TEXTS AND STUDIES OF THE JEWISH THEOLOGICAL SEMINARY OF' AMERICA, VOL. I GEONICA BY LOUIS GINZBERG The Library University of California, GEONICA OXFORD : HORACE HART PRINTER TO THE UNIVERSITY TEXTS AND STUDIES OF THE JEWISH THEOLOGICAL SEMINARY OF AMERICA, VOL. I GEONICA BY LOUIS GINZBERG I THE GEONIM AND THEIR HALAKIC WRITINGS NEW YORK THE JEWISH THEOLOGICAL SEMINARY OF AMERICA 1909 >M TO THE MEMORY OF ISAAC LEESEK FOUNDER OF THE FIRST AMERICAN COLLEGE FOR HIGHER JEWISH LEARNING THIS FIRST PUBLICATION OF THE JEWISH THEOLOGICAL SEMINARY OF AMERICA IS DEDICATED 1252856 PREFACE THE centuries between the final redaction of the Talmud and the beginning of Jewish culture in the West is one of the most obscure periods in the history of the Jews of post-Biblical times. If we regard the literary productiveness of a people as the only standard by which to measure its culture, then we must confess that /this was a period of decline ; the Geonic epoch has not brought forth one monumental work. Yet, a period which has produced such powerful religious movements as Karaism and mysticism, and has for the first time made a serious attempt to/ harmonize Hellenism with Talmudic Judaism cannot be considered as stagnant. The first step towards a correct under- standing of this period must be a clear comprehension of the institution which gave it its name : " the Gaonate." With the exception of K. Saadia,/who flourished toward the end of this period, we meet with no name of the first magnitude. But, the less important the Geonim were in themselves, the more important must have been the Gaonate to be able to impress its stamp upon several centuries. The fundamental question which we have to answer before we proceed to form an estimate of this period is : Were the Geonim only heads of Academies, or were they representatives of authoritative bodies ? The first volume of this book presents some material towards the solution of this question. Granted that we will never be able to form an adequate picture of the activity of the Geonim, for the contemporary sources are too meagre for this purpose, yet I hope to have shown that/the Gaon was more than the president of a scholastic institu- tion. The results of my studies are mostly directed Vlll PREFACE against the conception of the Gaonate as formulated by Isaac Halevy in the third volume of his Dorot harJRishonim (Pressburg, 1898), according to whom the Academies were only Talmud-schools, and the Geonim Talmud teachers. In spite of all his Kab- binic erudition and extraordinary critical acumen Halevy has contributed but little towards the under- standing of the Gaonate. His bitter attacks upon men like Kapoport, Frankel, Weiss, Graetz, and other Jewish scholars are but poor compensation for the lack of positive results. In accordance with my conception of the Gaonate as an authoritative body, I have, in dealing with the literary activity of the Geonim, confined myself to their HalaHc writings, since it is only in the Halakah thaj/the authority of the Geonim found its full expression. In the chapter, "The Halakic Literature of the Geonim" (pp. 72-205), I have given a survey of the literary activity of the Geonim along the different departments of the Halakah: Codification, Talmud exegesis, Eesponsa, and Liturgy. I hope that my investigation about the Seder E. Amram (pp. 119-54) will interest even those to whom the Halakah is either a terra incognita or a noli me tangere. Upon no other department was the activity of the Geonim so decisive and im- portant as upon the Liturgy, yet even this branch of research remained uncultivated. Conscious of the fact that in many respects I have chosen a way which not all will be ready to follow, I only claim credit for having undertaken anew the examination of some important questions relating to the history of the Geonim, which may lead others to study this very obscure period of Jewish history. A considerable part of the material utilised in my representation of the history and literature of the Geonim is taken from the Genizah. There is no PREFACE IX exaggeration in maintaining that the discovery of the Genizah by Prof. Solomon Schechter was in no other department of Jewish learning so epoch- making as in the history of the Geonim. Prof. Schechter's Saadyana (Cambridge, 1903) is a fair specimen of what we may expect from the Genizah for the understanding of the Geonic period. Indeed it is a veritable treasure trove for the history of this period. New Halakic material, however, has not been brought forth from the Genizah till now, and yet no one will doubt, except those who are given over to philological trifles or theological sophisms that it is the Halakah alone which gives us a true mirror of that time. Especially is this the case with the Responsa which deal with life in all its aspects. They enable us to penetrate into the study of the scholar as well as into the home of the everyday man. The second volume consists of Halakic Frag- ments from the Genizah now stored in the Taylor- Schechter collection in the Cambridge University Library, and in the Bodleian at Oxford l . The first thirty-eight fragments are Geonic Responsa 2 , which hitherto were entirely unknown, or which differ in some way from the form in which they have been known. I have disregarded such Geonic Responsa from the Genizah as are identical with those previously printed as well as those which are written in Arabic. With the exception of a few very badly damaged fragments, this book contains nearly all the Geonic Responsa from the Genizah in the above-mentioned libraries. The Fragments coming from the Bodleian were copied by myself, and I can therefore confidently 1 Comp. Index s.v. Mjw'n'u and froayv'vtrc. Pages 1-165 were first published in the Jewish Quarterly Review, XVI-XX. * Frag. XXXIV is a part of R. Nissim's Mafleah, which I have incor- porated in this book, as the Mafteah is mainly based on Geouic Responsa. X PREFACE vouch for their correctness in reproducing the original. For the copies of the Cambridge Frag- ments I am indebted to Ernest Worman, M.A., Cambridge. The Fragments reproduced here line for line, page for page, are preceded by short introductions describing the manuscripts and the nature of their varying contents. I have made it a point to call the reader's attention to certain interesting Halakic views expressed in the Fragments. I was brought up in surroundings where the understanding of the Halakah was the chief subject of Jewish learning, and even now I cannot free myself of the view that the Halakah ought to be no less important than the correct spelling of an Aramaic preposition. The Appendix to the second volume contains nine Fragments (XXXIX-XLVII) of the She&tot and Halakot Gedolot. The importance of these Frag- ments in the study of the early Geonic literature is fully dealt with in the first volume (pp. 91-3, 108-9), and also in the introductory note (pp. 349- 52) preceding them. To facilitate the use of the Fragments I have added two Indices. The first, arranged according to the Slmlhan *Aruk, gives the subject of the Kesponsa; those containing explanations of Tal- mudical passages are indexed at the end of this in accordance with the order of the Talmudical treatises. The second index is alphabetical, and deals with the historical or philological matter found either in the text of the Fragments or in the notes and introductions accompanying them. I desire to acknowledge my indebtedness to the authorities of the Cambridge University and Bodleian Libraries for courtesies shown me in connexion with the present work. CONTENTS THE GEONIM I. THE GAONATE. PAGE Palestine and Babylonia i Salient Features of the Gaonate 6 Friction between the Exilarchate and the Gaonate of Pumbedita 14 The Language of Nathan ha- Babli's Report .... 22 Nathan ha-Babli Identified 29 Nathan ha-Babli the Source for the Two Reports about the Babylonian Academies 34 The Supremacy of Sura 37 The Title Gaon originally the Prerogative of Sura ... 46 The Origin of the Gaonate under the Mohammedan Rulers . 52 Nathan ha- Babli's Account of Ukba . . . . 55 The Last Conflict between the Exilarchate and the Pumbedita Gaonate 62 The Predecessor of Saadia 66 The Chronology of the Geonim 69 II. THE HALAKIC LITERATURE OF THE GEONIM. Halakah the Main Feature of Geonic Literature ... 72 The Impulse to Geonic Literary Activity 73 Rabbi Aha of Shabha 75 The SheSltot and the Yenishalmi 78 Plan and Purpose of the SheSltot 86 Rabbi Jehudai the Earliest Halakic Writer in Geonic Times . 95 Conflicting Traditions about the Author of the Halakot Gedolot 99 Jehudai Gaon Author of the Original Halakot Gedolot . . 103 Later Amplifications of the Halakot Gedolot . . . . 108 Plan and Purpose of the Halakot Gedolot 1 1 1 Codification not Favoured 117 Prayers First Put in Writing 119 The Liturgical Part of the Seder Bab Amram . . . . 123 XU CONTENTS PAGE The Halakic Part of the Seder Bab Amram . . . . 144 Relation of the Manuscripts to the Printed Text . . . 151 Spurious Works attributed to the Geonim Nahshon and hia Son Hai 154 Works attributed to the Geonim Zemah, Hai ben David, and Hilai 159 The Importance of Rabbi Saadia in Halakic Literature . . 162 The Three Great Successors of Rabbi Saadia . . . . 167 Anonymous Codes of the Geonic Time 177 Origin of the Responsa Collections 182 The Importance of the Geonic Responsa 200 LIST OF ABBKEVIATIONS OF TITLES OF BOOKS . 206 ADDITIONS ....... . . 