, 31) erroneously by רופא. The small treatise is full of syntactical Arabisms. For a number of passages, however, where the edition presents us with readings which can hardly be called Hebrew, the Munich manuscript removes all difficulties. [Some select quotations and a discussion of the translation are available now in the study of Langermann 1996c . Throughout his long career, Steinschneider never flinched from his firm belief in the authenticity of this treatise, despite the doubts expressed by Graetz; nevertheless, the scholarly consensus is to regard the Treatise on Unity as a pseudepigraph. Langermann’s article reopens the question.]

Mss. of the Treatise on Unity attributed to Maimonides

Hamburg, Staats- und Universitätsbibliothek Cod. hebr. 310 (Hamburg 256/2) (IMHM F 1095), 38a–48a.

Mantua Comunita Israelitica Ms. ebr. 78 (IMHM F 864), 85a–90a.

Munich, Bayerische Staatsbibliothek Cod. hebr. 150 (IMHM F 1168), 1–15.

New York, Jewish Theological Seminary Ms. 2274 (IMHM F 28527).

New York, Jewish Theological Seminary 2407 (IMHM F 28660), 85b–95b.

Oxford, Bodleian Library Ms. Mich. 491 (Mich. 575) (Neubauer 1317/2) (IMHM F 22131), 23a–31a.

St. Petersburg, Russian National Library Evr. I 357 (IMHM F 50948), 6 fols.

Vatican, Biblioteca Apostolica Cod. Vat. ebr. 170 (IMHM F 232), 116a–22a.

Vatican, Biblioteca Apostolica Cod. Vat. ebr. 171 (IMHM F 8630), 79a–88a.]

§253. 4. A small Arabic treatise on felicity, in two chapters, is found in Paris, BN héb 719/4 , according to the catalogue; it appears to be only a fragment of a more comprehensive work. As far as I know, nobody has studied this ms. which is not noted in Steinschneider 1852 either (col. 1917).

This treatise was translated into Hebrew (by whom, when?) and printed under the title פרקים בהצלחה (Chapters on Felicity) in 1567. Mordechai Tama included it under the inaccurate title פרקים בהצלחה in the collection of legal opinions, translated by him from Arabic in 1765,⁴⁷⁴ which led to the belief that Tama himself was the translator.

[A manuscript of the Hebrew translation was identified by Fritz Baer in a codex then at Pamplona, now New York, JTS Ms. 2341/11 . Along with the copy of the Judaeo-Arabic original at Paris, it served as the basis of the doctoral dissertation of H. S. Davidowitz. D. H. Baneth considerably revised this thesis for publication, and the two texts were later published in facing columns under their joint authorship, along with an introduction and notes (Maimonides 1939). The Hebrew translator is named as “R. ZeraḤiah ha-Levi” and, after deliberating the various alternatives, Davidowitz (xxvii–xxx) leans towards identifying him with Zerahia ben She’altiel Ḥen; the main difficulty remains the appellation ha-Levi, which is not added to She’altiel Ḥen’s name anywhere else. A late copy (1781) of the Hebrew translation is London, London School of Jewish Studies 20. ]

The author talks about his Moreh and addresses his student (according to S. Rapoport, Joseph Ibn ʽAqnin).⁴⁷⁵ From this one may infer that this little treatise was meant to be a kind of appendix to the Moreh. There is nothing in it that proves the attribution to be false. The mystical shade which lies over this rather theological, at some points rhetorical, analysis does not directly contradict Maimonides’ spirit. This treatise (מאמר אחד קצר שחבר) is quoted already in the middle of the fourteenth century by Joseph b. Elazar (explaining Genesis 5:34) and somewhat later (as אגרת ההצלחה) by Don Benveniste b. Labi in a letter, so far unpublished.⁴⁷⁶

[Steinschneider’s claim that the treatise is authentic was strongly supported by W. Bacher 1896 . In support of the authenticity of this writing, Bacher (279, n. 6) even cites a parallel from Ma’mar ha-YiḤud , another tract that Steinschneider alone viewed as authentic! Davidowitz in Maimonides 1939 , however, mustered a long series of philological and other arguments (xiv–xxii) against the attribution to Maimonides, and since his publication, the scholarly consensus is that the tract is a forgery. For further bibliography see Dienstag 1986 .]

§254. 5. The Eight Chapters. Maimonides wrote a commentary in Arabic to the entire Mishnah. The treatise Abot (a compilation of ethical sayings) is preceded by an introduction [in eight chapters] on the soul, its faculties and their use for the attainment of a goal. This introduction was called “Eight Chapters” (שמונה פרקים), after their number.

It was, together with other parts of that commentary, edited in Arabic (with Hebrew letters) and in Latin in Porta Mosis by Edward Pococke (Maimonides 1654), and recently with German translation and notes, by M. Wolff (Maimonides 1863, title only with Arabic letters). Numerous corrections were made with the help of H. L. Fleischer.

The commentary to Avot was translated into Hebrew by Samuel Ibn Tibbon. His translation found a place in the Italian prayer book (Maimonides 1484), and again, in a separate printing, small quarto,⁴⁷⁷ in |438| the Greek prayer book (Romagna, around 1523 (Maimonides 1523), in an edition of the whole commentary to the Mishna and, finally, in all editions of the Talmud and the Mishna with the commentary of Maimonides.⁴⁷⁸ C. C. Uythage translated the commentary to Abot (Maimonides 1683) into Latin; a translation by Jacob Mantinus remains unedited; and there are several German translations from Hebrew of which we name only that by Gotthold Salomon (in Maimonides 1809). Samuel’s preface is extant only in part, in two different recensions. According to one of them, Samuel was asked by scholars of his hometown Lünel to translate this commentary, just as they had asked him to translate the Moreh. The date of the translation is given only in Parma, BP Cod. Parm. 2303/6 , which is Tebet 963 (Tebet 1 = Nov. 28, 1202). The eulogy of the deceased appended to the name of Maimonides (he died in December 1204) may have been added at a later date.⁴⁷⁹

[Two translations into modern Hebrew may be mentioned: that of Y. Qafih Maimonides 1962, as part of his edition and translation of the entire commentary to the Mishna; and that of Y. Shailat, Masekhet Abot (Maimonides 1994). For an English translation and study, see Joseph I. Gorfinkle 1912; a more recent translation is by Charles Butterworth and Raymond Weiss in Maimonides 1983.] Translations of the remainder of Maimonides’ Commentary to the Mishna, including the introduction to the chapter called Hʿeleq (which is where he enunciates his famous creed of thirteen principles), as well the Letter to Yemen and the Letter on Apostasy, both of which contain some philosophy, are deferred to paragraph §554, along with the Book of Commandments and some miscellaneous correspondence.

§255. Saadia (Gaon) b. Josef from Fayyum (Pithom). [A full accounting of Saadia’s literary output, outdated in places but unsurpassed in scope or quality, is available in the book of Steinschneider’s student, Henry Malter, Saadia Gaon, His Life and Works 1921 .] His Arabic name, as it is cited by al-Mas‘ūdī, was Sa‘id Ibn Yaʽqūb al-Fayyūmī. Head of the rabbinical school in Sura near Baghdad (died 941/2),⁴⁸⁰ he owed his appointment probably to the renown which he had acquired in Egypt for the scope of his learning as well as his opposition to the Karaites. He was knowledgeable not only in the sciences of his community, but also in those practiced by the Arabs. Abraham Ibn Ezra calls him “the premier speaker on all subjects.” In fact, among the Jews he is the first author who is well-versed in the different branches of literature. Inter alia, he translated into Arabic and commented upon the entire Old Testament. His commentaries indulge in long digressions, inflated by polemics, on different scholarly subjects; but to our regret they are almost completely lost. [Scholars have debated and continue to debate the question whether Saadia commented upon the entire Torah. In any event, significant portions of his commentaries to Genesis and Exodus have been identified and published: the former by Zucker in Saadia 1984, the latter by Ratzaby in Saadia 1998. Small portions of a Hebrew translation of his commentary to Genesis have been identified in Judah Barceloni’s Commentary to Sefer Yeẓira ; see Malter 1921 , item c ; another passage was recently identified by Y. Tzvi Langermann in a thirteenth-century text; see his note in 2004 .]

We limit our discussion here to only two of his works. The first, written in Egypt, soon met with criticism in his fatherland. Two or three centuries later, however, the first Hebrew translation was replaced by a compilation of a quite different character, printed under his name. The second work marks for us an epochal event in Jewish philosophy; hence we place it first in our entry, but only after this remark: After Maimonides, Jewish philosophy in the East did not rise above the basic doctrines of the mutakallimūn. In fact, we observe that Saadia, building his arguments first from the senses, secondly, from reason, and lastly, from revelation, just as the mutakallimūn do in their theorizing, goes quite astray, even as far as endorsing the notion of compensation of animals for undeserved pain.⁴⁸¹ It does not follow from this, though, that he should |439| have acknowledged all of the consequences of those doctrines. – Let us now turn to the works themselves:

1. כתאב אלאמאנאת ואלאעתקאדאת (Book of Religions and Dogmas),⁴⁸² in ten treatises, written in 933. The title is interpreted wrongly in the Hebrew translation. [Steinschneider gives the title in German as Buch der Religionen und der Dogmen . He makes it clear that this is the correct meaning, i.e., that Saadia’s book treats of religions. All other modern scholars follow the understanding of the Hebrew translator: Munk, e.g., in his Mélanges 1859 , has Livre des croyances et des opinions ; Samuel Rosenblatt called his translation from the Arabic The Book of Beliefs and Opinions (Saadia ben Joseph 1948) and Alexander Altmann entitled his abridged edition, translated from the Arabic, The Book of Doctrines and Beliefs (Saadia ben Joseph 1946 ). Finally, Yosef Qafih, in his edition of the Judaeo-Arabic with facing modern Hebrew translatio n (Saadia ben Joseph 1970 ), points out that the full Arabic title is כתאב אלמכ'תאר פי אלאמאנאת ואלאעתקאדאת, The Book of the Choicest Beliefs and Opinions ; his Hebrew title reflects this (ספר הנבחר באמונות ובדעות), though he gives the traditional title afterwards in parentheses.] ⁴⁸³ Until recently, only one copy of the original was known, Oxford, Bodl. Ms. Poc. 148 ; however, the Petersburg library has just now acquired another manuscript of the Judaeo-Arabic text (Firk. 627, see the following paragraph). [There are currently only two fragments of the work in the Firkovich collection: St. Petersburg, RNL Evr. Arab I 3038 and St. Petersburg, RNL Evr. Arab I 3084 . Oxford, Bodl. Ms. Or. 311 is yet another copy.] Joseph Gagnier published a specimen of the original, along with the Hebrew and his Latin translation (Oxford 1711), but even the Bodleian Library does not have a copy. S. Landauer (Saadia ben Joseph 1880b) edited the text on the basis of the two manuscripts, printing it in Arabic letters. Goldziher corrected a number of erroneous transcriptions and other mistakes,⁴⁸⁴ stressing all the while the importance of the book for the history of Arabic philosophy.

§256. Saadia’s book was translated into Hebrew by Judah Ibn Tibbon (in 1186, according to some manuscripts and the epigraph, missing in other editions, of the rare editio princeps, Constantinople 1562) without a translator’s preface. The Amsterdam edition (Saadia ben Joseph 1648) adds nothing in the way of textual criticism and interpretation; it even repeats a switch of folia (77 and 78) of the first edition. A very bad edition, though arranged in chapters, with a double commentary full of mistakes by “Leo b. Jeminis” (I. L. Bensew, or Benseeb), was published in Berlin (Saadia ben Joseph 1789).⁴⁸⁵ The Leipzig edition (Saadia ben Joseph 1859) purports to be a reprint of the editio princeps; it contains some notes by Adolph Jellinek, e.g., 23, and is not free from mistakes.⁴⁸⁶ I do not have access to the Leipzig edition (Saadia ben Joseph 1864), part of Slutzki’s series, accompanied by a short commentary that was continued by Y. Dynes (from p. 87) and an introduction that was omitted in the Cracow reprint (Saadia ben Joseph 1880a), along with the names of the commentators. – The German translation, or rather paraphrase, by Julius Fürst 1845, which omits the last book – appended only later, according to Landauer (p.xx) – is full of mistakes and does not deserve the trouble that has been taken in correcting some of them.⁴⁸⁷

Although the Hebrew translation gave occasion to leave some technical terms in their original Arabic, for instance, theological expressions whose precise meaning may have eluded the translator,⁴⁸⁸ almost none are to be found. The word טפרה (Saadia ben Joseph 1859, I, 23) is accompanied by a literal translation which leaves the reader uncertain about the specific meaning.⁴⁸⁹ We shall return to this translation. |440|

§257. (The anonymous paraphrase.) The character of another Hebrew translation, or rather paraphrase, already reveals itself in its title: פתרון ספר האמונות וחרצב (פילוס) הבינותממליצת מפענח צפונות.⁴⁹⁰ It appears, however, that this translation underwent several redactions, especially with regard to some chapters or less philosophical sections, which were printed from the beginning of the sixteenth century onwards (for which we have to refer the reader to the bibliographical works).⁴⁹¹ These are extant in some manuscripts as well.⁴⁹² – This paraphrase has not been published in its entirety, nor have all of the manuscripts been studied:

Manuscripts: Oxford, Bodl. Ms. Opp. 599 (deficient). Munich, BS Cod. hebr. 42/3 (deficient in chapter 23; Bloch used a copy); an abbreviated preface in Munich, BS Cod. hebr. 65/3 ; Paris, BN héb 669 (an abbreviated redaction, end deficient).⁴⁹³ Parma, BP Cod. Parm. 3018 (deficient, copy in Halberstam). Vatican, BAV Cod. Vat. ebr. 266 and Vatican, BAV Cod. Vat. ebr. 269 . Heidenheim 1 (where is it now? [In Jerusalem, JNUL Ms. Heb. 28 ⁰ 2132 ; Escorial, Escorial G-IV-6 ; fragments in Parma, BP Cod. Parm. 1265]).