207 1. THE GAONATE. PALESTINE AND BABYLONIA. "'THE staff shall not depart from Judah' the Exil- archs who govern the people with the ruler's rod ; ' nor a lawgiver from between his feet' the descendants of Hillel who instruct the people in the Torah." This brief, vivid characterisation of the two great Jewish institu- tions of the Talmudic time, by a Jewish sage living at the beginning of the second century 1 , remained no less true in the centuries that followed. In spite of friction / now and again between the later Patriarchs and the intellectual leaders of the Palestinian Jews 2 , the dissension never reached the point of causing a separation of the Cspiritual power from/ the worldly power] in Palestine. Though the Patriarchs were not always the actual presiding officers of the chief academy, de jure they were looked upon, in Palestine and outside, as the spiritual heads of the Jews. For instance, the last important achievement that / may be credited to the account of the Jewish scholars of Palestine, the fixation of the calendar, in the middle of the fourth century, is closely connected^/ with/the name of the Patriarch Hillel II, and, as late as the second half of the same century, the surveillance of religious conditions in [the Diaspora still lay in the hands _of the Patriarch, Jas we may learn from the account of a Christian author of the time. The Patriarch dispatched 1 Sanhedrin, 5 a ; this anonymous Baraita must have originated in the time of the Patriarch Rabbi Judah I ; the earlier Tannaim make no sort of mention of the Babylonian Exilarchs. * Comp., for example, Yer. Sanhedrin, beg. of second chapter. I B 2 THE GEONIM messengers, "apostles," not only for the purpose of col- lecting moneys, but/ also, in the words of Epiphanius 1 , " to maintain the observance of the law, and dismiss unfit archisynagogues, priests, presbyters, and ministers." In /Babylonia conditions were vastly different. From the earliest time there had prevailed a sharply marked JNdualismJ The Exilarchate, which/ could count upon the support of the non-Jewish government, was a political power and nothing more. It permitted no interference in its province, either from within or from without 2 . Beginning with the early years of the third century, the scholar's estate developed more and more into an essential element in the life of the Babylonian Jews, though it lacked a unified expression of its authority. There were, indeed, the Academies, especially the two great central 1 Epiphanius, Adv. Haer., XX, 4 and n, on the Jewish apostles. For details, comp. Harnack, Die Mission und Ausbreiiung des Christenthums, 237-40 ; Krauss, J. Q. R., XVII, 370-83 ; and Vogelstein, Monatsschrift, IXL, 427 et seq. Apparently the Babylonian Geonim followed this example and sent out apostles. Rabbi Nehemiah Gaon speaks, in his letter dated 962, of mbnpn btf nrnno rrus i:bn> TpDn munn jpin rrobtn 1*0 i:Tp' T by i:nVro n:m (J.Q.R., XIX, 106). Likewise Rabbi Samuel Ibn Hofni speaks of nrpBi ru'c'n po in one of his letters (J. Q. R., XIV, 308). This expression conveys the notion that the office of TpD was an old institution. In the year 750, we find Abi 'Ali Hassan, of Bagdad, as "the head of the congregation '' of Fostat (J. Q. R., XVII, 428). The idea suggests itself that he was sent upon a mission by the Babylonian authorities. In another Genizah fragment, J.Q.R., XIX, 740, ir^n Dioy 'n, "the apostle Eabbi Amram " is mentioned, who, however, seems to have been deputed by the Palestinian Geonim. On the other hand, Rabbi Eleazar Alluf, in Babylonia in 850, who gave the Geonim information about Spain, ' was not a returned emissary of the Geonim, but a native Spaniard, as we learn from the description of him in Harkavy, 201, and Schechter, Saadyana, 76: I:WD^I pi ^rac* n no -a rpto niyto an no 3n mrr nai. He went to Babylonia, and probably took up his residence there, for we find him there in 875 (Harkavy, I.e.). The custom of the Academies, discussed in G. S., p. 302, of disposing in the month of Adar of the questions submitted to them from all parts is probably connected with the dis- patching of the messengers, as the Patriarchs also sent their apostles out in this month, according to Krauss's correct observation (1. c., 374, note 4). a Sanhedrin, 5 a ; Yer. Baba Eathra, V, end ; and elsewhere. THE GAONATE 3 bodies at Sura and Pumbedita, but they wanted the means / of making effective powers of themselves. The Academy in Palestine, situated in the town in which/the Patriarch resided, was the highest (court of justic^ no matter who and what the president might be at a given time, thus in a measure representing the old Synhedrion l . In / Babylonia, on the other hand, the Qmportance of an Academy! depended upon /the learning of the presiding chief. So long as Rab Huna and Rab Hisda were con- nected with the Academy at Sura, it was in the lead, and Pumbedita was pre-eminent when/ it could boast of a R-abbah, a Rab Joseph, and other scholars of equal note. Yet, however brilliant the respective representatives otfthe Academies might be, neither of them could lay claim to exclusive authority. For instance, when the Academy at Sura, under the leadership of Rab Huna, was enjoying its palmiest days, many a scholar, like Rab Nahman and Rab Anan, refused to subordinate himself to its rulings 2 . / This was exactly as it should have been. The truth of the popular saying, Knowledge is power/Jjhas been verified abundantly in/the course of Jewish history. Since the destruction of the Jewish State, it has been Jewish know- ledge that/has always kept the Jews together, though they were scattered over all the continents. But to be a power, intellectualism must clothe itself in a concrete form/and for this there was no provision in the Babylonian Academies, 1 SanJiedrin, 31 b, where irirr nu is not, as Rashi holds, some place or other at which scholars foregathered, but the Academy over which the Patriarch presided, as may be seen plainly from Yer. Berakot, IV, 7 d, and Yer. Sanhedrin, II, beg. a Comp. Kelulot, 693, where Rab Anan addresses the head of the Sura Academy as jnan win, which evokes many an unpleasant i-emark. Rab Nahman also speaks of pin N:in, and, as Rabbi Sherira, in his Letter, 32, 13, observes with fine insight, Rab Nahman did not acknow- ledge the head of Sura as an authority superior to himself. Also the passage Kiddushin, 70 a, throws light upon the relation subsisting between Rab Nahman and Rab Huna. He did not consult with the latter when he cited Rab Judah, the chief of the Pumbedita Academy, before the court. 2 4 THE GEONIM as /long as they were purely (spiritual centresjdestitute of every vestige of temporal authority. Keeping this state of affairs in mind, we cannot find it surprising thai/ the Babylonian Academies were not yet able to take the place, as they afterwards did, of those in Palestine, when the latter entered upon a period of rapid decline, beginning with the dominance of the^religion of lovejthe adherents of which extirpated the Jewish culture o^the Holy Land with fire and sword 1 . The importance of the Babylonian Academies dates from /the so-called Geonic time. To be accurate, it is about/the end of the seventh century that they begin to appear as the paramount (power of the whole of the 1 In the Geonic time, the superiority of the Babylonian Talmud was acknowledged even in Palestine, in connexion with which the Responsum reproduced in G. 8., pp. 50-3, is of interest. Its author was a Palestinian scholar in the latter half of the eighth century, who, in his discussions, refers only to the Babylonian Talmud and the Gaon Rabbi JehudaL Also in the Ben-Mei'r controversy the Palestinians appeal to the Babylonian and not the Jerusalem Talmud. Rabbi Paltoi, y*e, 63 b, 40, expressed himself very harshly concerning certain Palestinian customs : p2'H avra NDbw DJTTO Dico nb mrni rv'yi 'inp an -im NITD py prim. His words give poignant expression to the decay of Palestinian supremacy in Babylonia. The last demonstrable case of Babylonians applying to Palestinians for a decision is that mentioned in Hullin, 59 b, for the Rab Samuel ben Abbahu of this passage is the Sabora of that name, who, according to Rabbi Sherira's statement, Letter, 34, 18, died in 505. Neubauer's text has the incorrect reading rrnrr 11 instead of iron % as Wallerstein has it. On the other hand, Neubauer's reading in the previous line, 'NOIIT:, is preferable to nairn, as appears from MSS. M and O of 'Enibin, na, which have tiiro, while in the parallel passage, Menahot, 33 b, MS. M also reads iDim a corroboration of Rabbi Sherira's statement that the name has been transmitted in two forms, n, the third in rank 1 , enjoyed the privilege. In a Responsum, probably from the hand of Rabbi Natronai 2 , printed in G. S., p. 31, we have the following: NB>n t6n w^ ana N^ni rvb mm t&3 pn xhx " That he [Rabbi Simonai] did not write you regarding this question is due to the circum- stance that he was not the head [of the Academy], but only the Resh Kalla 3 ." Even in a case like the one dealt with in the Responsum under consideration, in which the countrymen in Spain ; comp. D*n, 130, and y"c, 26 b, 23. Rabbi " Asaph " (J. Q. R., IX, 689, top) is not to be emended to Joseph ; he is the Rabbi Asaph who was the VID 'T during the Gaonate of Rabbenu Hai ; comp. R.&J., LV, 50. His opinion was probably given orally to Rabbi Elhanan. Notice that in J.Q.R., 1. c., he is called simply 10, while the authorities preceding and following him bear the title Gaon. 1 Besides "the seven mto >tto" (Rabbi Nathan, in his report, 87, i6j, the title of the seven most prominent members of the Academy, there must have been also " the xba rcn," who took an active part in the instruction given at the Academy. It seems that Rabbi Hai occupied this office before becoming -\"yn ; comp. Saadyana, 118. I do not know whence Harkavy, Saadia, 144, note 7, derived his statement that Rabbi Hananiah, the father of R. Sherira, became Gaon only after having occupied the offices of D"T and VaN. 2 Comp. 3*n, 15, and 'rou.'N, III, 49. 3 The subject of 3.13 may possibly be Rabbi Haninah, so that the passage would read, "that he [Rabbi Haninah] did not write it to you [that the Nto 'i was of his opinion] is due to the fact that, &c." In any event, the inference to be drawn from the passage is that the 3*1 replied to no question, and even in a case like the one under consideration, the Gaon made no mention of him. THE GAONATE 9 testimony of the Resh Kalla was of importance, the Gaon does not refer to him with a single word. The Amoraim had found it unbecoming conduct in the Patriarch Rabbi Simon ben Gamaliel that, using the singular in a formal announcement, he failed to include his colleagues (Sanhedrin , na-b). What would they have thought of the official style of their successors, the Geonim ? Personal arrogance, it need not be said, can be charged neither against Rabbi Simon nor against the Geonim. In a college of scholars, the presiding officer is primus inter pares, but the Patriarch in early times, and later the Gaon, were the representatives of an institution that acknowledged one head alone 1 . In attempting to appraise the Gaonate, the transmission of the office from member to member in a limited number of families, is a most suggestive feature 2 . During the last three centuries of the Geonic period, or what was the Geonic period properly so called, we have, for example, the following data concerning the Gaonate of Pumbedita. The Gaon Dodai (761), brother of the celebrated Gaon Jehudai, bequeathed his office to his son Rabba, and no less than six of Rabba's descendants occupied the position after him his grandson Joseph ben Mar Rabbi and his great- grandson Mattathias in one line, and in another line four of his descendants belonging to successive generations, Judah, Hananiah, Sherira, and Hai, the first of them representing the fourth, or perhaps the fifth generation removed from Rabba 3 . Out of a total of 277 years, Dodai and these descendants of his enumerated here occupied the Gaonate 102. 1 There are cases on record which the Geonim decided in opposition to the opinion of the Academies, see Nahmanides, Milhemet, Kiddushin, 9, and n"j, 82, 226. The frequent references made by the Geonim to the customs of the Academies are to be taken not as marks of respect shown to colleagues and disciples, but rather to the institution as such. 2 The data upon the Geonim families that follow, unless other references are given, are taken from the Letter of Rabbi Sherira as their sole source. 3 Comp. below, pp. 70-1, on the de ree of kinship between Rabbi Judah and Rabba. 10 THE GEONIM Besides this prominent family, claiming Davidic descent, there was another family of Geonim of great influence, the priestly family 1 to which belonged Rabbi Abraham Kahana (about 75)) i n all probability the successor to his brother Natronai 2 . Rabbi Abraham himself was fol- lowed first by his son Hanina and his grandson Kahana. and then by his other son Abumai. Furthermore, the Geonim Ahai 3 , his son Kimvi, and his grandson Mebasser, seem to have been descendants of the same Rabbi Abraham. Sherira, our only source, was not interested in family relations, except as his own were affected, and whatever information we glean from him upon the subject he gives incidentally. There is no telling, therefore, to what extent the above Geonim families were interrelated among them- selves 4 , or how those Geonim who now appear isolated, outside of the charmed circle, are really connected with it. For instance, we are not acquainted with Rabbi Zemah ben Paltoi's relation to the Geonim families, but Sherira tells us by the way that he gave his daughter in marriage to Rabbi Judah Gaon, the grandfather of Sherira. In Sura the Gaonate was in the almost exclusive pos- session of three families for a period of about two centuries. The Geonim Mari (777), Hilai, Natronai, Hilai. Jacob, and Joseph 5 (942), belonged to one family ; Zadok (823), Kimoi, Nahshon, Zemah, and Hai (889), to the second ; and the third 1 In connexion with this, it may be mentioned that the Palestinian Gaonate also was in the hands of a single priestly family. 2 Comp. below, pp. 21, 41, where arguments are given in favour of this conjecture. 3 Perhaps Rabbi Kohen-Zedek and his son Rabbi Nehemiah, Geonim of Pumbedita, as well as the grandson of the former, Rabbi Samuel ben Hofni, belong to the same family as Rabbi Mebasser, so that the quarrel between the last and Rabbi Kohen-Zedek, both of whom are described as Kohanim, was between two branches of the same family. Rabbi Nehemiah (J. Q. R., XIX, 105) seems to allude to his origin from a Geonim family in the words i;vneiro Vrun. * Rabbi Hezekiah ben Samuel (J. Q. R., XVIII, 402) reports that he was descended from a Sura as well as a Pumbedita Geonim family. 5 In Harkavy, Saadia, 228, he is called D' 1 :''**; p p*o. THE GAONATE II was the priestly family which furnished the Gaonate with four incumbents, Jacob (801), Abimi, Moses 1 , and Kohen- Zedek (845). Whatever view may be held on the subject of hereditary genius, it cannot be applied to the case in hand. Among the Geonim it must be admitted that it was not always intellectual force, but rather the office, that was transmitted from one member of a family to another. What explana- tion could otherwise be offered of the circumstance that during the whole extent of the Amoraic period a single instance occurs of father and son, Rab Ashi and Mar, being presidents of an Academy, while the Gaonate was controlled by a few families throughout its who^e history ? There is no intention of blinking the fact that the claims of sons upon the offices and dignities of fathers have always received somewhat more than due consideration among the Jews since the most ancient times 2 . But this would still leave the frequent succession of the Gaonate from brother to brother unexplained 3 . For instance, Jacob and Abimi, brothers, were Geonim, and so were Zadok and Kimoi, though the father of neither pair had been in office. It remains, then, to explain the close transmission of the Gaonate only by the assumption that it came to be looked upon as the prescriptive right of certain influential families. The same explanation would cover the phenomenon that the Ab Bet Din, the Resh Kalla, and the secretary of the Academy, so far as we know about them, also belonged to the Geonim families mentioned above 4 . 1 That Rabbi Moses was a son of the Gaon Rabbi Jacob is obvious from the Genizah fragment published in G. S., p. 214. 2 Comp. Sifra, Afiare, 83 b, ed. Weiss, and Midrash Tannaim, ed. Hoff- mann, 106. 3 An interesting analogue to this succession by brothers is offered by that of the high priests in the Herodian time ; comp. Biichler, Priester und Cultus, 107 et seq. * Of the -Tax, we know only six by name : Rabbi Joseph ben Mar Rab (Letter of Rabbi Sherira, 38, ia\ Rabbi Zemah (comp. G. S., p. 203), Rabbi Tob (J. Q. R., XVIII, 402), Rabbi Hofni, father of Rabbi Samuel 12 THE GEONIM In this respect the Gaonate approached the[institutions of the Patriarchate and Exilarchate,) which/were the preroga- (J. Q. B., 1. c.), Rabbenu Hai, and Rabbi Abraham (R. E. J., LV, 52). All these, with the exception of the last, of whom we know nothing, were members of Geonim families, and three of them became Geonim them- selves in view of which it is hard to understand how Halevy, 1. c., 266, can maintain that the i"i succeeded to the office of Gaon only in extremely rare instances. The three whom we may be said actually to know. Rabbi Joseph, Rabbi Zemah, and Rabbenu Hai, occupied the Gaonate. Indeed, in two passages, Rabbi Sherira (38, 12 and 15) remarks how extraordinary it was that the Y'lN Rabbi Joseph was disregarded in filling the Gaonate, upon which he had a claim by virtue of being n"a . What the duties and the nature of the office of the ViN were, it is difficult to determine now. Its importance is attested by the fact that certain announcements and regulations were provided with the official seal of the Exilarch, the two Geonirn, and the two Y"iN, as we know through Rabbi Natronai, 'Ittur, I, 44 d. Another Geonic Responsum by Rabbi Natronai, or by his colleague of Pumbedita, Rabbi Paltoi, in j"n, 20, also speaks of the 'iTur WCTO pn TQ nyaiN, "the four courts of justice of the two Academies," that is, the courts of the Geonim and of the Y'ix, and in Harkavy, 187, we find the two courts presided over by Sherira as Gaon, and Hai, his son, as Ab Bet Din, described as TO u'Vru D':n 'm :c 'jN-mr 1 ; while from the Genizah fragment published in G. S., p. 386, we see that only the court presided over by the Gaon was called the m Vnjn JH. Apparently it was a courtesy extended to Rabbenu Hai personally, to give the appellation to his court in spite of its lower rank. The expression nytorr irtr, or its Aramaic equivalent, Nruv.an naa, is identical with Vn;n jn rn, as can be seen from Harkavy, 156 and 215, and i"cn. II, 31. The n"; was, as is well known, sai n rt, which stands for NraTrr; aa H NTI. The chief judge of the Exilarch was also called N33 n ri, in his case shortened from unnoi *m n *H, which office, it is needless to say, has nothing in common with the other in spite of the similarity in the names of the two offices. We are equally at sea as to the position of the Nta '~\ . Apparently the N"?O 'TI I"IN jwj of the Geonic time have some sort of correspondence to the triad of directors presiding over the Tannaitic Sanhedrin, c:n n"a N'ir:, and the 'c^oi i"2N p in the Palestinian Gaonate. But as we have no definite information about the office of the can (see the present writer's article upon the subject, "Jewish Encyclopedia," s. v. Hakam), this correspondence gives us no clue to that of the N?3 'i. As will be shown below, pp. 47-50, the title D*S was conferred upon the heads of the Pumbedita Academy, in the time before they were called Geonim. Besides these, we know the 3*1 Rabbi Samuel, the great- grandfather of Rabbi Sherira, and Rabbi Amram, the maternal uncle of Rabbi Sherira. The nbs 'YCTD mentioned in Harkavy, 201, the THE G AON ATE 13 tive each of a family. Another common point characterising the three institutions is a fiscal system. The Gaon received moneys like the Exilarch, and like the Patriarch in earlier times. In the Judaism of ancient days, and for hundreds of years after the extinction of the Gaonate, no fees were attached to/the office of a teacher, especially a teacher of advanced