None of the manuscripts names Berachiah ha-Naqdan (c. 1260, see §275) as the translator. One or two of his works, along with our book, are extant in Parma, BP Cod. Parm. 2106 . [None of the Parma manuscripts contains both the paraphrase and “one or two” works by Berachiah, but Parma, BP Cod. Parm. 2106 contains Berachiah’s philosophical treatise, which cites large portions of the paraphrase under consideration. See Gollancz 1902.] Rapoport’s conjecture (in his biography of Saadia) that Berachiah paraphrased the Arabic original still will not go away, even after its origin and invalidity have both been demonstrated.⁴⁹⁴ The Paris catalogue, speaking of Paris, BN héb 669 , cites Berachiah as if his name appears in the manuscript.⁴⁹⁵ Talk persists of Berachia’s translation,⁴⁹⁶ or of “Pseudo-Berachia”. It has been suggested that we differentiate between two or even three authors bearing this name in order to distinguish the author of the paraphrase (which is in fact anonymous) from that of the fables etc. True, Berachiah does display some passages from our translation, but he does not present them as his own doing.⁴⁹⁷

We know nothing about the anonymous translator. The published extracts from the paraphrase give no clue as to its [geographical] origin, apart from its very particular literary style to which we shall return.⁴⁹⁸

The problem of the geographical origin of the paraphrase is connected with that of its date. Unfortunately those quotations which one could suppose to be the oldest occur in writings whose own dating is not yet established with certainty. I have found quotations in Jakob b. Reuben’s anti-Christian work, which is dated 1170 and written probably in France; but this date has been called into question, and the doubts have not yet been resolved. In any case that work, especially its |441| final section, which contains the quotations, is not much younger.⁴⁹⁹ Our translation was exploited in the composition of a hymn on God’s unity found in the German liturgy; the identity of its author is uncertain.⁵⁰⁰ A third witness is Isaac Naqdan, who knows a number of philosophical terms taken from it. He probably lived in France,⁵⁰¹ perhaps in Germany, where this translation was known at least from the end of the twelfth century through the first half of the thirteenth. That is the period in which the latest author who quotes from it, Berachiah, lived in France.

Hence this translation probably originated in or near the Provence, since we do not know of any translation from the Arabic that was transmitted from Italy to Germany at that time. Spain is not a possibility, in our view, because the translation lacks a scholarly character. Moreover, its style is reminiscent of the liturgical poetry of the contemporary French and German ritual. As far as its character is concerned, this translation remained an isolated phenomenon. Its language had no influence, and it was almost unknown after the first half of the thirteenth century, even in Germany. There is no trace of it at all in Italy and Spain throughout the Middle Ages. It is the German Elijah Levita (sixteenth century) who probably saw it.⁵⁰² [These and other questions connected with the anonymous paraphrase were taken up in Kiener 1986 . Kiener dates the translation sometime before the end of the eleventh century, and claims that it originated east of the Provence. He asserts that it was prepared by a Rabbanite scholar, eager to counter Karaite anti-Saadia propaganda, which was flowing out of Byzantium in the form of Hebrew translations of Karaite works; and he emphasizes its linguistic debt to Kallir, its longwindedness, mystical scent, and influence on the German pietists. The text remains unpublished; more manuscripts are available (including the elusive Breslau-Heidenheim copy) than were known or utilized by any student of the text, including Kiener.]

When was it written? Is it older than Judah Tibbon’s translation? It probably is, but this has not been sufficiently demonstrated. It is improbable, however, that Ibn Tibbon would have known it, as some have maintained.⁵⁰³

With regard to its style, and some striking terms in particular,⁵⁰⁴ I refer the reader to the characterizations given by Bloch and Zunz, because I do not think that this kind of style matters for the general history of translations (but see the end of this paragraph). I only make this one remark: The books into which it is divided are called מגלה.⁵⁰⁵ It contains only a few Arabic words, like צדר אלכתאב and נסך' אלשרע (Steinschneider 1852, 2277–28).

This translation can be of help towards a better understanding of the original. More important is the fact that, according to Landauer, it exhibits one particular redaction of the text, namely that of the Bodleian manuscript, whereas Ibn Tibbon’s translation tallies more with the St. Petersburg manuscript. This means that one should not simply emend it |442| on the basis of the other redaction (as does M. Wolff).⁵⁰⁶ Landauer illustrates the differences between the two redactions, especially in the seventh book.⁵⁰⁷

Moses Taku did not find Saadia’s name in his manuscript and doubts the authenticity of the work,^507b but then Moses frequently displays a lack of critical analysis. Saadia’s book touches on various opinions which were familiar enough for his contemporary readers so that he could dispense with a more precise presentation. Saadia, who was quite prone to controversy, has as his main concern the refutation of those opinions.⁵⁰⁸ A satisfactory translation, accompanied by explanatory notes, such as Landauer has promised, must take into account both [Judaeo-Arabic] texts as well as both translations.

Meanwhile some younger German scholars have attempted to elucidate our book, or parts thereof, in various ways. An indication of their titles and tendencies may round up this section.

M. Eisler 1870–73 presents, in his lectures on the medieval Jewish philosophers, section 2 (Vienna 1876), an analysis of our work

In his dissertation, David Kaufmann treats the theory of (divine) attributes in our book. Later on he incorporated this into his Geschichte der Attributenlehre 1877a. He presents an analysis of the texts with notes, using Ibn Tibbon’s translation.

Philipp Bloch (Saadia ben Joseph 1879) translated and explained the introduction and the “Cosmology” (Book One), for the most part in the Jüdisches Literaturblatt, on the basis of Ibn Tibbon’s translation, but also making use of the anonymous one.⁵⁰⁹

J. Guttmann 1882 presents Saadia’s philosophy by means of paraphrases or analyses of passages from our book, prefaced by some general observations, and accompanyed with notes which refer to sources and parallel passages. Dr. Simonsen has compared the passages under discussion with the Arabic text and Ibn Tibbon’s translation.⁵¹⁰

David Kaufmann 1883a ⁵¹¹ compared Ibn Tibbon’s translation of the introduction (using manuscripts) with the [Judaeo-Arabic] text and the anonymous paraphrase in order to arrive at a judgment on the translation.⁵¹² We wish to underscore his correct observation that Judah is not responsible for many mistakes of the copyists, multiplied by easily corrected printer’s errors. Kaufmann characterizes the paraphrase very well (232) as “variations on Saadian motifs,” nevertheless I would not be prepared to call the author a “linguistic artist,” because his phrases and metaphors are imitations of quasi-poetry or rhetoric which, even when evaluated within their proper context, i.e., ritual, beg for apologies rather than |443| admiration. By academic standards, this verbose style, whose meaning must be guessed, this self-repeating paraphrase, reminds one of Voltaire’s dictum, “Paraphrase is the mark of a poor tongue.” The anonymous translator may have been learned, but he did not have the talent that his rival possessed to render ideas and abstract terms in a concise, scientific form. Kaufmann recommends an edition of this paraphrase because it is, as he says, an incomparable mine for a lexicon of medieval Hebrew. It exhibits, it is true, a lot of forms and phrases which, due to better taste and an advance in the art of translation, have remained singular instances without imitation. Once one has collected all the expressions and phrases existing in the literature, some amateur specialist may add to them some oddities found in our paraphrase.

Mss. of the Anonymous Paraphrase

Biblioteca de San Lorenzo de El Escorial G-IV-6. (IMHM F 10467), 103 fols.

Munich, Bayerische Staatsbibliothek Cod. hebr. 42/3 (IMHM F 1612), 301a–526a.

Munich, Bayerische Staatsbibliothek Cod. hebr. 65/3 (IMHM F 1130), 20b–1a.

Oxford, Bodleian Library Ms. Heb. Opp. 599 (Ms. Opp. 1185) (Neubauer 1224) (IMHM F 22038).

Paris, Bibliothèque Nationale héb 669 (IMHM 11548), 1–72.

Parma, Biblioteca Palatina Cod. Parm. 3018 (De Rossi 769) (IMHM F 13747), 75b–90a.

Vatican, Biblioteca Apostolica Cod. Vat. ebr. 266 (IMHM F 323), 1b–131a.

Vatican, Biblioteca Apostolica Cod. Vat. ebr. 269 (IMHM F 326), 140 fols.

§258. 2. Commentary to the Book of Creation. We have already on another occasion (§227) touched upon the Book of Creation (ספר יצירה). The oldest extant commentary to it is Saadia’s, which was written in Arabic. For the original we have only the Bodleian manuscript, Oxford, Bodl. Ms. Poc. 256 (formerly Uri 370, of which Dr. Löwe in Brighton possessed a copy; another one was made by B. Goldberg), which carries the title תפסיר כתאב אלמבאדי;⁵¹³ some parts of this have been published.⁵¹⁴ [The Arabic text was published by M. Lambert (Saadia ben Joseph 1891 ) and again by Y. Qafih (Saadia ben Joseph 1972 ); none of the editions take into account all of the available fragments of the original text. The Arabic manuscripts are as follows: Paris, AIU 170; Paris, AIU 69; Cincinnati, HUC 567; St. Petersburg, RNL Evr. Arab I 3070; Oxford, Bodl. Ms. Poc. 256; St. Petersburg, RNL Evr. Arab I 3085; St. Petersburg, RNL Evr. Arab II 1068; St. Petersburg, RNL Evr. Arab I 3071; Oxford, Bodl. Ms. Heb.d.62 ]. In a long introduction, eight theories concerning the origin of the world are presented; the eighth is that of the Book of Creation, attributed to the patriarch Abraham, who, however, is said only to have taught this theory. Only later was it committed to writing by scholars. In this regard it is similar to the case of a part of [King] Solomon’s sayings.⁵¹⁵ The seventh and eighth theories contain only part of the truth, which is expressed completely only in the ninth, namely the Torah. This introduction elaborates in a clearer fashion some of the opinions that are refuted in Saadia’s Book of Religions (i.e., the Emunot ve-De‘ot)

The work contains the complete Hebrew text, divided into eight chapters, each of which consists of a number of laws (הלכות). Each coherent unit of text is accompanied by a translation (תפסיר) and a double commentary (שרח) explaining both the words and the ideas.

We shall not enter here into the problem of the date of composition. We know that the author quotes from several of his own linguistic, exegetical, and legal works.⁵¹⁶ Most probably the book was first written while the author was still in the Fayyūm (i.e., before 928).⁵¹⁷

§259. A Hebrew translation of this commentary (פירוש ס' יצירה) exists in |444|:

Manuscripts: Munich, BS Cod. hebr. 92; Munich, BS Cod. hebr. 221 (deficient; a transcription of Werbluner which Kirchheim quotes is now in Breslau, and probably is identical with the one mentioned by Guttmann 1882 (26)? Halberstam possesses a transcription of Chaim Meir Horowitz. Parma, BP Cod. Parm. 3018 .⁵¹⁸ [New York, JTS Ms. 1911; New York, JTS Ms. 1912; New York, JTS Ms. 1902 ]

The translator – according to the Parma manuscript, he is “Moses b. Joseph b. Moses ha-Dayyan, b. Nathan ha-Dayyan, b. Moses of Lucena” – translated the work for Aaron b. Elijah b. Isaac.⁵¹⁹ The Jewish community of Lucena was expelled in 1148; we do not know if the translator himself or his ancestors lived there. But neither alternative is likely; the translation may just possibly have been done before 1148, but most probably it is to be dated some four or five generations after 1148.