disciples, and still more especially if the teacher's office was connected with/ the exercise of judicial authority 1 . Now, we know from Nathan ha-Babli (82, 5 from below), that the Gaon received a fixed salary for his personal use, and also Rab Amram, in the Introduction to his Seder, tells us that one-half, or, according to another reading, one-fourth, grandfather of Rabbi Sherira (end of his Letter ; not the grandfather of Rabbi Hai, as Harkavy, 409, calls him), was not a xSa 'i, but secretary to the Academy, as we are informed explicitly in a Genizah fragment (J. Q.R., XVIII, 402). The same office was filled by the great-grandfather of Rabbi Sherira, Rabbi Judah, before he was appointed Gaon, the Genizah fragment just cited being authority for this statement, too. Again, the grandfather of this Rabbi Judah occupied the same position of secretary to the Academy, as we are told by Rabbi Sherira in his Letter (comp. below, p. 71). What the position of Rabbi Nathan was, the paternal uncle of Rabbi Sherira, it is hard to say. The latter calls him F]i'?N, which may stand for N^ '"\ (comp. G. S., p. 237), but as his father, Rabbi Judah, was secretary to the Academy, it is probable that the son may have occupied the same office. In a Genizah fragment (Saadyana, 60) a ni'C'rt ax jra n is mentioned, whom Professor Schechter is disposed to identify with Rabbi Sherira's uncle (great-uncle is probably a printer's error). But this identification is opposed to the fact that Rabbi Sherira calls him rpx, and not -\"m. Perhaps this Rabbi Nathan is identical with the Egyptian scholar Rabbi Nathan, Saadyana, 113. The MD' '~\ mentioned in a Responsum by Rabbi Hai, in Harkavy, 137, may be a Nta '-\ or an VaN. He is probably identical with ncx 'i, the father of the two Geonim, Rabbi Zadok and Rabbi Kimoi, who is the author of a Responsum transmitted to us in TOXTN, II, 77, as the present writer has proved in the JRevis. Israel., V, u. The reading in Voc should be pa F]CV . . . -nn 'tnn. This is Rabbi Joseph ben Abba, Gaon of Pumbedita in 814. A son of Rabbi Samuel ben Hofni, Israel (?), likewise was secretary to the Academy (J. Q. ., 1. c., 404, where ':vnnn means "our young son," as in Saadyana, 118). Perhaps Israel is to be read instead of Samuel in Neubauer, Chronicles, 198, end. In the fragment in the J. Q. R. just cited, as well as in J. Q. R., XIX, 106, the sons of the Geonim appear " as an estate by themselves." 1 Comp. Maimonides, Commentary on Abot, IV, 5. 14 THE GEONIM of all donations sent to the Academy fell to the share of the Gaon l . Rabbi Nehemiah, in a letter addressed to the communities, begs them to send money for himself and the Academy 2 . Thus we have three witnesses, independent each of the others, testifying to the relatively large revenues of the Geonini. The same Nathan informs us that/Babylonia and the adjacent countries were divideo^into parishes, a part of them under the jurisdiction of the Exilarchate, a second part of them under the Academy of Sura, and a third part under the Academy of Pumbedita. 1 In their respective parishes the Exilarchs and the Geomrn exercised the right of appointing the judges and other communal officers, and in acknowledgment of their sovereign rights a fixed annual revenue was exacted and delivered into the coffers of each 3 . FRICTION BETWEEN THE EXILARCHATE AND THE G AGNATE OF PUMBEDITA. These jhreej points roughly stated, the pre-eminence of the Gaon wltnin the Academy, the quasi-hereditary character of his office, and the equipment of the Academy with power to levy taxes and appoint communal officers prove abun- dantly that the Gaonate was by no means a purely scholarly 1 Comp. Marx, Untersuchungen zum Seder des Gaon Rob Amram, I, n. 8 J. Q E. , XIX, 106 ; on 1 ?! i:b. He speaks of run:, free-will offerings, mp'CE, fixed dues (comp. Rabbi Abraham Ibn Daud, 68, 4, bottom, np'Ds), and D"iroirT, "fifths." What is meant by the last cannot readily be determined. Perhaps the name originated in the fact that the con- gregations had five kinds of taxes to pay, viz. for the Exilarch, each of the two Geonim, and each of the two Academies. Dr. Poznan ski's conjecture (1. c., 401), that a fifth part of the whole income of the members of the congregations was paid to the Academies, is very improbable, if only for the reason that the Rabbinical law does not permit more than a fifth of one's income to be set aside for alms and related purposes. If the members of the congregations had sent one-fifth of their income to the Academies, there would have been nothing left for the home needs. Comp. also Saadyana, 118, where irpbrra probably means "the portion due us." 8 Concerning landed estates and the revenues of the Academies, see J. Q. R., XIV, 389, an 1 XVIII, 402. THE GAONATE 15 institution. What has been adduced enables us also to reach a better understanding of the continual friction between the Exilarchate and the Gaonate, and the not infrequent conflicts that arose among the pretenders to the Geonic office. Scholarly zeal, family pride, and material interests are factors of too great potency in the life of individuals not to leave their impress upon the course of history. In the Talmudic time, while the Exilarchate was supreme, without a rival, dissensions might happen to occur now and again between the temporal power and a scholar here and there, but with the Academies as such the Exilarchs had nothing to do. The whole aspect of affairs changed in the period of the Geonim, when the influence of the body of scholars found concrete expression in the Yeshibot, the vested privileges of which constituted them dangerous rivals of the Exilarchs. The only historian of the Geonic time, Rabbi Sherira (36, 13), has this to say regarding the older epoch of his period : " The succession of the Geonim at Sura, up to the year one thousand (689), is not quite clear to us, by reason of the disorders and revolutions caused by the Exilarchs; who depose Geonim and install them again 1 .'' This statement of Rabbi Sherira's, regarding the relation between the Exilarchs and the Geonim of Sura, is rather startling, for, leaving out of account the quarrel between Rabbi Saadia and the Exilarch David, which sprang from personal opposition rather than a conflict of powers, Rabbi Sherira himself makes no mention of any sort of discord between the Geonim of Sura and the Exilarchate for the three centuries following the date given by him. The appointment of Rabbi Samuel and Rabbi Jehudai, scholars of Pumbedita, to office at the Sura Academy (Letter of Sherira, 36, end, 37, 5), is surely not to be taken as an act of hostility on the part of the Exilarch Solomon ben Hisdai against the Academy at Sura. It appears, on the con- = Nnaicnn, "revolutions" ; this passage is badly corrupted in some versions of the text, and many an error has been caused by the confused reading. l6 THE GEONIM trary, that the Exilarch was desirous of securing the most prominent scholars of the day for the Sura Gaonate, as Sherira himself observes. The vacancy at Sura in 843-4, caused by dissensions (Letter, 39, 10), cannot be set to the account of the Exilarch ; Kabbi Sherira would not have kept us in the dark had it been so. It must have been due to some internal disturbance in the Academy, which, it seems, was divided into two factions, partisans of the family of Rabbi Zadok and partisans of the family of Rabbi Jacob. The end was that Rabbi Moses, the son of Rabbi Jacob, gained the upper hand, while the son of Rabbi Zadok, a younger man than Rabbi Moses, assumed the Gaonate fifty years later. On the other hand, Rabbi Sherira records a number of conflicts between the Exilarchs and the Geonim of Pumbedita. About Rabbi Natronai I (719), Sherira says (35, 6, below), that, encouraged by his kinship with the family of the Exilarch l , he ruled the Academy so vigorously that the scholars of Pumbedita took refuge in Sura, and did not return to Pumbedita until after his death. A generation later (about 755) we hear again that the Exilarch, actuated by personal animosity 2 , passed by the claims of Rabbi Aha, later famous on account of his work Sheeltot, and instead installed his secretary 3 , Rabbi Natroi Kahana, as Gaon of Pumbedita. A serious conflict broke out in 771 between the Exilarch and the Gaon of Pumbedita, Rabbi Malka. Rabbi Sherira (36, 4) writes: 'wan na wnoab rrw uate an torn noa *an 'Dp mm N'M winx an no na 'at by wna^aa KB>J an -IDBKI 'mnajn K'tw 'ar oy Knsa'riD pmn jaaaw pas? >TK K't?J "WnBJI py pi'. In view of the historical 1 The exact relationship is not given by Rabbi Sherira. He probably was a son-in-law of the Exilarch. 2 Ibn Daud, 63, 14: rvito irso roao ':BQ. Rabbi Sherira must have meant the same, though he does not express it in so many words. 3 mrac , comp. 