Recently the problem has become even more complicated. At the end of the commentary of Judah b. Barzillai [1885] (before 1140), there is a postscript which contains the introduction to and the beginning of the commentary by Saadia up to the end of the first halakha of chapter 1, in a different translation.⁵²⁰ Judah himself (184) refers to this postscript and remarks that the style of the (unnamed) translator is not good Hebrew (unclear?).⁵²¹ At other places in his book Judah quotes probably from the same translation, which I suppose covers the whole book; and it warrants his stern judgment. [The copy recently identified by Langermann in Vatican, BAV Cod. Vat. ebr. 236 does in fact cover parts of the entire book.] Perhaps, however, another translation of parts of the book circulated,⁵²² since also Moses Taku (around 1230) quotes a passage from the beginning of chapter 4, following yet another translation.⁵²³ Perhaps one of these old |445| translations reached southern Italy, where Moses b. Solomon from Salerno (around 1240) mentions Saadia’s commentary, unfortunately without quoting a passage from it.⁵²⁴ [According to Malter 1921 , 355–357, we have evidence for five, perhaps six, translations of Saadia’s commentary: (i) that of Moses b. Joseph of Lucena; (ii) two, perhaps three, translations mentioned by Judah b. Barzillai, two of which are cited in his commentary; (iii) the translation employed by Moses Taku; (iv) a translation cited by Berechia ha-Naqdan. See also Habermann 1947 , where the author displays in facing columns the two translations of Saadia’s introductory essay, using the citations in Judah b. Barzillai for the anonymous version and Munich, BS Cod. hebr. 221 for the Moses b. Joseph rendition. Like most scholars before him, Habermann is no less interested in Saadia’s text (sometimes called “the Saadia version”) of Sefer Yeẓirah as he is in Saadia’s commentary. Two recent articles on the Saadia version are by Weinstock 1982 and Allony 1982 .]

As there is little promise of finding a better basis, any observations on the fragmentary and unreliable material of this translation that can be made must be accompanied by some reservations. The completely extant introduction is frequently abbreviated. It offers complete Arabic sentences (Judah b. Barzillai 1885, 214 bottom and 215), as if the translator was not sure what he was doing or had not completed his task. There is also no lack of single Arabic words, often present next to the Hebrew translation: רהו'ג (p. 213 last line, 346) אזו'גלא (214, l. 14, probably םימואת is missing after לזמ), רהזו'ג (209, 345, cf. supra 494), םיהמכ ךאבגלא (230 l. 12), 'ץרגלא (221 l. 4 from bottom, cf. 346), הסדנה (214 l. 9 read לשמה, הסדנה לא יפ רבדה ןכו, p. 272 read הסדנהלא (in Moses הסדנהה) and סדנהמ יברע ןושלב, ֞היצאכלא (230 l. 3 and bottom line, wrongly ץאנלא l. 4 from bottom), ךלמ (Categ. 248, cf. 353), ינומקס (for 'קסא? 214 line 3 from bottom), רבונצ םינטוב (222),^524b יבנע ינפג יברשלא (230, 348), ינארבלא רמת (230, l. 3, 348).

Instead of שוריפ we find שרדמ (even ושרדמ ןורתפ 229), also ןורשפ (245, 274), רשפ רבדה (214 l. 3 from bottom), ךל רשפנש (213 l. 4 from bottom). Finally, we note these forms and expressions: םונרקח 162, l. 6 from bottom, הישות ילעב 174, 255, 266 (also in Moses b. Joseph), 177 שרפתש ומכ 177 (cf. 345 from Moses, and Taku 1860, 66), ףנטתמ 178, אטבמב (logic, cf. 346) 213 one line up אשומ רבד 214 l. 1, תוריזנ l. 12 (from נזור, declination [If this is meant to be an astronomical term, then declination is almost certainly wrong; the word usually refers to retrograde motion.]; one word is missing before), מזל העולה (Arabic אלטאלע) and הגנה בעולה,בזנחו העולה and בעו' האריה l. 17 from bottom; ונקשר ונכשר l. 7 from bottom; עם 230 l. 20 (for Arabic קום, people, cf. 348 םדא),⁵²⁵ תורוחצ 349 (for תולוגס, also in the anonymous paraphrase of Emunot in Zunz 1876, 235).

Mss of the Hebrew translation of Saadia’s commentary to Sefer Yeẓirah

Cincinnati, Hebrew Union College 567 (IMHM F 19492).

Munich, Bayerische Staatsbibliothek Cod. hebr. 92 (IMHM F 23122, PH Scholem 21), 75a–99a.

Munich, Bayerische Staatsbibliothek Cod. hebr. 115 (IMHM F 23126, PH 90 (selected pages)), 110b–26a.

Munich, Bayerische Staatsbibliothek Cod. hebr. 221 (IMHM F 1104), 50b–82b.

New York, Jewish Theological Seminary Ms. 1902 (Loewe Ms. 20) (IMHM F 11000).

New York, Jewish Theological Seminary Ms. 1911 (IMHM F 11009).

New York, Jewish Theological Seminary Ms. 1912 (IMHM F 11010), 4a–53a.

Oxford, Bodleian Library Ms. Heb.d.62 (Ms. Uri 370) (Neubauer 2850) (IMHM F 21397, PH 3417), 13–18.

Oxford, Bodleian Library Ms. Poc. 256 (Ms. Uri 370) (Neubauer 1533) (IMHM F 16901).

Paris, Alliance Israelite Universelle H 69 A (IMHM F 3159).

Paris, Alliance Israelite Universelle H 170A (IMHM F 3235).

Parma, Biblioteca Palatina Cod. Parm. 3018 (De Rossi 769) (IMHM F 13747), 75b–90a.

Vatican, Biblioteca Apostolica Cod. Vat. ebr. 236 (IMHM F 292), 95a–142a.

§260. The Hebrew commentary, printed under Saadia’s name ever since 1562, is, according to Munk,^525b the greatest insult that one could possibly do to Saadia, a work not worthy of a sublime mind, nay, of a reasonable human being. In fact, this false attribution was already recognized more than a century ago by the learned critic Jacob Emden.⁵²⁶ We have already remarked that the anonymous author relates a physiognomical anecdote which perhaps derives indirectly from the Arabic Secretum secretorum; an incorrect reading in the printed book, substituting Mainz (Mayence) for Athens (§260, 253), led Rapoport to locate the author in Germany. Nobody now upholds the authenticity of this sham (“factum”, as Munk calls it); however, it does display quotations under the name of Saadia, and these are probably responsible for the misattribution. In fact, they prove the opposite – something not at all rare! Were it possible to establish a connection between these quotations and one of the two translations, it would be useful for the problem of their dating. |446| A quite different, and now more accessible, source has been discovered, it is true, but we cannot carry out here the difficult and complicated investigation that may lead to a clear conclusion. Since no such clear result is now at hand, we shall, for the time being, explicate the most important points.

In the printed Ps.-Saadia (which we shall call “Ps.”) the abbreviation פר“ס (sometimes פ“ר), which is פירש רב סעדיה (R. Saadia explained), appears more than sixty times.⁵²⁷ According to S. Sachs⁵²⁸ this abbreviation is ubiquitously a mistake of the copyist for ש”רפ (יאתבש 'ר שריפ), i.e., Shabbetai (Donnolo), author of a commentary (946) which Sachs knew from one copy⁵²⁹ and which has recently (Florence 1880) been edited by Prof. David Castelli 1880. I cannot agree with this opinion unconditionally. One still has to take into account the manuscripts of Ps. that are complete and arranged differently. Furthermore, the quotations from Sabbatai (Donnolo) in Ps. do not occupy the same place throughout, and a number of them do not have a place at all in the printed Donnolo (the references follow in an endnote). Thus it remains to be investigated whether some of these quotations may possibly be found in one of the two translations of the authentic Saadia. I believe I have found at least one which, however, concerns the text of Sefer Yeẓira.

§261. The translation of Moses b. Joseph and even the freer (anonymous) translation which I call, for the sake of brevity, a “paraphrase,” differ in their whole character from the anonymous paraphrase of the Book of Religions which some wished to attribute to Moses.⁵³⁰ Moses does not offer a paraphrase, but rather, in the words of Neubauer, “an all too slavish translation.”⁵³¹ At that time, I had only Halberstam’s copy of the Munich manuscript, which is in small script, at my disposal, and this did not allow me to carry out a thorough investigation. Now I have a copy of the introduction and some excerpts, e. g., of the two grammatical passages (see note 531) from the same manuscript. To these, Kaufmann’s excerpts in his notes to Judah b. Barzillai may now be added. When we compare Saadia’s texts as preserved in the translation to the two redactions of Sefer Yeẓira, we see that his version differs from both with regard to the order of chapters. Chapter 7 corresponds to chapter 5 of the second redaction, f. 104⁴, line 7 from the bottom אויר רויה גויה; chapter 8 in the beginning נוצר עם is composed of pieces from the end of chapter 3 to 6 of the same redaction; chapter 4 para. 1 = I, 9 of redaction no. 1, eight paragraphs. Chapters 5 (corresponding to 3, 5 bis, f. 73 redaction no. 1) and 6 (= V, 2 redaction 1) are not divided into paragraphs (halachot). Since the Munich manuscripts are missing chapter 2 and the beginning of chapter 3, I cannot indicate the paragraphs.⁵³² Saadia’s Arabic translation of the text has of course not been |447| retranslated into Hebrew except for words and passages indicated in the commentary. The commentary is called פתרונה and פירושה, and the explanation of single words פשטה;⁵³³ when the meaning is clear it is simply called פשטה כמשמעה.⁵³⁴

The translator uses Arabic words from the original, or ones in Hebrew form, as, e.g., םיטטספ (I, 1 טיטאספ Arabic?), תונהמה תוכאלמ (֞הינהמ? II, 1, cf. Steinschneider 1869a, 246), נתדרג and perhaps נצור (IV, I, 2, Kaufmann (Judah ben Barzillai 1885), 340), מסדר (IV, 3 for רדצמ) ודר י⁵³⁵ (אודרו? VIII), האיפלק for תומדיג or תומדג (VIII from ףלק), אסדנה (Introduction, notes 1 and 9 חבור,צורה and הנדסא). In II, 2, where Saadia utilizes Arabic words in order to illustrate his remarks on the letters, the translator adds (?), “and all these words belong to the Arabic language.”

Very frequently a phrase or a word is translated and explained by another one, mostly with the use of כלומר, e.g., in the introduction n. 8, וצריך לתוספת אני מספרה זה כלומר עתיד אני לספרה אחר זה; in Arabic only ויחתאג' אלי זיאדה אנא ואצפהא בעד הד'א; I, 1 (Kaufmann (Judah ben Barzillai 1885), 339) בערים ממנה כלומר שלא ידעוה and רשמו כלומר ענפו; II, 2 שף עליהם זכר הגלוי כלומר כדי הם לזכרם בגלויים; III, 6 בקירב ענין כלומר בדרך העברה; ibid. (end of chapter) ולו המטריא כלומר המכאיב; but for יענון, I, 1 (f. 59, B. f. 81) the Arabic text would seem to have מעני.⁵³⁶ Frequently we find חכמי התושיה (beginning of the introduction I, 1, 2, also in the paraphrase; §259, end),⁵³⁷ also סרעף for מחשבה and the like (also in the paraphrase of Emunot), I n. 4: מדרך הסרעף והמזמה (paralleled on 271 לפי דעת והמזמה).