'Erubin, 1 1 b, and Yebamot, 42 a, where Amoraim are called NSQIE, which naturally cannot mean house servants, &c. THE GAONATE 17 importance of this passage it is the only instance trans- mitted to posterity of the Geonim interfering in a contest about the Exilarchate it is worth while discussing it thoroughly, all the more as it has been completely misunderstood heretofore. Graetz renders Rabbi Sherira's account in the following words (Geschichte, V 3 , p. 386) : "[Rabbi Malka] had deposed Natronai ben Habibai, when he [Natronai] was about to usurp the dignity from Zakkai ben Ahunai, who had been in possession of the office of Exilarch for some years past. The two Academies united in supporting Zakkai ; they deposed Natronai, and he had to flee to Maghreb." Weiss, in his Dor Dor we-Doreshaw, IV, 29, goes a step farther. He gives the following description of the incident: "In the time of Rabbi Malka a dispute occurred between him and the Exilarch Natronai ben Zabinai 1 , by reason of the fact that the Gaon had determined to make Zakkai ben Ahunai Exilarch. In this purpose he was aided and abetted by the Gaon of Sura. With united forces they worked to remove Natronai from his office, and put Zakkai ben Ahunai in his place, and they succeeded. Natronai was forced out, and, grieved by the dishonour done him, he left Babylonia, and settled in Palestine 2 . The cause of 1 Weiss accepts the incorrect reading wit, while Graetz properly has wan. Albargeloni, nvyn 'D, 256, writes the name wan, as Rabbi Isaac of Vienna does in i"i, I, 114 d, though the source followed by the last, DTIC , 28 a, reads wan . a Graetz again displays his insight here, when he translates nyo with Maghreb, that is, Spain and North Africa, for Albargeloni, I.e., and the correspondents of Rabbi Hai (D':pi cnc, 56, where ncirr is a printer's error for TIBC, the Parma MS. and Albargeloni, rrvs' 'D 'r, 108, having the correct word TIED) have the tradition that Rabbi Natronai went to the Maghreb. My colleague Dr. Friedlaender tells me that the Arabic writer Ibn Hazm, a contemporary and acquaintance of Rabbi Samuel ha-Nngid, makes sport in his Milal wa'n-Nihal, I, 156, and V, 4, of the Jews who say that one of their sages went from Bagdad to Cordova in a day, and horned an enemy of their people there. There can be no doubt that this sage was Rabbi Natronai, of whom Albnrgeloni and Rabbi Hai alike report that he went to Spain by means of -pin rcjcp. It is true I C 1 8 THE GEONIM the conflict was, as we can see from the Letter of Rabbi Sherira, that Natronai was a scholar, and the Geonim did not care to have a learned Exilarch in office." In the first place. Rabbi Sherira makes the explicit statement that Zakkai had been Exilarch many years before Natronai. Then, even if it were true that the Geonim opposed Natronai, which I hope to show was not the case, they were not conspiring against the Exilarch in office. On the contrary, they were giving him their support in his struggle with an usurper of his dignity. Graetz, who speaks in the body of his book (p. 1 74) somewhat vaguely of the conflict between Natronai and Zakkai as a " quarrel about the Exilarchate between two pretenders," is more precise in his note on the passage, in which he properly denominates Natronai a usurper. Halevy, in a long tirade against "the German scholars" (231-2), accuses Graetz of having perverted facts only to cast a slur upon the Geonim, yet he himself agrees with Graetz in his statement of the affair between the Gaon and the Exilarch. The truth is that Graetz, and Halevy as well, misunderstood the case that Rabbi Hai does not give credence to the story told him about Rabbi Natronai, but his incredulity extends only to the miraculous manner of his removal from place to place, not to the fact of his emigration to Spain. Albargeloni furthermore relates that Rabbi Natronai wrote the Talmud down, from memory, for the use of the Spanish Jews. The statement of the great-grandson of Rabbi Paltoi, J.Q.R., XVIII, 401, that Rabbi Paltoi sent the Spanish congregations copies of the Talmud and Talmudic explanations, in no wise contradicts Albargeloni. Even if it is true that Rabbi Natronai wrote the whole Talmud down for the Spaniards, it would not be at all remarkable to find that copies of the Talmud were rare in Spain a century later. One hundred and fifty years after Rabbi Paltoi, Rabbi Samuel ha-Nagid (Ibn Daud, 72, 2, bottom) had copies of the Talmud made and distributed. On the subject of the circulation of copies of the Talmud in the time of Rabbi Paltoi, see G. S., p. 295. The Responsum discussed there (p. 294) was probably given by Rabbi Natronai, the contemporary of Rabbi Paltoi. Briill, Jahrbucher, IX, 117, attributes the opposition of the Geonim to Rabbi Natronai to the fact of his putting the Talmud into writing. They insisted upon oral transmission. But how could they have divined what he would do after leaving Babylonia ? THE GAONATE 19 completely. With historic insight Graetz (1. c.) recognised the difficulty in Sherira's words : p^ tota 21 10BK1 "rmayi myc6 f>TX x rvnriN does not mean "to depose," but, on the contrary, to install one in office in opposition to another. Sherira himself corroborates this linguistic usage on the next page (38, u): no ^y TTinnNi . , . prra" 1 m no ^o mrni *1DV *m "and after him Rabbi Isaac officiated as Gaon, whom they [the Academy and the Exilarch] installed in opposition to Rabbi Joseph." Sherira goes on to explain that Rabbi Joseph, by reason of his position, learning, and descent, had a claim upon the Gaonate, but that the Exilarch had ordained Rabbi Isaac as Gaon " over him." In the light of these facts the passage regarding Rabbi Malka in Sherira's Letter reads as follows: "And he [Rabbi Malka] installed Natronai ben Habibai as Exilarch in opposition to the Exilarch Zakkai ben Ahunai, who had been vested with the office for some years. The two Academies, on the other hand, assembled in joint session, Zakkai also being present, deposed him. Accordingly, when c 2 20 THE GEONIM Rabbi Malka departed this life, the Exilarch Natronai emigrated to the West." This case anticipates the later one of Rabbi Saadia, when he made Hassan Exilarch in opposition to David, who had been holding the office for many a long year. And as, at the time of Saadia, the two Academies, yielding to the pressure brought to bear by the Exilarch David, divested Saadia and Hassan of their dignities, so also it happened at the time of Rabbi Malka, for Til"Qyi, as the correct texts read, refers to Rabbi Malka: "They [the Academies] together with the Exilarch deposed him [Rabbi Malka]." Later copyists, who went astray in the same way as the modern historians, added wnB3;> after Til-OVl l . Naturally, it cannot be supposed that Rabbi Malka acted single-handed in his opposition to the reigning Exilarch and the Academies. He must surely have had his followers, like Rabbi Saadia during his suspension from office, and it is not at all unlikely that he would have come out victor in the end, as Rabbi Saadia suc- ceeded in his struggle, had he not fallen during the fray. And his death was the reason that made Rabbi Natronai go to the West. He had to give up the contest after his main support, Rabbi Malka, had passed away. The accusation against the Geonim, that they incited quarrels with the Exilarchate when the incumbent was a scholar, is wholly unfounded. If history were written according to such methods, the inquirer would reach the opposite result, that the partisanship of the Geonim for one close to them in intellectual interests led them to 1 But even if N:vno:b ^mis 1 ! were proved to be the correct reading, the other assertion, that Rabbi Malka was not the opponent, but rather the friend, of Rabbi Natronai, remains unassailed. It is, however, inconceivable that Rabbi Sherira should have used the expression 'rmori of an usurper, seeing that with him, as for instance 36, 9, it has the meaning of removing one from an office legitimately held. And it would be an absurdity to say that " the Exilarch removed the counter- Exilarch from office," as though a pretender would acknowledge the legitimacy of his opponent. THE GAONATE 21 prefer a learned to an unlearned Exilarch. Now we know that the quarrel about the Exilarchate at the time of Zakkai ben Ahunai grew out of far other motives. From the Genizah fragment given in Saadyana, 76, it appears that Zakkai was a descendant of Bostanai and a Persian princess, a marriage the legitimacy of which was questioned by many. For this reason, Rabbi Malka was prepared to support Natronai, whose descent was unblemished. From the Genizah fragment we learn also that the descendants of the princess tried to force the recognition of their legitimacy by resort to the power of the non-Jewish government. Accordingly Rabbi Malka was justified in his opposition to Zakkai. Scarcely ten years pass (782), and again we hear of the Exilarch's deposing the Gaon of Pumbedita, Rabbi Haninai ben Abraham. Rabbi Sherira, who usually drops a hint at least as to the cause of such disputes, has not a word to say about this occurrence. It is fair to take this as corroborat- ing the supposition made above (p. 10), that Rabbi Abraham Gaon, the father of this Rabbi Haninai, was a brother of Rabbi Xatronai, and, as he belonged to the Sura Academy, as will appear later, and received the Gaonate of Pumbedita against the wish of the Academicians there, the assumption is not unwarranted that the deposing of Rabbi Haninai was due to the wishes of the Academy, which was not inclined to accept an outsider. As to Rabbi Sherira, he had good reason for not desiring to enter into a detailed discussion of the case ; it hardly redounded to the credit of his own Academy. In the year 828 we hear once more of interference with the affairs of the Academy at Pumbedita on the part of the Exilarch. The two pretenders to the Exilarchate, Daniel and David, each had " his " Gaon at Pumbedita, with the result that even when David maintained the upper hand, Pumbedita was supplied with two Geonim, Rabbi Abraham and Rabbi Joseph. It is not possible to define the part played by the 22 THE GEONIM Exilarchs in the disputes at Pumbedita between the Geo- nim Kabbi Isaac and Rabbi Joseph ben Rabbi in 833, and between Rabbi Menahem and Rabbi Mattathias in 859. About Rabbi Isaac, Sherira says (38, 14) that the Exilarch David ben Judah had installed him, but that does not guarantee Isaac's having been his candidate as opposed to Rabbi Joseph, because the expression used by Sherira is irwinNl, " and they appointed him [Rabbi Isaac] as Gaon." " They " probably means the members of the Academy *. Finally, a feud of many years' duration broke out between the Academy of Pumbedita and the Exilarchs, under the last of them, David, who appointed Rabbi Kohen-Zedek to be the Gaon, while the Academy invested its own candidate, Rabbi Mebasser, with the dignity. THE LANGUAGE OF NATHAN HA-BABLI'S REPORT. To the student who regards history as more than a mere stringing together of disconnected events, the friction between the Exilarchs and the Geonim of Pumbedita presents an interesting problem in various