We present more details following the order of the book (‘P’. refers to the paraphrase in Judah b. Barzillai). Introduction, opinion no. 1 עם שדולם, P. 269 line 1צחצוח הבאור ; בחנם,|448| P. לרצות פשרונו;⁵³⁸ n. 2 ועל הרדיפה למה שאמרנו, כלומר על סבר נוהגי מנהג בה,the passage is missing from P.; וכן המקרה בקפוי (for the Arabic גמאד), P. וכן הכל בנבך, perhaps garbled from שלש פרידות ?מתכת, P. חלקים; n. 3 גושם (elsewhere, גושמים וגלמים I, 1, גושמי I, 2); יפטירו ויפשירו ויורו וירמיזו, P. 271, line 11 לפרש מהם וללמוד מהם על דרך העברה ודמיון;n. 4 הרוחניים, for animals; P. חיות; cf. n. 5 הרוחניים המדברים, P. נשמת כל חיים מדברים (probably alluding to the prayer נשמת כל חי); n. 6 לכל המזגים והפיתוכים; missing in P. 272; n. 7 זה זולת עובר משני פנים (Arabic ג'איז, cf. Steinschneider 1857b, 121), missing in P.; n. 9 מוקצים אל הארץ, P. 273 סמוכים. – For the following passages the variant readings are missing from P. The quotations in the notes of Kaufmann to Judah b. Barzillai, viz. from I, 1 338 to 119, 339 to 155 (where we find לח ונצוק), I, 4 345 to 209, IV, 1, 2 340–42, IV, 4 345 bottom, offer little material. I note from my excerpts: I, 1 רשול, סדור ועריכה, names of the Categories (see Endnote), פרק ויבדד; end חשבון האישטרונומיא (f. 59, cf. II, 1 f. 64 b); I, 2 עימות and מעומתים (from לעומת), thus also to be read מעומתים in III, 6; המתויכים גוף האדם, and parallel, ממציע האדם; ורצה בזה התיוך כי הוא באמצע הכל; I, 4 טובע טבעם ככה וקוצב קצבם כן;⁵³⁹ בעל כורח כל יצור צריכים למו; the same החתים ⁵⁴⁰;עצלות כל חכם מלהוסף עליהם and יונצח (read ינוצח?). II, 1 שער גדול ועצום מן ההכר; II, 2 התכפל ; III, 6 f. 70 b דיאפראגמא; IV, 6 עולם קטן;IV, 8 מי שקרא ס' הנתוח והועלת [ותועלת] האברים (probably Galen is meant); IV, 12 גזרתו פולשת; V (in Jellinek 1852, 33) המפייטים and (f. 79) קמענים; VI (Jellinek 1852, 73 Zunz 1865, 318) איול.

§262. (Pseudo-Ibn Ezra.) A small work, tainted with superstition, bearing the title ספר העצמים (Book of Substances, or Beings), is extant in the following manuscripts:

Berlin, SPK Or. oct. 244/4 (Catalogue 56, formerly Luzzatto). Oxford, Bodl. Ms. Mich. 238 . Florence, BM-L. Ms. Plut.II.25 (deficient in the end). Mantua, CI Ms. ebr. 78a ⁵⁴¹ Parma, BP Cod. Parm. 2615/3 . – Schönblum 81 (II, 72).⁵⁴² [The latter manuscript has not been identified. Other known manuscripts are Cambridge, UL Add. 1186/7 , New York, JTS Ms. 2321/3 , New York, JTS Ms. 2323/1 , New York, JTS Ms. 2349/1 , London, BL Add. 16934 , London, BL Add. 27038 , Moscow, RSL Günz. 338/3 , Naples, BN VE III F12 .]

The translator is nowhere named. Samuel Zarzah (1369) displays almost all of chapter 3⁵⁴³ as well as a long passage from the beginning in which the author tells us what he was told by an eminent religious person “who in the true sense of the word deserves the title philosopher.”⁵⁴⁴ Zarzah admits⁵⁴⁵ that he does not know Arabic. He asked Jacob Ibn Alfandari ben Solomon⁵⁴⁶ to translate for him the explication of some passages in Ibn Ezra’s commentary to the Pentateuch (the explication is by Solomon Ibn Yaish?⁵⁴⁷; al-Fandari was probably also the translator of the note of Joseph Ibn Wakkar cited in the same passage by Zarzah)⁵⁴⁸ as well as the Book of Substances by Abraham Ibn Ezra.^548b [The Book of Substances was printed by M. Grossberg 1901; Grossberg used only one manuscript, namely the copy found in the British Library. Jacob al-Fandari is named as the translator in Naples, BN VE III F12 , dated 1492 and possibly the earliest copy of the treatise.] |449|

Abraham is also credited with the Arabic Book of the Substances by Samuel Ibn Motot (1370) who presents some passages from it in Hebrew. [The passage is found in Milan, BA P 13/7 Sup. , fols. 319bff., and corresponds to Grossberg’s printing, 12 ff. The text cited by Ibn Motot is almost certainly the same translation found in the manuscripts, though there are some variant readings.] But this work is not by Ibn Ezra. It may even be the work of a Muslim, perhaps modified by the translator. [The text refers to Jewish ritual and scripture.]

It is divided into six chapters (ןפוא): (1) On the First Cause; (2) On the influence of the Higher Spheres (of the Separate Intellect) on the Lower ones; (3) On the followers of positive (i.e., not a revealed) creed⁵⁴⁹ like the Sabians, the Nabataeans (corrupted into טבכ and טוב); (4) On the Soul; (5) On the Animals (not a separate chapter in the Parma manuscript, so that Perreau in his analysis of the work lists only five chapters);⁵⁵⁰ (6) On the Spheres.

The traces of this book vanish soon after its translation. No direct quotation is known after the fourteenth century. [However none of the thirteen extant copies are earlier than the fifteenth century. Moreover, with the exception of New York, JTS Ms. 2323/1 , all copies are in Italian hand; the JTS manuscript is an Ashkenazi hand, but it mentions Lombardy in the incipit, so it too seems to have an Italian connection. Thus interest in the treatise seems to be concentrated in two very specific moments: the supercommentators to Ibn Ezra’s biblical commentary who worked around the middle of the fourteenth century (Zarzah, Ibn Motot), and Renaissance Italy. At least two incipits relay information about the text and its transmission. It was called in Arabic Kitāb al-Jawāhir , and it was written by Ibn Ezra for Judah Hallevi. This second point cannot be confirmed from any other source; it may reflect the observation of one reader that the tenor of the text is similar to that of the Kuzari . Interestingly both of these manuscripts are missing the last chapter, which deals with astronomy. Here follow the two incipits:

• New York, JTS Ms. 2323/1 , f. 31a: הספר הזה חברו החכם ראב“ע ז“ל בלשון הגרי ושלחו לחכם ר' יהודא הלוי ולא נמצא בגלילנו זולת ההתחלה מצאתי בישוב (?) לומברדאה העתקתי בידי שנות והקריב“ה לעולה לפ“ג.

• Naples, BN VE III F12 , f. 104: התחלת ספר העצמים לאבן עזרא ז“ל ספר העצמים כתא“ב אלגוהי“ר(!) לראב“ע ז“ל חברו החכם השלם בלשון הגרי ושלחו לר' יהודה הלוי ז“ל ולא ימצא בארץ הזאת זולת ההתחלה הזאת והעתיקו ר' יעקב אלפנדרי.

Mss. of Pseudo Ibn-Ezra, Book of Substances

Berlin, Staatsbibliothek (Preussischer Kulturbesitz) Or. oct. 244 (Steinschneider 79) (IMHM F 1996), 17–21.

Cambridge, University Library Add. 1186/7 (SCR 575) (IMHM F 17052), 101a–7b.

Florence, Biblioteca Medicea Laurentiana Plut.II.25 (IMHM F 17666), 1–132.

London, British Library Add. 16934 (Margoliouth 793/12) (IMHM F 5478), 146a–80b.

London, British Library Add. 27038 (Margoliouth 1073/7) (IMHM F 5716), 45a–54a.

Mantua, Comunita Israelitica Ms. ebr. 78 ebr. 78 (IMHM F 864), 91a–99a.

Moscow, Russian State Library Ms. Günzberg 338 (IMHM F 47620), 33a–42b.

Naples, Biblioteca Nazionale Vittorio Emanuele III F 12 (IMHM F 11526), 104a–7a.

New York, Jewish Theological Seminary Ms. 2321/3 (NY JTS Acc. 2979) (IMHM F 28574), 29a–36b.

New York, Jewish Theological Seminary Ms. 2323/1 (NY JTS Hi 109) (IMHM F 28576), 31a–37b.

New York, Jewish Theological Seminary Ms. 2349/1 (NY JTS Acc. 75327) (IMHM F 28602), 31a–37b.

Oxford, Bodleian Ms. Mich. 238 (Ms. Mich. 316) (Neubauer 1234/3) (IMHM F 22048), 34b–44b.

Parma, Biblioteca Palatina Parm. 2615/3 (De Rossi 1355) (IMHM F 13316), 30a–38a.

[An anonymous defense of creationism. Ms Vatican ebr. 236, ff. 62a-64b, contains a series of refutations of the eternity of the universe. We learn from the incipit that it has been translated from the Arabic by an unnamed individual at the request of “my master and brother Moshe ben Yo’av “: עוררתי בקשתך רבי ואחי/ משה בן יואב רי“ת ונערתי חצני להעתיק הספר הזה/ מלשון ערב ללשון עברי. The translator worked from a borrowed copy of the original. Though he complains that the stipulations imposed upon by the owner (תנאי השאלה) prevented him from copying “part of it”, in the very next sentence he states that he “copied it in its entirety”. In any event, the one extant manuscript is incomplete, breaking off with the words לא מצאתי יותר תם ונשלם. This, however, could possibly be the comment of the copyist rather than the translator.

The opening sentence names פיתאגרוס וטולמיוס as the authorities who denied the possibility of creatio ex nihilo . Though no available Pythagorean text known to us argues against creation, the name of Pythagoras was closely associated with the pagan neoplatonists, so his inclusion is not surprising. The name of Ptolemy, however, makes no sense in this context. טולמיוס must be an error for טימאוס; and indeed the Timaeus was one of the major sources for the view of the pre-eternality of matter. Moreover, the Timaeus is Plato’s most “Pythagorean” text; hence it is much to be expected that “Pythagoras and Timaeus” would be mentioned together in this context. Indeed, later on Plato and “his evil disciples” are named as the authors of this view.

A second school named at the beginning of the treatise is the בעלי התולדות. In their view, the causes of everything are to be traced to the four elements and their mixture (מזג); in other words, they deny the existence of a first cause. Later on a third school of thought is named: that of Aristotle, who considers hyle and form to be “the Agent” (הפועל). The treatise contains both rebuttals (תשובה) of the pernicious doctrines as well as proofs (ראייה) for the existence of a single creator.

The formulations of the arguments, and perhaps some of the strategies as well, are unusual; the text awaits close study. This much, however, can already be said with some degree of certainty: the text bears many similarities to the Saadia’s Beliefs and Opinions and arose out of the same milieu. Most notable is the pride of place given to Plato’s Timaeus as the chief authority for the eternity of the world (not mentioned by name in Beliefs and Opinions , but certainly the source of the “second school” refuted there), and the complaint registered in both treatises that some Jews have been persuaded to accept Plato’s teachings. Note that in Beliefs and Opinions , book I, before setting forth his rebuttals of all the schools known to him to have denied creation ex nihilo , Saadia mentions additional arguments that he proffered in other writings; and one cannot rule out the possibility that the treatise under discussion is the work of Saadia, perhaps from this refutation of Hiwi al-Balkhi. A full discussion of the plausibility of this suggestion is beyond the purview of this notice. Let it be said briefly that in the extant portions of Saadia’s reply to Hiwi there is no discussion of creation; and the fact that a rhymed Hebrew text does exist does not negate the possibility of a prose writing in Arabic as well.

Some of the Hebrew terminology employed is unique to this writing, and the meaning of some words remains a mystery. Here are some examples: והשורש קודם למה תומדב ויתחתש; שדוחמה לע אלא לע יוחימה ןיאו יוחימה אוה יונישהו; סוטיפהו גהנמה. The translator employs םרג for jawhar and אצומ for madba' .

This text was first identified by Y. Tzvi Langermann, and his edition, translation and analysis appears in Langermann 2013.

The Karaites

§263. We collect here the writings of the scholars of this sect, because the sources for the few authors to be considered are the same and, accordingly, only with difficulty can one decide to which one of them a given source may relate. Moreover, the character of the translations is quite peculiar, so that the quotations in later works, which are written under the influence of the Rabbanite translations, stand out on account of their strikingly unusual character.⁵⁵¹ [For a recent and authoritative study of the peculiar Hebrew of the Karaite translators, see Simon Hopkins 1992; also noteworthy is the M.A. thesis of A. Maman on the Hebrew of Tobias b. Moses, submitted at Jerusalem, 1978, and his essay on Karaite Hebrew Maman 2003. For an exhaustive treatment of the Karaite translators in their communal and intellectual context, see Zvi Ankori 1956, passim . A recent comprehensive guide to Karaite philosophy is surveyed in Ben-Shammai 2003 . Maman and B. Shammai’s essays are part of the close to thousand-page guide to Karaite Judaism edited by Meira Polliack 2003 , and with a comprehensive bibliography by B. Walfish. Walfish and Kizilov have greatly expanded the bibliography in 2010 .]

§264. Until recently, it was customary to speak of two Karaite authors by the name of Joseph whose date was not known exactly, but taken approximately to be shortly after Saadia Gaon (941). It would lead us too far to discuss thoroughly all that has been brought forward concerning these authors and their works in Arabic, part of which were translated into Hebrew. Therefore we confine ourselves here to a discussion of a few key points, i.e., to identify, if possible, the authors of some philosophical works translated into Hebrew. We must, however, preface this discussion by some remarks concerning the authors.

The confusion between the two Josephs, or at any rate the uncertainty as to how to distinguish between the two, seems to go back to the twelfth century. The older Joseph, living perhaps around the last years of Saadia (937) or somewhat later, is Joseph al-Qirqisānī, called already in the eleventh century Abū Yaʽqūb.⁵⁵² [The confusion of which Steinschneider speaks no longer exists. Al-Qirqisānī’s name was Jacob, or, more precisely, Abū Yūsuf Yaʽqūb. Steinschneider knows of a source that cites precisely this name, but he rejects it; see §266. Throughout the remainder of his entry, Steinschneider erroneously refers to “Joseph” Qirqisānī. There are other errors as well in his report, and we shall draw attention to them. However, in the concluding sentence of §266, Steinschneider wisely urges caution with regard to all the conclusions and inferences that he has drawn, given the state of knowledge in his time. Indeed, much of the confusion had been dispelled in his Die arabische Literatur der Juden , published 9 years after HUe . There al-Qirqisānī receives a lengthy notice (79–84); he is identified as the author of Kitāb al-Anwār ; and many manuscripts are listed. On the other hand, Steinschneider still gives his name as “Joseph Abū Yaʽqūb”. The “second” Joseph, conventionally referred to now as Yūsuf al-BaṢīr, receives a separate entry on 89–91. Al-Qirqisānī's Kitāb al-Anwār edited by Leon Nemoy in 1939; see his corrections and emendations 1959–60. An English translation of the first treatise by W. Lockwood, with two important introductory essays by B. Chiesa, is found in Yaʽqūb al-Qirqisānī 1984 . Since Nemoy, the most important studies are those by Vajda in the REJ and the doctoral dissertation of Haggai Ben-Shammai on al-Qirqisānī and Yefet b. ‘Eli 1977. For the works of Yūsuf al-BaṢīr, see Basir 1985 , which lists many of Georges Vajda’s studies on al-BaṢīr, and David Sklare and Haggai Ben-Shammai’s catalogue on the works of Yūsuf al-BaṢīr in the Firkovich collection (Hebrew) ( 1997 ).] One quotation has the name Jacob b. Joseph |450| Qirqisānī⁵⁵³; among the first Karaites a certain Jacob b. Isaac Qirqisānī is named.⁵⁵⁴ The second Firkovich collection, unfortunately little known up to now, contains two Arabic works that were previously known only by title. Firkovich⁵⁵⁵ names as the common author of both works “Abū Yūsuf Ya‘akov al-Qirqisānī”, for the first one even Jacob b. Isaac b. שמעיה.⁵⁵⁶ The second is a commentary to the passages in the Pentateuch which are not concerned with the Law, under the title אלריאץ' ואלחדאיק [ Book of Heaths and Gardens ], allegedly composed in 937. A. Harkavy⁵⁵⁷ confirms the author’s name and connects the Arabic title with the Hebrew ס' הנצנים (Book of Flowers); this latter has been ascribed to both Joseph Qirqisānī and Joseph b. Noah (?).⁵⁵⁸ Perhaps the treatise on the Decalogue, ascribed to Joseph b. Jacob Qirqisānī in ms. Paris BNF héb 755, is part of this commentary? [Indeed the author’s name is given in this manuscript as יוסף…המאור קרקסאני; however, it is now believed that the commentary is the work of Elazar b. Eliezer.] ⁵⁵⁹ According to Harkavy the author of the Arabic work quotes Saadia Gaon. Considering the uncertainty and confusion governing this whole field one may venture the conjecture that the Arabic manuscript has seen a switch of names, so that the name of Joseph, displayed in other sources, would be that of the author, and Abū Yaʽqūb his cognomen. The combination Abū Jacob (for Ibn J.) with Joseph⁵⁶⁰ has been unjustly attacked; but, then again, I do not believe that someone changed the names intentionally in order to arrive at the more common combination.

Finally, there is an author known only as Abū Jacob who is most probably one of the two Josephs. We shall soon return to him, as well as to a second work ascribed to one Jacob and a number of other designations of one or the other Joseph.

§265. (Joseph b. Abraham) Almost everything appertaining to the second Joseph is likewise not free of doubts. He is usually called Joseph b. Abraham ha-Kohen.⁵⁶¹ Firkovich objected to the last element |451| in his name for no good reason.⁵⁶² In an Arabic work he also carries the cognomen Abū Yaʽqub.⁵⁶³ Being blind, he was called with a euphemism common to Jews and Arabs⁵⁶⁴ “the seeing” (הרואה, אלבציר).⁵⁶⁵ That epithet was applied to Joseph Qirqisānī; this was due to some sort of confusion, probably connected to the titles המאור or המאור הגדול which were attached to his name. This leads us back to the Arabic work that was mentioned in the preceding paragraph.

§266. The Book of Lights (כתאב אלאנואר),⁵⁶⁶ no. 493 of the second Firkovich collection, is – according to its owner – ascribed to Abū Joseph Jacob b. Isaac b. Shemaya al-Qirqisānī and composed in 937. This is very dubious. [As noted above, no one any longer doubts the correctness of the name or date as reported in the Firkovich manuscript.] In an account by Neubauer,⁵⁶⁷ the author bears the name “Joseph of Kirkisia”; Israel Maaravi⁵⁶⁸ calls him Qirqisānī. Simcha Isaac attributes the book האורים, quoted without a name for the author by Levi, to Joseph b. Jacob Qirqisānī⁵⁶⁹; others speak of a Book המאור or המאור הגדול, written in 910 or 930 by Joseph (הרואה) who, according to them, is earlier than Qirqisānī (later on mentioned in one of the two sources)⁵⁷⁰ and engages in a refutation of Saadia Gaon. This then would be Joseph b. Abraham who would not, however, be missing from the list of the Karaite scholars.⁵⁷¹ Nevertheless Simcha Isaac attributes the book המאור הגדול to Qirqisānī. – Moses b. Solomon ha-Levi הצבני ⁵⁷² is the author of a compendium of the Book of Lights, called מכ'תצר אלאנואר. [The current view is that the proper spelling of his name is הצכני; some twenty-two copies of his abridgement have been identified.] Moses Bashyazi, to whom we owe this reference,⁵⁷³ quotes and |452| translates passages from this Arabic work, saying “Moses b. Solomon relates in the name (משם = בשם) of Joseph Qirqisānī.” [As noted, the Arabic text of Kitab al-Anwār was edited by Leon Nemoy in 1939. Portions of al-Qirqisānī’s Book of Lights were translated into Hebrew probably by Moses Bashyazi, son of the great Karaite leader Elijah Bashyazi. To date six small portions of his translation have been registered in the holdings of the Institute for Microfilmed Hebrew Manuscripts; for the most part they are translations of selected passages that have been penned into the margins of the Judaeo-Arabic original. However, by far the largest section of Moses’s translation is found in St. Petersburg Oriental Institute of the Russian Academy D4 (IMHM F 69408), a manuscript that has not yet been catalogued at the IMHM. Fols. 3a–64b contain Bashyazi’s translation of book VI in the right column, usually headed העתקת משה; the left displays the Judaeo-Arabic. At the bottom of f. 6b Moses reports that he found the manuscript used for the translation in Egypt, but that it was missing an entire quire: אמר משה בשיצי המעתיק עד פה מצאתיו שלם במצרים ומכאן ואלך חסר קונטרוס אחד שלם. Another portion of this translation may be found on fols. 110a–116b of the same manuscript. It may be added that the first section of book VI discusses the “premises of the Alexandrian” ( muqaddamāt al-Iskandarāni ); and it has been suggested that “the Alexandrian” is none other than Philo of Alexandria, whose nonpresence in medieval Jewish philosophical writing is striking.] From this formula it cannot be inferred that Qirqisānī is the author of the basic text of the compendium,⁵⁷⁴ since Qirqisānī could well be quoted by Joseph b. Abraham if, as we shall see, the latter was the author of the original work. This, indeed, is the more likely possibility, according to everything which we have collected concerning “the blind one” who would thus have authored the Book of “Lights”. Similarly, one has supposed that Isaac the Blind, son of Abraham b. David (beginning of the thirteenth century), is the author of the mystical book “ha-Bahir” (the Shining). [This may be a reference to M. H. Landauer’s attribution in the Literaturblatt des Orients ( 1845 ), col. 215, mentioned by Scholem 1987, 253 n. 116; since Scholem, most scholars prefer to talk of redactors of ha-Bahir. See Anonymous 1994.] However, the date 937 in the ms. Firk. (if it is not simply taken over from manuscript copies of the book 'אלריאץ would decide against [the authorship of] Joseph b. Abraham who lived, according to Firkovich and Harkavy, almost a century after Saadia (and Qirqisānī). Harkavy thought that the author of the Book אלאסתבצאר (or יהי אור, fiat lux, again an allusion to the blind) polemicizes against Samuel b. Ḥofni (d. 1034).⁵⁷⁵ Thus this book is different from the translated compendium of the work on dogmatics.

If I understand correctly the latest information from Harkavy (which does not fully agree with his earlier reports),⁵⁷⁶ the Book of Lights is an introduction to the commentary on the Pentateuch. It treats the history of the Jewish sects, contains occasional polemics against Christianity and closes with a complete Book of Laws (כתאב אלשראיע). The last item appears to be the Book of Laws which Yefet b. Zair attributes to Joseph b. Jacob Qirqisānī.⁵⁷⁷

These involved discussions demonstrate that we can talk about the earliest Karaite authors only with strong reservations.

§267. (The translated writings.) Joseph b. Abraham composed the following works of which the first one is the most important.

1. אלמחתוי (The Encompassing). The original was thought to be lost. D. Kaufmann recently purchased an almost complete manuscript Budapest, MTA Kaufmann A 280 ; fragments exist in St. Petersburg. [For a sample catalogue and description of the fragments, see Sklare with Ben-Shammai 1997 , 77–89. The Budapest ms, which is missing the first chapter, was the basis of Vajda’s edition of the Arabic text; on the basis of the St. Petersburg fragments, an edition of the missing material has been prepared by H. Ben-Shammai in the sample catalogue 1997 , 113–26.] The Hebrew translation is called ספר הנעימות (Book of Agreeable Things); Hadassi (chapter 258) calls it also זכרון הדתות (Book of Sects?). [On the translation of ספר נעימות see Jacob Mann 1935 II, 290 n. 10.] ⁵⁷⁸

Manuscripts. Leiden, BR Cod. Or. 4779 (a copy in Trigland, no. 24, another one in L. Dukes).⁵⁷⁹ Paris, BN héb 670b ⁵⁸⁰ St. Petersburg RNL Ms. Evr. I 687; St. Petersburg, IOS B 241b (with compendium); St. Petersburg IOS B 394 (compendium); St. Petersburg, IOS C 103b (with compendium). Ms. of Abraham Misri in Petersburg [ = St. Petersburg, IOS B 241b?].⁵⁸¹ |453|

Although this work has been brought closer to us by a century, it has lost nothing of its significance for the history of Arabic and Karaitic dogmatics, for it is the sole extant model of a work of Mu‘tazilite kalam that could just as easily have been written by a Muslim. The enlightening analysis of the book, which takes into account the compendium (§268) as well, published by F. Frankl in the Sitzungsberichte of the Royal Academy in Vienna,⁵⁸² was meant to be followed by an edition based on the original. Unfortunately the industrious and expert scholar in this field died (August 1887) after all of the preparations (for this sequel) had been completed.