respects. Many a question evoked by the combative relation between Gaonate and Exilarchate clamours for a reply. In the first place, why was it that the Academy at Sura was not troubled by the interference of the Exilarchs in the course of a period during which the Academy at Pumbedita felt their heavy hand half a dozen times'? What was the reason that the Exilarch, who lorded it over the Academy at Sura until the end of the seventh century, assumed so peaceable an attitude toward it during the three centuries that followed ? And, in the third place, what explanation can be adduced for the fact that all the wrangles between 1 Halevy, who regards the Exilarchs as universal scapegoats, holds (p. 271), without advancing any proofs, that it was again the Exilarch who appointed Rabbi Isaac as Gaon in opposition to the wish of the Academy. The words maoD 121 prove nothing, because the official ordination was always performed by the Exilarch. THE GAONATE 23 the Exilarchs and the Geonim of Pumbedita occurred in a single century, from 719-828 ? l These questions can be answered only when we have attained to intimate knowledge of the rise of the Gaonate and its relation to the Exilarchate on one side and the two Academies on the other, and knowledge of this sort is accessible to us only through closer acquaintance with the sole and only account of the Academies that has come down to us. Rabbi Samuel Shulam, in his additions to Rabbi Abraham Zacuto's Yohasin, gives an account of the Babylonian Academies and of the Exilarchs Ukba and David, after one Rabbi Nathan the Babylonian. An Arabic fragment of the report concerning Ukba was published by Dr. Israel Friedlaender in the J. Q.K, XVII, 747-61. The great historical value of this document makes the language in which it was written originally a matter of prime importance, and it behoves us to give our attention to this question first of all. Dr. Fried- laender, in his learned and instructive introduction to the narrative, is decidedly of opinion that it was written in Arabic originally, but I venture to believe that the proofs adduced by him are not conclusive. The expression D^D^a jnu . . , N DIN is admittedly an Arabism, but it had become so fluent a locution with the Arabic-speaking Jews that it cropped up in their Hebrew and Aramaic writings as well. Its use by Nathan, there- fore, proves nothing. In Rabbi Sherira's Letter it occurs three times (35, 6, below ; 40, i ; and 40, 5), yet no one is inclined to doubt that the Letter has been transmitted to us in its original language 2 . Dr. Friedlaender further 1 The controversy between Rabbi Mebasser and Rabbi Kohen-Zedek is of quite another character, as will be demonstrated in detail further on. 2 The expression .... a jrro occurs frequently in original Hebrew works ; comp., for instance, ruDErr, I, 61 ; crteiT i:3, III, 15 b ; J.Q.R., XIX , 106, 730, 734. The phrase, derived from the Arabic, was the model for -\3:, "known under the name"; comp. Harkavy, D':\r' CJ O'vnn, II, 10. In the inscription on the Cattaui synagogue in Old Cairo, reproduced 24 THE GEONIM claims the phrase DC^NI by 1^ 13rD'1 (79, 19) as a translation of the Arabic nriDfrO") HPJH. The expression, occurring three times in close succession, has a Hebrew equivalent in each of the three contexts: Btn WilN WWl B*n ?ni uvum "pvy hf inrrum. If the use of D^KI . . . mron proves anything, it would rather indicate that the one who trans- lated the document from Hebrew into Arabic did not understand it, and left the original untranslated. What Nathan says in this passage is that the Gaon of Sura sent word in writing to his followers, either to offer their congratulations personally to David ben Zakkai, on his assumption of office (irwiTtP), or, if there were any l who for some valid reason could not appear before him, to express their gratification at his success in a letter to the Exilarch Dt^QJ by "6 'GrD v i. In one way or another they all were to manifest their assent to his choice as Exilarch K>xn iniN uvwi. In the description that follows, of the public presentation of the Exilarch, Nathan properly omits all reference to the written homage ordered by the Gaon. Nathan is equally precise in his account of the homage paid the Exilarch by Kohen-Zedek. The two dignitaries met face to face, hence the expression used by Nathan, "pvy by inrrurn, where DVJ? is a synonym for the K>SJ used before. For the rest, the phrase employed by Nathan to express the public recognition of the Exilarch as such, nta twi JTUit, throws new light upon an expression occurring in the Talmud several times ami Nnpiy w:ni> KIDH n mms which has caused the lexicographers no little difficulty 2 . The Aramaic "DIN by E. N. Adler, "Jews in Many Lands," 30, mrr does not mean "the famous," but "named." Comp. also Harkavy, Saadia, 114, .... p rim, and 227, note 6, and Steinschneider, Jubelschrift, 139, line 8 from bottom, and Harkavy, 186, where ID: = 3 rim. 1 On the Tannaim mentioned in this passage, comp. Marx, J.Q.R., XVIII, 771, to which should be added that Rabbi Hai in the Responsum appearing as an appendix to Rabbi Sherira's Letter, ed. Mayence, speaks of ... o'wnn (65) ; comp. also '^nv 'c, 130, ed. Neubauer. 2 On the locution irrnnN c'Mn, in the Seder 'Olam Zutta, see Lazarus, Die Hdupter der Vertriebenen, 100-1, and Briill, Centralanzeiger, 67. THE GAONATE 2 5 corresponds exactly to the Hebrew rnan of Nathan. Accordingly, the translation would run: "Rabbi Hisda proclaimed Rabban Ukba as Exilarch, on which occasion the new Exilarch spoke as follows." The Arabic 1"ipy^ nnDxn would be rather colourless, while the Hebrew i'run is the very term one would expect to find here. The expression by Toy is not an Arabism ; it is found in the Talmudim and the older Midrashim with con- siderable frequency. I shall adduce only a few of the passages. D'con i^y iioj?^ i^r N^>I py spm rnrrwn pa, " Twilight lasts but an instant, so that the scholars could not determine its duration" (Ter. Berakot, i, 2 b, 35, and parallel passages ; Babli, ibid., 2 b, end), by Tioy^ trp'Qt? TinP ?W p'JO, "He wanted to determine the number of Israelites," which corresponds exactly to the expression used by Nathan (Yer. Taaniyot, II, 56 d, 44). The Tal- mudic equivalent for errando discimur is by tciy mx px era braa p DK N^N mm nm, "Man cannot fathom the words of the Torah until he has made mistakes " (Gittin, ^3a). Regarding the motion of the celestial spheres, Rabbi Simon ben Yohai says: nmni? IK'SX n ny^inoi jcnsn fl (83, 16). So far as I know, this use of fo occurs only in works translated into Hebrew, not in Hebrew originals, and it gives considerable weight to Dr. Friedlaender's opinion as to the original character of the Arabic text. In any event, the Arabic contains some 1 Nissi, the son of the Exilarch and brother-in-law of the Gaon Sar Shalom, is mentioned by Kabbi Hai in his Kesponsum appended to the Letter of Kabbi Sherira, ed. Mayence, 63. bNitra m mi D':, in DTIE, 380, is derived from the Seder Rab Amram, as can be seen from Marx, Uiitersuchungen, &c., 8, Hebrew part, but 'n'l 'no (32) reads 'ii instead of 'D':. I have only to add that the Genizah fragments of the YerushaJ.mi read 'p: in all passages in which our texts have wr: or nD3. THE G AON ATE 29 readings that are preferable to the Hebrew in corresponding passages, and they are of great value in the study of Nathan's account. NATHAN HA-BABLI IDENTIFIED. Another important question must be settled, and a more difficult one. Who was this Nathan, the Babylonian, the author of the report we are considering ? Graetz's hypo- thesis (Geschichte, V 3 , 47 12), that he was one of " the four captives," and the founder of Jewish learning in Provence is, it need hardly be said, wholly untenable. From the Genizah fragments, we know first of all that Rabbi Shemariah ben Elhanan, one of the four captives, was a pupil of Rabbi Sherira (J.Q.jR., VI, 222). But Nathan, as Graetz himself observes, wrote his account during the Gaonate of Rabbi Aaron, and knows nothing of Sherira. Moreover, Rabbi Hushiel's Letter, published by Professor Schechter (/. Q. R., XI, 643-50), stamps the whole story of the four captives as a legend, at least in the form in which it has been transmitted to us by Rabbi Abraham Ibn Daud. There may be an historical kernel in it, but not more. Furthermore, the hypothesis advanced by Graetz rests on a false construction put upon a sentence in Zacuto's Tohasin (ed. London, 174), where a sentence is quoted from a "Rabbi Nathan, the Babylonian, in Nar- bonne." The practice of applying the name Babylon to Rome is not limited to the New Testament (Rev. xiv. 8 ; xvi. 1 9 ; xvii. 5). It is current in the Midrash as well (Cant. -R., I, 6), and there can be no doubt that Zacuto is referring in the passage under consideration to Rabbi Nathan of Rome, the author of the 'Aruk, who studied in Narbonne under Rabbi Moses ha-Darshan. To clinch the identification, the very sentence cited by Zacuto in the name of Rabbi Nathan, the Babylonian, is to be found in the 'Aruk of the Roman Rabbi Nathan l . 