The translation continues to be beset by a number of (text-)critical problems. The translator is anonymous. The end of the Leiden manuscript reads ונתקן בירושלים, which may mean that either the work or the translation was written in Jerusalem, or else that this copy was corrected there. We know too little about the translator for us to decide whether his travels took him as far as Jerusalem, something that seems improbable.⁵⁸³ The translation is the work of a scholar who lived in or in the vicinity of Greece, since he frequently resorts to Greek words.⁵⁸⁴ At the beginning of the same manuscript, at the end of the list of forty-five “gates” (chapters), there is an index of the topics that are discussed. I took the trouble to actually count the chapters, a cautionary measure dictated by the number forty given by Simcha Isaac. According to Pinsker, 1860, 195, the Misri manuscript has thirty-five “gates” and three “chapters” (four, according to Frankl’s table). St. Petersburg RNL Ms. Evr. I 687 in Firkovich’s catalogue numbers forty-three chapters. Doubtlessly, it was neither the author nor the translator who numbered the chapters. Apparently Pinsker himself added the numbers, erroneously giving two of them the number sixteen (ט“ז and י“ו). He gives no chapter titles at all after no. 3, even though they are displayed in the Leiden manuscript. Frankl divides the work into two parts, with nineteen and twenty-four chapters respectively.⁵⁸⁵

The author himself remarks (chapter 16b is in Pinsker, “That God is abundant”)⁵⁸⁶: “We have counted (מנינו) [this theme] among [or: taken it up within] the chapters on theodicy (משערי הצדק והיושר). Others have dealt with it together with the chapter on God’s unity.⁵⁸⁷ The deviant designation |454| of the final three “chapters” is not to be found in the original; they would have had their counterpart in the Arabic work, had they been the work of the author. Pinsker uses this, without justification, to support his hypothesis that the three “chapters” were supplied either by Joseph’s student, Jeshua b. Judah, or else by Tobias, who translated the book. On the basis of this assumption Pinsker maintains that the books משיבת נפש and אוצר נחמד, to which the author refers only in these three chapters, are works of Jeshua. Frankl, whose analysis covers the three last chapters, did not raise this question beforehand. Fürst had the rare luck to hit upon a correct observation of his own; he replies to Pinsker that one can find in the compendium chapters that correspond to those three.⁵⁸⁸ Unfortunately he forgets this remark in the course of his very own article, when he presumes that the passages which contain references to the two books just mentioned are interpolations of Jeshua, “as Pinsker has correctly proven!”⁵⁸⁹ – But this is a different hypothesis, one which concedes the authenticity of the three chapters, except for the quotations in question, which Jeshua would now have interpolated. Fürst, however, does not tell us whether these interpolations were done in the Arabic original or in the translation, nor whether he adopts Pinsker’s opinion that either Jeshua or Tobias is the Hebrew translator. In the end of the section on Jeshua, Fürst completely forgets his objection and his own view; Jeshua is the Arabic author of the three chapters, and in the section on Tobias, this person makes his entry as the translator!⁵⁹⁰ So we have to deal with two different questions: the authenticity of the three chapters, upon which the authorship of the two books mentioned therein depends, and the identity of the translator.

One quotation, in Steinschneider 1858, 172, which Fürst has forgotten to take into account, proves that the author of one of these chapters is identical with the writer of a passage of the book אוצר נחמד, Oxford, Bodl. Ms. Opp. 26 , attributed to Tobias. It is too early, however, to judge whether this compilation was translated from Arabic or is an original, what its sources are, etc. No other manuscript of this work is known to exist.⁵⁹¹ [Tobias b. Moses is now presumed to be the author; a second copy has come to light in ms. St. Petersburg, Institute of Oriental Studies C 116.] For the quoted משיבת נפש, see §271.

For us, the question of the translator of the entire book is the more important one. Most of the old Karaite translations do not name the translator; from the first period (until mid-twelfth century) almost none is known by name.⁵⁹² Only Tobias (§268), a student of Jeshua and the “translator” par excellence, is well known, but he hardly translated on his own all of the writings that we have (apart from those which are lost or which we do not know), nor even all those which by using Greek words betray their origin in Greece or its neighboring countries.⁵⁹³ In order to maintain (following |455| Pinsker) that Jeshua was the author of our work, we need additional proof, beyond the few passages in the three chapters, which allow for diverse interpretations. One would rather assume that in the Orient, where the tradition of composing in Arabic continued, the need to translate Arabic works was not yet felt. Perhaps Tobias was the first translator and Hebrew compilator,⁵⁹⁴ because he had been in Jerusalem and lived in Constantinople, the only region where Karaite immigration was significant enough to bring about translations. As a matter of fact, the first traces of Karaite literature in Constantinople may be observed approximately in Tobias’s time, [a coincidence] to which we shall return shortly.⁵⁹⁵ This period is of interest to us primarily because of an entirely new question, i.e., whether the first Rabbanite translators may have borrowed words and forms from old Karaite translations. I shall tackle this very complicated inquiry when the study of some pertinent manuscripts will be granted to me. [Steinschneider apparently never carried out this intention.]

The character of our translation is described in detail in the Leiden catalogue 1858, 167 ff.; Pinsker 1860, Appendix, 199, has collected a number of unusual words, and Fürst blends them into a stew of mistakes, as is his habit. The translation has a highly Arabicizing character, leaving a great number of Arabic words and even phrases as they are and structuring the Hebrew according to Arabic forms. Attention has already been called to Greek words. Frankl (Nachrede, 7) thinks Tobias is the translator, since he has translated the compendium (§268); the identity of the translator, he says, is “quite evident” – but under current circumstances, as we have explained, a more detailed argumentation would not be superfluous. More important is his remark that the three (or four) final chapters, discussed above, were considerably abridged, probably because Tobias could refer in their place to his own writings משיבת נפש and אוצר נחמד (sic). One may still resolve all of the difficulties if those writings in their principal contents were not written by Tobias, but are rather compilations from Arabic texts, to which Tobias may have added something (§271.5).

Joseph ha-Roeh Sefer ha-Ne‘imot

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779 (IMHM F 28071).

Paris, Bibliothèque Nationale héb 670 (IMHM F 11549), 50b–51b.

St. Petersburg, Institute of Oriental Studies B 394 (IMHM F 53555), 44a–63a.

St. Petersburg, Russian National Library Evr. I 687 (IMHM F 51269).

St. Petersburg, Institute of Oriental Studies B 241 (IMHM F 53371), 4a–124b.

St. Petersburg, Institute of Oriental Studies C 103 (IMHM F 69335), 1b–48a.

§268. 2. כתאב אלתמייז ⁵⁹⁶ Book of the Compendium [?]. The original was assumed to be lost. Recently the British Museum purchased a defective manuscript from |456| [Moses Wilhelm] Shapira (chapters 9–20 and 23–29). In the preceding work the author quotes it under the title אלמנצורי; Joseph Ibn Zaddik, who calls the author Abū Yaʽqub, follows his method in part.⁵⁹⁷ [For the question of the identity of אלמנצורי with כתאב אלתמייז, see Sklare with Ben-Shammai 1997 , 65, which introduces the catalogue and description of the Judaeo-Arabic fragments in St. Petersburg, 65–89.] I shall not comment upon Fürst’s hypotheses concerning the title of the work.⁵⁹⁸

The Hebrew translation, which carries the title מחכימת פתי (Psalm 19:8), is extant in the following:

Manuscripts: Paris, BN héb 670/2; Oxford, Bodl. Ms. Poc. 213 (defective)⁵⁹⁹; Oxford, Bodl. Ms. Heb.f.12 ; Leiden, BR Cod. Or. 4779 /3 (1858, 179). Paris, BN héb 670/2. St. Petersburg RNL Ms. Evr. I 688 , St. Petersburg RNL Ms. Evr. I 689 (defective). – St. Petersburg IOS B 342b , St. Petersburg, IOS B 241/4 , St. Petersburg IOS B 394 Misri.⁶⁰⁰ Deinard (1888). [For selections of the Hebrew translation, see Vajda 1976.]

According to the epigraph in the Leiden manuscript the book contains additions by Tobias,⁶⁰¹ who would then be the translator from the Arabic. – The Paris catalogue writes that “this treatise was apparently originally written in Arabic.” As a matter of fact, it contains few Arabic words, but stylistically it exhibits many arabisms. The Greek words may just possibly be traced back to the author, but it is easier to assume they were added by Tobias. On the other hand, a double Arabic title does not fit a Hebrew work, and we do not know of any other Hebrew work by this author. Neubauer⁶⁰² supposes our book to be identical with אלאסתבצאר, but he himself notes a quotation from the eleventh treatise of the latter⁶⁰³ which proves that it is not our compendium; in any event, its chapters are not numbered. Neubauer exhibits part of the epigraph of St. Petersburg RNL Ms. Evr. I 688 ; he was led to assume, on the basis of the equivocal המעתיק, that the book was translated three times.⁶⁰⁴ The manuscript was copied by Elijah b. Isaac who says, towards the end, talking about some historical facts, particularly concerning the Karaites, that 500 Karaite books were burned in Constantinople in 1735; he himself had lost between 50 and 60 books in Gosloff⁶⁰⁵ and could not find a complete copy of the copied book. The prototype of this manuscript was copied, not translated, by Elijah b. Barukh Yerushalmi, a well-known copyist (around 1654); Elijah is probably also the author, not translator, of some Hebrew works.⁶⁰⁶ He always tried to improve his copies; probably we owe to him also some alterations to the text. He supplies prologues and epigraphs which sometimes have been given more authority than they deserve. In our manuscript, at the end of the foreword, Elijah apologizes for leaving out some words or adding letters, something that the translator Tobias had already done, according to the observation of Elijah Bashyazi (died 1490) in his copy (not “translation”). He (Elijah Bashyazi) has also, according to Elijah b. Baruch, |457| placed the conspectus of chapters at the end of the book. After counting he found them to number thirty-three, and so he marked them with the mnemonic ל”ג השמן (Lev. 14:12). He himself, he says, has put the register at the beginning of the book and indicated the subject of every chapter at its beginning.⁶⁰⁷

Thus there is only one single translator called Tobias, or, to give his full name, Tobias b. Moses, called ha-Oved (Arabic אלעבד, the servant, namely of God), also “the Scholar” (הבקי), from Constantinople, probably the student of Jeshua in Jerusalem (mid-eleventh century?), one of the first translators from Arabic known to us.⁶⁰⁸

Pinsker 1860 (Appendix, 198) gives us the contents of the thirty-three chapters. Generally they tally with the “comprehensive” work (no. 1 in this entry), except for the polemic against the sects, which is not found in the compendium, for that book was directed at those who had no need of it.⁶⁰⁹ Since each of the two books refers to the other, it is impossible to decide which one was composed first. Frankl⁶¹⁰ deduces from one passage of the compendium missing from the more elaborate book that the latter is the more recent version.

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779 (IMHM F 28071).

Oxford, Bodleian Library Ms. Heb.f.12 (Ms.Heb.f.12) (Neubauer 2789) (IMHM F 17286).

Oxford, Bodleian Library Ms. Poc. 213 (Ms. Uri 129) (Neubauer 323) (IMHM F 17242).

Paris, Bibliothèque Nationale héb 670 (IMHM F 11549), 50b–51b.

Paris, Bibliothèque Nationale héb 670/2 (IMHM F 11549), 77a–107a.

St. Petersburg, Institute of Oriental Studies B 342 (IMHM F 53537), 81a–122b.

St. Petersburg, Institute of Oriental Studies B 394 (IMHM F 53555), 44a–63a.

St. Petersburg, Russian National Library Evr. I 687 (IMHM F 51269).

St. Petersburg, Russian National Library Evr. I 688 (IMHM F 51269), 69a–95a.

St. Petersburg, Russian National Library Evr. I 689 (IMHM F 51332), 69a–95a.

St. Petersburg, Institute of Oriental Studies B 241 (IMHM F 53371), 4a–124b.

St. Petersburg, Institute of Oriental Studies IOS B 241/4 (IMHM F 53371), 180a–226b.

St. Petersburg, Institute of Oriental Studies C 103 (IMHM F 69335), 1b–48a.

§269. (Doubtful works.) Here I collect those works whose author or original language are doubtful, continuing my numbering and beginning with those attributed to one of the two Josephs.