1 On the sojourn of Rabbi Nathan, the author of the "jnr, in Narbonne, comp. Gross, Gallia Judaica, 409-10, and Geiger, Heb. BibL, III, 4. The 30 THE GEONIM On one point Graetz is doubtless right in assuming that Rabbi Nathan wrote his account, not in Babylonia, but in some other country, the Jewish inhabitants of which he wanted to enlighten concerning Babylonian conditions. In all probability the country in which the Babylonian wrote was North Africa. His account, as it appears in the Yohasin, and also in the Arabic Genizah fragment, begins with the words : " This is what the Babylonian Nathan, son of Isaac, told [=IDN, 'reported by word of mouth'], what he himself partly saw and what he partly heard in Babylonia, relative to the Exilarch who came to Africa, Ukba, the descendant of David." Now, only a small part of Nathan's account deals with Ukba, and it is difficult to understand why, in the first place, Ukba should be named as the hero of the narrative, but par- ticularly why it should have been stated so emphatically that he had come to Africa, a circumstance which naturally comes out in the course of the narrative. It is therefore not a far-fetched supposition, that this Babylonian Nathan himself came to Africa, and the Jews there questioned him about the celebrated exile who had once lived in their city, for at Nathan's arrival he was probably deceased. About the controversy of the Exilarchs, Nathan could tell them but a few facts known to him by hearsay, 1"13D1 nvpED. It had happened before his time, or at least in his earliest childhood. On the other hand, he was well versed in the details of the dispute between the Gaon Kohen-Zedek and the Exilarch David, and again between Rabbi Saadia and the same Exilarch. Therefore he passed adroitly from Ukba to his successor. The description of Ukba's exile serves as nothing more than a foil and intro- duction to the events under David. That he began his account with Ukba shows equal astuteness, for Ukba it was who interested the African Jews in particular. name Nathan ha Babli was probably suggested to Zacuto by the celebrated Tanna of the same nnme; '\ '?N f c, ed. Friedmann is taa = 'ori, R. Joshua was in Rome, comp. Gittin, 58 a. THE GAONATE 3! These conjectures, which to me seem obvious, are sup- ported by a Responsum by Rabbi Mei'r of Rothenburg 1 . itray *iy i>"r Kpnsso fro Yn mi^ra mro 'bv owtan nuuwn nmx PS^TEI 2^n Bnaoi nono nwirb i^nnn&'D ^>as -urn m i:n: nniN ^Kt? V3 ^3 p-UD 13N "In my collection of Responsa of the Geonim, I found the following by Rabbi Nathan of Africa : Until now it was customary to permit the eating [of butter made by non-Jews], but since they have begun to bring it from Hamath and Giscala, where it is adulter- ated [with fat], we excommunicate all who use it." First of all, we are here introduced to an African scholar of the Geonic time by the name of Nathan. One is tempted to identify him with the Rabbi Nathan ben Rabbi Hana- niah, a Responsum by whom is abstracted (T"lN, I, 176 b) by Rabbi Isaac ben Moses of Vienna, the teacher of Rabbi Mei'r of Rothenburg, from the "African" collection niyxpon 'D, pro- bably the same Geonic collection referred to by Rabbi Mei'r himself in his mnien ny>, 193. Muller in his Mafteak (157) assigns this Responsum to Rabbi Nathan Alluf, the uncle of Rabbi Sherira, an identification that cannot hold water, for several reasons. With the exception of Rabbi Hai, who replied to a number of questions addressed to his father, by reason of the advanced age of the latter, there is not, in the whole extent of Geonic Responsa literature, a single Responsum by an Alluf 2 . Besides, Rabbi Isaac of Vienna calls the author Rabbi Nathan ben Hananiah, and the uncle of Sherira was Rabbi Nathan ben Judah. Miiller's emended reading, rp::n YIK, instead of 'n '"H nna, cannot be endorsed. What reason can there be for designating the 1 Quoted by Rabbi Aaron of Lunel in his n"n 'rn, II, 333. Rabbi Nathan, whose views on liturgical questions are cited very frequently by Rabbi Aaron in the first part of his work, was, as appears from 'n 'rnr, I, 43 b and io6a (bottom), a grandson of Rabbi Azriel, doubtless Rabbi Azriel ben Nathan, the great-grandson bearing the name of the great-grandfather. Gross, Gallia Judaica, contains Rabbi Azriel, but not his grandson, Rabbi Nathan. * For details comp. above, p. 7, n. i. 32 THE GEONIM son and brother of a Gaon as the brother of his brother, instead of in the universal way as the son of his father ? We have, besides, positive and explicit evidence regarding an African authority by the name of Rabbi Nathan ben Hananiah. Such an one was a correspondent of Rabbi Natronai Gaon, as we learn from Rabbi Samuel Ibn (jama 1 , and also of the Gaon's younger contemporary, Rabbi Zemah ben Solomon, the chief judge of the Exil- arch 2 . In a question addressed from Kairwan (fur, 84 a, 3) to Rabbi Zemah [ben Paltoi?], Rabbi Nathan and Rabbi Judah are characterised as "the scholars of Kairwan V In another Responsum in the same col- lection, i8b, 12, the sons of Rabbi Nathan are referred to in a letter to Rabbi Saadia. Moreover, it is highly probable that the Rabbi Nathan whose opinions are cited in three passages in the Seder Rob Amram is this African Rabbi Nathan, and not the uncle of Rabbi Sherira 4 . Nevertheless, I hesitate greatly to identify the Rabbi Nathan quoted by Rabbi Mei'r of Rothenburg with the Kairwan scholar Rabbi Nathan ben Hananiah, and for the following reasons : The passage about the butter made in Hamath and Giscala by no manner of means bears the interpretation that butter was exported from Palestine to Northern Africa in the ninth century. The remark by Rabbi Nathan becomes intelligible only when it is 1 In Graetz, Jubelschrift, 17. 2 Dukes, from an Oxford MS., in Ben Chananjah, IV, 142. 3 This passage was referred to by Zunz, Situs, 191, and he properly identified this Rabbi Judah with Rabbi Judah ben Saul, the contemporary of Rabbi Nathan. The same Rabbi Judah is described in t'w, II, 171 b. together with Rabbi Nathan ben Hananiah, as a correspondent of Rabbi Natronai. He is there called bistt? '-\ '2 rmrr 'n , which is better, it seems, than "JINC 'T 'i rmrv '-\ 'a mirr 'i, in Luzzatto's is^n rva, 109. In Rabbi Mei'r of Rothenburg's n"r, 193, he is also called Rabbi Judah ben Saul. Is ci"n to be read for the corrupt CTD in Parties, 21 b ? 4 Comp. below, pp. 149-50. In this Responsum c'Tobn does not mean young students, but, according to the general usage of Arabic-speaking Jews, prominent scholars. Comp. Harkavy, Saadia, 43, note 5, and y"ic, 3 a, end. THE G AGNATE 33 brought into connexion with the fact that in Babylonia butter made by non-Jews was considered as belonging to the forbidden varieties of food, though it was permitted in Palestine. Hence Rabbi Nathan reports that even in Palestine the use of such butter was prohibited, since it appeared that it was adulterated in Hamath and Giscala, being mixed there with forbidden ingredients. Whence this specific acquaintance with Palestinian conditions on the part of Rabbi Nathan of Kairwan ? If we were to assume, what is not very likely 1 , that the Kairwan scholars of the ninth century were in close relations with those of Palestine, it would still have to be explained what occasion there was for the Palestinian scholars to communicate with the Kairwan scholars regarding the custom prevailing in their country. Thus the probabilities multiply for identifying Rabbi Nathan of Africa with the Babylonian Rabbi Nathan, the author of the account of the Academies. This Babylonian, who must have reached Africa by way of Palestine, had to satisfy the curiosity of his African fellow-Jews and a real desire for knowledge as well. The scholar from foreign parts on the one hand told them about the Exilarchs and the Geonim, and on the other doubtful ritual cases were referred to him, such as that in the Responsum quoted above, in which Rabbi Nathan, inclined as a Babylonian to agree with a prohibition forbidding the use of butter prepared by non-Jews, strengthens his natural inclination by reference to the fact that even the Palestinians, ac- customed from of old to a more lenient practice, refrained from eating it in changed circumstances 2 . 1 Rabbi Mei'r of Rothenburg in his n*UJ, 193, writes: 'iairra '3111x1 bin ':wan ibsTTC .... s-pnDM ronoo, which would indicate that this African Responsa Collection contained decisions only by Babylonian, not by Palestinian authorities. 3 On the use of such butter, comp. the Geonic Responsa in D*n, ip-ar, and G. S., p. 153, according to which the prohibition against it had not always been recognised even in Babylonia. Comp. also Miiller, r^fa rra, 16. I D 34 THE GEONIM The assumption that Rabbi Nathan was an oral reporter on Babylonian conditions, rather than an author who re- corded his reminiscences in writing, would reconcile the differences between the Hebrew and the Arabic version of his narrative. The question as to the original language would then be set aside in favour of the supposition that the two versions are independent of each other. In the Kairwan audience that listened to Rabbi Nathan, some used Hebrew and some Arabic in their literary com- positions, and thus his narrative reached us through the medium of two languages. NATHAN HA-BABLI THE SOURCE FOR THE Two REPORTS ABOUT THE BABYLONIAN ACADEMIES. The above will throw light for us upon the relation that exists between Rabbi Nathan's narrative proper and the piece about the Babylonian Academies preceding it. Graetz, whose view is espoused by Weiss and other scholars, considers Rabbi Nathan the author of the de- scription of the Babylonian Academies at the head of the narrative, in the same sense in which he is the author of the narrative to which his name is explicitly attached. Halevy, on the other hand, identifies the piece about the Academies with a report quoted by Zacuto from Rabbi Samuel ha-Nagid's Introduction to the Talmud. Graetz's historical tact stood him in good stead here as so often, while Halevy cannot see the wood for the trees. There can be no doubt, as Halevy properly remarks, that the two are merely versions of one and the same account ; and also there can be no doubt that Samuel ha-Nagid's document goes back to Rabbi Nathan. It is certain that the description of the Babylonian Academies pre- ceding Rabbi Nathan's account cannot have been taken as it stands from Rabbi Samuel's Introduction, which, Halevy maintains, seeing that it contains two important points missing in Rabbi Samuel's the description of the THE GAONATE 35 "reception Sabbath" of the Exilarch,and the dispute between the Academies regarding the division of the revenues, at the time of Kohen-Zedek 1 . Halevy (Dorot ha-Rishonim, III, 363) passes the first point over in silence, and with regard to the second he maintains that it dropped out of Rabbi Samuel ha-Nagid's narrative in Yohasin through an oversight of the copyist. But whence could Rabbi Samuel Shulam, the editor of Zacuto's Yohasin, have supplied the passage which was missing in his model 1 2 We see thus that not only is the account transmitted by Shulam independent of Rabbi Samuel ha-Nagid's, but a comparison of the linguistic peculiarities of the description of the Academies with those of the narrative proper by Rabbi Nathan proves beyond the peradventure of a doubt that they have the same origin. For instance, in both accounts pN is used in the meaning of city (78, 5 ; 79, 31). The statement about the rights of the Geonim of Sura during an interregnum in the Exilarchate is the same verbatim in Nathan's narrative proper (86, n, below), and in the description of the Academies preceding it (78, 15), 1 The following point forms an essential difference between the two narratives. According to Rabbi Samuel ha-Nagid it was a question of "parishes," nvvm, those under Sura being twice as large originally as those under Pumbedita. But according to the account published by Shulam, it was a question of the donations, which were put into a common fund for the Academies, two-thirds being allotted originally to Sura and one-third to Pumbedita. The rather indefinite expression in Shulam's report, D'pSn w nVDi: , was misunderstood by Rabbi Samuel ha-Nagid, who took the nvren of the previous sentence as the subject. This view is proved incorrect by the words of Rabbi Nathan, WTO no to . 