3. פרק צדוק הדין (Chapter on Theodicy):

Manuscripts Leiden, BR Cod. Or. 4790 ³ (1858, 227). Paris, BN héb 670b , St. Petersburg RNL Ms. Evr. I 679; St. Petersburg, IOS B 241/5 [formerly belonging to Abraham Mitschri], St. Petersburg, IOS B 241/5; St. Petersburg IOS B 394

The author is not named in the Paris manuscript, copied by Simcha Isaac, who attributes it in his bibliography to Joseph b. Abraham. Pinsker 1860 (Appendix, 198) also takes this to be probable; for Fürst (1862 II, 69) and Gottlober and Chwolson 1865, 176, it is a fact that the book is translated from Arabic, which contradicts my opinion (1858, 227). In the Firkovich manuscript there is a note near the beginning by Elijah b. Baruch Yerushalmi⁶¹¹ who adds to the title, “by R. Aha” (if this is not one of Firkovich’s forgeries).⁶¹² The copyist took it for the Book הגבולים, mentioned by Judah Hadassi (as anonymous),⁶¹³ whose author was, according to him, Nissi b. Noah. [Concerning this author, see Jacob Mann (Falaquera 1935 ), II, 305 and 1413.] ⁶¹⁴ Firkovich refutes this combination with the book הגבולים or חכמת הגבולים in a note; the latter, he says, is a geographical (!) work, whereas our chapter deals with definitions, since a variant reading displays גדרי instead of גבולות, and in fact it contains definitions (according to Firkovich, ninety-four). Our chapter, according to him, refers to ten things that are explained at the beginning of the book (according to Fürst, the ten articles of faith!); it is therefore none other than the book בינת המשכילים or פלס ביאור המצוות, a work on the commandments whose introduction, under the title םירבדה תרשע, |458| is found in manuscripts Firkovich 610 and Geiger 12. Nissi, the purported author, lived in 790 and was the first Karaite to write in Hebrew, his predecessors having written in Aramaic. However, some fragments of this introduction, or rather of an introduction to an interpretation of the Decalogue, as well as a part of the latter, edited by Pinsker, are more than dubious.⁶¹⁵ Schorr⁶¹⁶ supposes interpolations in some philosophical terms; I think the piece is more recent.⁶¹⁷ Concerning the identification of Nissi with a Rabbi Aha (אחא), who is supposed to have been a Masorete,⁶¹⁸ Fürst’s explanation of the error is plausible.⁶¹⁹ The introduction (37) has: אני נסי בן נח באין חי [חיל] ולא כח, הנקרא ר' אח להקריב ניחוח; for אח,40, someone substituted ר' אחא. This kind of rhyme, however, is accepted only in the Franco-German school.⁶²⁰ The passage seems even more suspicious in view of the fact that Firkovich used it to fabricate a title, composed by Nissi b. Noah in 688.⁶²¹ Simcha Isaac seems to have emended it to אחי להקריב ניחוח.⁶²²

Now if the combined discussion of theodicy along with the interpretation of the Decalogue were to be confirmed – on the basis of their styles, which, however, seem very different to me – one would have every reason to date both of them several centuries after Nissi, in whose name Joseph Bagi (beginning of the sixteenth century) already cites a passage from the introduction.⁶²³

Finally, we remark that God’s ten promises listed in the theodicy are perhaps related to the signs of the Messiah.⁶²⁴

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779 (IMHM F 28071).

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4790 (Warner 52) (IMHM F 28074), 13a–41a.

Oxford, Bodleian Library Ms. Heb.f.12 (Ms.Heb.f.12) (Neubauer 2789) (IMHM F 17286).

Oxford, Bodleian Library Ms. Poc. 213 (Ms. Uri 129) (Neubauer 323) (IMHM F 17242).

Paris, Bibliothèque Nationale héb 670 (IMHM F 11549), 50b–51b.

Paris, Bibliothèque Nationale héb 670/2 (IMHM F 11549), 77a–107a.

St. Petersburg, Institute of Oriental Studies B 342 (IMHM F 53537), 81a–122b.

St. Petersburg, Institute of Oriental Studies B 394 (IMHM F 53555), 44a–63a.

St. Petersburg, Russian National Library Evr. I 679 (IMHM F 51336).

St. Petersburg, Russian National Library Evr. I 687 (IMHM F 51269).

St. Petersburg, Russian National Library Evr. I 688 (IMHM F 51269), 69a–95a.

St. Petersburg, Russian National Library Evr. I 689 (IMHM F 51332), 69a–95a.

St. Petersburg, Institute of Oriental Studies B 241 (IMHM F 53371), 4a–124b.

St. Petersburg, Institute of Oriental Studies IOS B 241/4 (IMHM F 53371), 180a–226b.

St. Petersburg, Institute of Oriental Studies B 241/5 (IMHM F 53371), 228a–50b.

St. Petersburg, Institute of Oriental Studies C 103 (IMHM F 69335), 1b–48a.

§270. 4. שאלות (Questions) which Abū Jacob (Yaʽqūb) put to all scholars of the world, “Israelite and non-Israelite,” concerning the central issues of dogma, a work whose Arabic original is not known,⁶²⁵ in Hebrew [Not a few collections of responsa in Arabic by al-BaṢīr have been identified; however, none of those that have been inspected are the original of the Hebrew שאלות.]:

Manuscripts. Leiden, BR Cod. Or. 4779/5 , 104, copy in Trigland. Paris, BN héb 670/4; St. Petersburg, IOS B 67 ; NY, JTS Ms. 3409 . St. Petersburg RNL Ms. Evr. I 682; St. Petersburg RNL Ms. Evr. I 684 Tischendorf 5 (defective).⁶²⁶ YeraḤmiel Fried in Odessa.

The name Abū Jacob may refer to different Karaite authors, e.g., Joseph b. NoaḤ,⁶²⁷ Joseph b. Bakhtawi (?),⁶²⁸ |459| Joseph Qirqisānī, and Joseph b. Abraham, as well as to Isaac b. Bahlul⁶²⁹ who, as a matter of fact, is straightforwardly credited with our book by the Firkovich catalogue, and by Gottlober and Chwolson 1865, 146. Pinsker finally opted for Joseph b. Abraham, and Fürst relays this identification as a fact⁶³⁰; and although I do not see any good reason to attribute the Questions to any other author, there is still no sufficient basis for [attributing them to] Joseph b. Abraham. According to the title in the Leiden manuscript there are thirteen questions. However, they are not numbered, and, in fact, there are twenty-one. The Firkovich manuscript counts fifteen in the title, 15.⁶³¹

The translation whose author is not named uses few Greek words.

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779 (IMHM F 28071).

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779/5 (IMHM F 28071), 133b–38a.

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4790 (Warner 52) (IMHM F 28074), 13a–41a.

New York, Jewish Theological Seminary 3409 (ENA 2771) (IMHM F 53002), 146a–61a.

Oxford, Bodleian Library Ms. Heb.f.12 (Ms.Heb.f.12) (Neubauer 2789) (IMHM F 17286).

Oxford, Bodleian Library Ms. Poc. 213 (Ms. Uri 129) (Neubauer 323) (IMHM F 17242).

Paris, Bibliothèque Nationale héb 670 (IMHM F 11549), 50b–51b.

Paris, Bibliothèque Nationale héb 670/2 (IMHM F 11549), 77a–107a.

Paris, Bibliothèque Nationale héb 670/4 (IMHM F 11549), 121a–34a.

St. Petersburg, Institute of Oriental Studies B 67 (IMHM F 53002), 92a–103b.

St. Petersburg, Institute of Oriental Studies B 342 (IMHM F 53537), 81a–122b.

St. Petersburg, Institute of Oriental Studies B 394 (IMHM F 53555), 44a–63a.

St. Petersburg, Russian National Library Evr. I 679 (IMHM F 51336).

St. Petersburg, Russian National Library Evr. I 682 (IMHM F 51005), 1a–15a.

St. Petersburg, Russian National Library Evr. I 684 (IMHM F 51374), 1a–15a.

St. Petersburg, Russian National Library Evr. I 687 (IMHM F 51269).

St. Petersburg, Russian National Library Evr. I 688 (IMHM F 51269), 69a–95a.

St. Petersburg, Russian National Library Evr. I 689 (IMHM F 51332), 69a–95a.

St. Petersburg, Institute of Oriental Studies B 241 (IMHM F 53371), 4a–124b.

St. Petersburg, Institute of Oriental Studies IOS B 241/4 (IMHM F 53371), 180a–226b.

St. Petersburg, Institute of Oriental Studies B 241/5 (IMHM F 53371), 228a–50b.

St. Petersburg, Institute of Oriental Studies C 103 (IMHM F 69335), 1b–48a.

§271. There are several Karaite authors by the name of Jeshua (not to be confused with Joshua).⁶³² The most famous is Jeshua b. Judah. The Arabic form of his name is very probably Abū l-Faraj Furqan b. Asad⁶³³; he is supposed to have been the student of Joseph b. Abraham and the teacher of the translator Tobias (that is to say, second half of the eleventh century). The chronological problems besetting that teacher have been resolved by recent research (§264).⁶³⁴ Other than this, very little is known about his person; everything that Fürst presents⁶³⁵ is found only in some anonymous works, falsely attributed to him, e.g., in a Hebrew (translated?) commentary to Exodus and Leviticus which is attributed to one Jeshua b. Ali (otherwise unknown), supposedly composed in 1088 (St. Petersburg RNL Ms. Evr. I 588 ). In another place, I have briefly discussed Pinsker’s opinions about the work – which are mutually contradictory.⁶³⁶

5. משיבת נפש (Consolation of the Soul), a dogmatic and ethical work similar to the writings of Joseph b. Abraham, is extant in Hebrew (probably translated from Arabic) in the following:

Manuscripts: Paris, BN héb 670a , St. Petersburg, RNL Evr. I 690 , [St. Petersburg, IOS B 339 , St. Petersburg, IOS B 241a , St. Petersburg, RNL Evr. II A 40 , St. Petersburg, IOS C 103a , St. Petersburg, IOS B 340 .]

St. Petersburg, RNL Evr. I 690 has been redacted by Elijah (b. Baruch Yerushalmi?); he is then the first person to attribute the work to Jeshua. Following after him in this attribution are Simcha Isaac (who copied the Paris manuscript; nevertheless, SimḤa mentions (צ”א 24^b) a book by the same title which, he says, is mentioned by Joseph b. Abraham in his Ne‘imot). It is very improbable that a translator should have given the same title to two different books; therefore, the quotations in the last chapters of the Ne‘imot have been taken to be interpolations. Frankl assumes Tobias to be their author (§267). |460| Curiously enough, no other medieval author seems to know of the work; therefore, we have too little information about it to enter into a detailed discussion.⁶³⁷

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779 (IMHM F 28071).

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4779/5 (IMHM F 28071), 133b–38a.

Leiden, Bibliotheek der Rijksuniversiteit Cod. Or. 4790 (Warner 52) (IMHM F 28074), 13a–41a.

New York, Jewish Theological Seminary 3409 (ENA 2771) (IMHM F 53002), 146a–61a.

Oxford, Bodleian Library Ms. Heb.f.12 (Ms.Heb.f.12) (Neubauer 2789) (IMHM F 17286).

Oxford, Bodleian Library Ms. Poc. 213 (Ms. Uri 129) (Neubauer 323) (IMHM F 17242).

Paris, Bibliothèque Nationale héb 670 (IMHM F 11549), 109a–20b.

Paris, Bibliothèque Nationale héb 670 (IMHM F 11549), 50b–51b.

Paris, Bibliothèque Nationale héb 670/2 (IMHM F 11549), 77a–107a.

Paris, Bibliothèque Nationale héb 670/4 (IMHM F 11549), 121a–34a.

St. Petersburg, Institute of Oriental Studies B 67 (IMHM F 53002), 92a–103b.

St. Petersburg, Institute of Oriental Studies B 342 (IMHM F 53537), 81a–122b.

St. Petersburg, Institute of Oriental Studies B 394 (IMHM F 53555), 44a–63a.

St. Petersburg, Russian National Library Evr. I 679 (IMHM F 51336).

St. Petersburg, Russian National Library Evr. I 682 (IMHM F 51005), 1a–15a.

St. Petersburg, Russian National Library Evr. I 684 (IMHM F 51374), 1a–15a.

St. Petersburg, Russian National Library Evr. I 687 (IMHM F 51269).

St. Petersburg, Russian National Library Evr. I 688 (IMHM F 51269), 69a–95a.

St. Petersburg, Russian National Library Evr. I 689 (IMHM F 51332), 69a–95a.

St. Petersburg, Institute of Oriental Studies IOS B 241 (IMHM F 53371), 250b–68b.

St. Petersburg, Institute of Oriental Studies B 241 (IMHM F 53371), 4a–124b.

St. Petersburg, Institute of Oriental Studies IOS B 241/4 (IMHM F 53371), 180a–226b.

St. Petersburg, Institute of Oriental Studies B 241/5 (IMHM F 53371), 228a–50b.

St. Petersburg, Institute of Oriental Studies IOS B 339 (IMHM F 53458), 16 fols.

St. Petersburg, Institute of Oriental Studies IOS B 340 (IMHM F 53449), 65–77.

St. Petersburg, Institute of Oriental Studies IOS C 103 (IMHM F 69335), 83b–91a.

St. Petersburg, Institute of Oriental Studies C 103 (IMHM F 69335), 1b–48a.

St. Petersburg, Russian National Library Evr. I 690 (IMHM 51333), 22 fols.

St. Petersburg, Russian National Library Evr. II A 40 (IMHM F 64039), 9a–20b.

§272. 6. מרפא לעצם (Healing for the Bones), on God and His attributes:

Manuscripts: Paris, BN héb 670/6 , St. Petersburg, RNL Evr. I 686 , [St. Petersburg, IOS C 103/3 , St. Petersburg, IOS C 69/3 , St. Petersburg, IOS B 241/2 ]. – Rabin. 1886 n. 55 [ = Cincinnati, HUC 848 ?]

Perhaps this is translated from Arabic. Hadassi (chapter 33 and 100, in Hadassi 1836 edition, at the end of f. 98)⁶³⁸ mentions the title; thus it is a work belonging to the first period (before 1148). All that we know about the author is the fact that he lived in Jerusalem, which he says himself, and that he has not visited Babylon or Constantinople. Simcha Isaac (Luẓki) does not know of him; Pinsker attributes the book to Jeshua b. Judah without a valid reason, but with enough [plausibility] for Fürst to gather details about Jeshua’s biography from it. In order to make some use of his own manuscript, Firkovich wants to attribute it to Aharon Abū l-Faraj, or even to a tenth-century author.

This treatise is composed of twelve chapters in the Paris manuscript, but in the Firkovich manuscript, according to Pinsker, of three chapters and twenty-five gates, apart from the introduction. The technical terms are frequently given in Arabic and Greek; the former probably derive from the original, the latter may perhaps be due to the translator.⁶³⁹

Cincinnati, Hebrew Union College HUC 848 (IMHM F 11336), 29 fols.

Paris, Bibliothèque Nationale héb 670/6 (IMHM F 11549), 151a–72b.

St. Petersburg, Institute of Oriental Studies IOS B 241/2 (IMHM F 53371), 160b–25a.

St. Petersburg, Institute of Oriental Studies IOS C 69/3 (IMHM F 69215), 56a–71a.

St. Petersburg, Institute of Oriental Studies IOS C 103/3 (IMHM F 69335), 69a–83a.

St. Petersburg, Russian National Library RNL Evr. I 686 (IMHM F 51373), 20 fols.

§273. 7. שער הצדק (Gate of Justice): St. Petersburg, RNL Evr. I 683 and St. Petersburg, RNL Evr. I 685 ⁶⁴⁰ (six folia in quarto), contain thirty-four questions, but carry slightly different titles. [Another ms. fragment is St. Petersburg, IOS B 67/5. ] The manuscripts name as author חסן b. MashiaḤ, and Firkovich adds that the book is mentioned, along with five others by the same author, by Judah Hadassi (f. 98 ³), but he does not notice, or does not want to notice, that these works are mentioned in chapter 258, and the author in chapter 257.⁶⁴¹ In fact Simcha Isaac names this as the title (f. 26) of an anonymous work mentioned by Hadassi, without indicating the author. To me, this title seemed to be that of a chapter rather than of a book, since a part of the extensive book by Joseph b. Abraham (not a single chapter, as in Pinsker)⁶⁴² is indicated in the same way. Hadassi’s remark may then refer in general to all works treating of the subject of theodicy, e.g., supra, no. 3.

I know too little about this small treatise to decide whether it has been translated from Arabic; but as it seems, the Karaites employed that language when discussing dogmatics, at least in the first period.

The name of the author, as it is assumed on the basis of an erroneous combination, is corrupt almost throughout; Pinsker and Fürst have not taken heed of the evidence that I presented. Ḥasan,⁶⁴³ or, in the diminutive, |461| Ḥusayn (חסין)⁶⁴⁴ b. MashiaḤ was, according to Sahl,⁶⁴⁵ a contemporary of Saadia and probably wrote in Arabic.

Ḥasan ben MashiaḤ (attributed to), Sha‘ar ha-ẓedeq

St. Petersburg, Institute of Oriental Studies B 67/5 (IMHM F 53002), 103b–7b.

St. Petersburg, Russian National Library Evr. I 683 (IMHM F 51005), 15b–20a.

St. Petersburg, Russian National Library Evr. I 685 (IMHM F 51005), 15b–9b.

Endnotes to Jewish Authors

|999–1000| Endnote 29 (to 386, n. 118) Choice of Pearls [In this endnote Steinschneider matches up the aphorisms from published collections, including his book of translated Hebrew poetry, Mannah 1847, with those of the Choice of Pearls in Ascher’s edition. The abbreviations are explained above.]

• (A) My Manna, the chapter of Ascher’s edition of Choice, followed by the aphorism number therein [I.e. ‘§87 = 1.3’ should be read as ‘ Manna , aphorism 87, corresponding to Choice , ch. 1, aphorism 3. ] Manna, §87 = 1.3; §88 = 1.13; §89 = 1.16; §90 = 1.28; §91 = 1.29; §92 = 1.36–37; §93 = 1.42; §94 = 1.53; §95 = 1.58; §96 = 1.57; §97 = 1.59; §98 = 1.65; §99 = 1.66; §100 = 3.79; §101 = 3.107; §102 = 4.111; §103 = 5.117; §104 = 20.258; §105 = 21.268; §106 = 25.229; §107 = 25.281; §108 = 27.310; §109 = 29.315; §110 = 31.332; §111 = 32.328; §112 = 32.332; §113 = 32.317; §114 = 32.350; §115 = 32.352; §116 = 32.355; §117 (תעלות); §118 36.376; §119 (תוספת); §120 = 39.404; §121 = 43.456; §122 = 44.502; §123 = 44.523; §124 = 44.516; §125 = 47.563; §126 (Dz. 6).

• (B) KimḤi (see above, p. 384) [the chapter numbers from Ascher’s edition are omitted]

  • (a) Dz. [1842]: §1 = ? (Pr.: [1990] 8, 9 fol. 11, Dukes 1851, 64); §2 = ?; §3 (incorrectly listed חרכהה רעש) = 69; §4 = 61; 5 = ?; §6 = 74; §7 = 414–15 (Choice, p. 166); §8 = ? (not 36–37); §9 = 60; §10 = 208; §11 = 547? (the first line is not in ED 1852, 6; §12 = 522; §13 = 523; §14 = 547? 15 = ? (Pr. 9: 19, f. 13); §16 = ? (§16b = ?); §17 = ?; §18 = 601 (cf. 460, 260, where the text is corrupt); §19 = 310 (read: המכה תרזג?); §20 = 436; §21 = 338; §22 = 354.

  • (b) Dl. [1846b]: §1? (Dukes 1853, 55, Dukes 1844, 536, incorrectly paginaged as 542, from ןוגיה ירצ, not in the Cremona edition); §2 = ?; §3b, §4 = ?; §5 = 12?; §6 = 13; §7 = 15 §8 = 16; §9 = 17; §10 = ? (Dukes 1851, 8); §11 = ?; §12 (is the final strophe); §13 = 93; §14 = ?; §15 = 165? §16 = ?; §17 = ?; §18 = 549; §19 = ?; §20 = 178; §21 = ?; §22 = 194; §23 = 204; §24 = 136; §25 –; 27 = ? §28 = 756; §29 = 398; §§30–32 = ? §33 = 370.

  • (c) Dll. [Dukes 1850c]: §1 = 29; §2 = 624 (also Pr. 12: 9); §3 = ?; §4 – ? (ED. 11); §5 = 50; §6 = 58; §7 = ?; §8 = 544.

  • (d) ED. [H. Edelmann 1852] [§§1, 2, 4, 5 (Dll. 7) 11, 16, 19, 22, 24, 25, 26, 28, 29 in Dz. 1 and 4 in Dl. 3 (Paraphrase) = 202; §6 = 63; §§7–10 = ? §12 = 156–157; §§13, 14 = ? §15 = 49; §17 = 264; §18 = 266; §20 = 363; §21 = 364; §23 = ? §27 = ? §30 = 366; §31 = 367; §§32, 33 = ?

  • (e) Pr. [KimḤi 1990 ] The aphorisms are numbered according to their sequence in KimḤi’s commentary on Proverbs, with reference to chapter and verse [and page number in the Talmage edition KimḤi 1990 ]; then they are matched to other editions and to the Ascher edition of Choice: §1, 2:2, 11 Dz. 7; §2, §3, §4 םדא יח (Dl 4, 5, 12, or 15?); §5 השעמ ילב המכח = Ascher, 59. §6, 2:11, 6 = ?; §7 שיא לכ = 117; §8, 3:16, 17 = 7.25, 7.30 8? 9?; §10, 4:22, 25; §11, 8:11, 37 Dz. 1 §12, 10:19, 46 Dz. 15; §§13, 14, 11:2, 49 = 14.328–332 [11:15 רמא םירישה רש is not Choice?] §15, 11:2, 49 = 14.316–317; §16, 11:16, 52 = ib?; §17, = ib. 518. §18 11:25, 54 ED 12 §19, 11:25 = ib.158; §20, 12:9, 59 (Dukes 1850a, 507, Dukes 1853, 49 n. 31, where in Sarsa? The anecdote of the King precedes it; see Steinschneider 1870f, §21, 14:12, 70?; §24, 14:30, 74 – 21.594 (Dukes 1850a, 378; §25, 16:32, 86, Dl 13; §26 ib. (?); §27, 18:23, 94; 28 ךתלוזב, 20:6 = 27? (Dukes 1850a, 389); §29, 22:1, 111, prose, = 540 (see 387, n. 122 above [where Steinschneider corrects Ascher’s text to read “8000” dinars rather than “80,000”; the proverb here has “1000 gold dinars.”]) ; §30, 24:21, 121 = 32.366; §31, 25:14, 128? (corrupt); §32, 25:17, ib. = 405; §33, 27:19, 138 (Dukes 1850a, 391); §34, 29:19, 145 = 244.

    In addition, a note to no. 86 (read במעלות הבט ;(?ממדות טובות Dukes 1853 49, no. 30.

• (C) Ibn Gabirol’s Ethics [Wise’s comparison between the aphorisms in Ibn Gabirol’s Ethics and the Choice of Pearls in Wise 1901, 108–113, supercedes Steinschneider’s comparison here.]

Endnote 30 (to 428, n. 410) [In this endnote, Steinschneider gives various Hebrew terms for the Aristotelian] Categories. (Key: ‘a’ refers to Ru’aḤ Ḥen, ch. 10; ‘b’ to Emunot ve-De‘ot II, 2, according to Judah Ibn Tibbon’s translation; ‘c’ to Samuel Ibn Tibbon’s Glossary, s.v. איכות, ‘d’ to Moses Ibn Tibbon’s translation of Maimonides’ Logical Terms; ‘e’ to Jacob b. Makhir’s translation of Averroes’ Epitome on Logic; ‘f’ to the translation of Abraham Ibn Daud’s Emunah Ramah Ibn Daud 1852, f. 5 ; where מצב is counted fourth; as a result, our numbers 4–6 are counted there 5, 7, 6; cf. the anonymous work cited above on 500. [Actually, this statement appears to refer to n. 205 on 500, which discusses the Kelalei Higgayon of David Ibn Bilia found in the Bodleian, Oxford Mich. 88, ff. 104a–129a; there the order of the Categories is also slightly different from the standard one]; ‘g’ to the translation of Moses Ibn Ezra’s Arugat ha-Bosem Dukes 1843a, 118, where our nos. 5–8 are rearranged as 6, 5, 8, 7, cf. Kaufmann 1877a, 64.

1. עצם – everywhere, 2. a, b כמות, even כמה, g ספירה. – 3a etc. איכות, d איך, g תואר. – 4 a, b, d מצטרף, c הצטרפות והמצורף, e f g הצטרפות – 5 a etc. מתי, b זמן, g עת, in b, g as 6. – 6 a etc. אנה, b מקומות, g מקום, in b f g as 5. – 7 מצב everywhere, in b as 8. – 8 ול, b and g (as 7) ןינק. – 9 a, etc. לעפיש, b and g לעופ: c and e place 10 before 9. – 10 a etc. שיתפעל, b נפעל, g פעול. – In Saadia Commentary on Sefer Yeẓirah there are various translations: יסוד, כמויות, איכות, טפול, (Dukes 1853, 25, Kaufmann 1877a, 141), מקום, קניה or קנין, פועל, נפעל, or, הרכוש והקניה מופעל,יחס (Dukes 1853, 3.)

Endnote 31 (to 436) Variants to Maimonides, מאמר הייחוד, see Steinschneider 1892, 86.

Endnote 32 (to 446) Pseudo-Saadia on פר“ס. Steinschneider 1892, 79; Epstein 1892, 75.