2 Halevy might have learnt from Coronel's introduction to the meort D'cmcnip that the MS. of this report used by Neubauer for his edition had been written in 1509, while Shulam published the Yohasin at Constantinople only in 1566. On the MSS. of this report comp. Marx, in Z.H.B., V, 57-8, and IX, 140. Steinschneider, in Geschichisliteratur, 21, likewise entertains the supposition that Shulam's report goes back to Rabbi Samuel ha-Nagid. It need not be said that the great historian was too circumspect to assume, as Halevy does, that Shulam had simply copied Rabbi Samuel's narrative from Zacuto. He is of the opinion that the source made use of by Shulam is traceable to Rabbi Samuel's Intro- duction, which, however, as has been shown, is equally unwarranted. D 2 36 THE GEONIM while Rabbi Samuel ha-Nagid has the somewhat pompous expression iD^iy rvu^ nta >xn IDB^DI for nta ^N"i nic% and in the same sentence he uses niKSjnn niB>n for the lh? nwnn 13ni* of the other two sources. But as, on the other hand, Rabbi Samuel ha-Nagid's presentation in the main agrees literally with the descrip- tion of the Academies preceding Rabbi Nathan's report, we are safe in assuming that Rabbi Nathan is the source for both. The development must have been thus : Rabbi Samuel, in his Introduction to the Talmud, where he had to speak of the two Academies, abstracted Rabbi Nathan's account, which may have come under his notice through the Jews of Kairwan, with whom, it is well known, he was in constant communication 1 . Another author, who had heard Nathan's account from his own mouth, tried to make up a brief sketch of the Academies. He gave a few facts regarding their origin at the time of the Amoraim, and then, to lend his compilation an air of completeness, he eked out Nathan's report by the addition, at the beginning, of a chronology from Adam to David, the last of the Exilarchs, taken from the Seder 'Olam Zutta. According to the notions prevailing- in the Middle Ages as to literary practices, this compiler, who patched together three pieces from three different sources, deserved the name author, and, without burdening his conscience, he could maintain silence regarding the sources used by him. This "opus" he made the introduction to the narrative-which he had taken down from the mouth of Nathan, honestly introducing it with the words " and what Nathan said 2 ." 1 Even his questions addressed to the Babylonian Geonim were trans- mitted by the Kairwan scholars ; comp. Harkavy, 107. The literal agreement of Rabbi Samuel ha-Nagid's report with Rabbi Nathan's disposes of the theory that the former made use of Ibn Hofni's Intro- duction to the Talmud. 8 In his mm' Eac, 42, Ibn Verga quotes a report on the installation of an Exilarch from D':ncwi D'aiw rvniicn, which seems to be independent of Rabbi Nathan's, while the passage about the Exilarch Ukba, in Rabbi Abraham ben Nathan's Manhig, 32 a, probably goes back to Nathan. THE GAONATE 37 THE SUPREMACY OP SURA. We return to our starting-point. The relation of the two Academies to each other, and their relation to the Exilarch, can in a measure be defined now. Rabbi Samuel ha-Nagid, as well as the anonymous author in Shulam, who, as we have seen, is none other than Rabbi Nathan the Babylonian, are explicit upon the subject. Originally, the head of the Academy at Pumbedita could be appointed only with the concurrence of the Gaon of Sura. If the heads of the two Academies met anywhere, the Gaon of Sura was given the precedence. This was particularly marked when they paid their respects to the Exilarch on his "reception Sabbath." In their correspondence, the head of Pumbedita had to address " the Gaon and the scholars of Sura," while the head of Sura wrote simply " to the scholars of Pumbedita." In case the Exilarchate had no incumbent temporarily, its revenues fell to the share of the Gaon of Sura. Sura received two parts of the donations contributed for the maintenance of the Babylonian Academies, and Pumbedita but one part. This fiscal arrangement was changed in 926, under the Gaonate of Kohen-Zedek 1 , when Pumbedita was made equal sharer with Sura, on account of the increase in the number of disciples in the former Academy. On the basis of these facts, Graetz properly makes the assertion that/6riginally the title Gaon was the prerogative of the head of the Academy at Sura, the Gaonate not being a duumvirate, but an institution with a single chief, and its origin must be explained with these facts in mind. In opposition to this sane view Halevy (p. 151 et seq.) puts up a theory, which sets forth that in the Geonic 1 There is not the remotest warrant for supposing that Kohen-Zcdk, the Gaon of Pumbedita, was here confused with his namesake of Sura. The important change in favour of the Academy at Pumbedita could naturally not have been connected with the name of the Gaon of Sura. 38 THE GEONIM time Pumbedlta held the leading place, and the above- mentioned privileges of Sura applied to the time of the Amoraim, probably of Rab Ashi, with but few exceptions not being in force in the Geonic time. But how, in the name of common sense, can it be said that the claim upon the larger share in the donations to the Academies appertains to Talmudic times ? We know from Talmudic data (Grittin, 60 b) that the revenues of the Academies consisted of voluntary contributions deposited in boxes, which were put up for this purpose in the house of the head of the Academy. We should be accusing Rab Ashi of highway robbery pure and simple, if we supposed that he ordered the removal of two-thirds of the contents of the box at Pumbedita to the coffers of Sura. It is hardly necessary to defend the great leaders of the Jews against such charges. Halevy, in particular, has no ground under his feet when he relegates the privileges of Sura to Tal- mudic times (p. 263), because he gives the preference to Rabbi Samuel's version, which bases the distribution of the moneys between the two Academies upon the parish divisions for judicial purposes \ and such divisions, it is well known, did not exist in the Talmudic time, as the appointment of communal officers was in the hands of the Exilarch. Besides, as applied to the Talmudic epoch, what does it mean to say that the head of the Sura Academy was addressed as Gaon by his colleague ? Even if Gaon is not taken literally, but as an equivalent for NmTiD B*"I, it is not a term used in the Talmudic period in addressing a scholar, wan and man are the titles applied to scholars in that time 2 . The parts assigned to the heads of the Academies on the "reception Sabbath" of the Exilarch are altogether incongruous with the time of Rab Ashi, about whom we are told explicitly that the Exilarch Huna 1 Comp. above, p. 35, n. i. 3 Ketubot, 69 a, pin ; Shebu'ot, 36 a, im ; comp. also Hullin, 95 b, Dip i:-an mb .... ir.i. THE GAONATE 39 ben Nathan subordinated himself to him (Gittin, 59 a), while in the narratives under examination, the respect shown the Exilarchs by the Geonim is dwelt upon in unmistakable words. However, Halevy adduces reasons for his opinion, that the prerogatives of Sura do not apply to the Geonic time. And astonishing reasons they are ! From the letter of Sherira we know that two scholars of Pumbedita, Rabbi Samuel and Rabbi Jehudai, occupied the Gaonate of Sura 1 . The reverse situation is not mentioned as a fact. But, as Dr. Elbogen justly says, " Lack of knowledge on our part is not a counter-argument " (Die neueste Construction der judischen Geschickte, 33). Sherira, belonging to Pumbedita, was particularly proud of the distinction that fell to the lot of two members of his own Academy, and records it with great satisfaction. On the other hand, he had absolutely no occasion to report the appointment of scholars from Sura at Pumbedita. Quite apart from this consideration, the installation of scholars from Pumbedita at Sura has nothing to do with the question before us. On the contrary, from the fact that the greatest scholars of Pumbedita were invited to Sura, we might justly infer that Sura excelled the other Academy in importance and dignity, and there- fore those of Pumbedita regarded their appointment as a distinction. The right of veto in connexion with the appointment of a new Gaon in Pumbedita, which the sources mention as a privilege of the Sura Gaonate, does not affect the question as to whether, in the course of centuries, two or three scholars hailing from Pumbedita were installed in office at Sura. For the rest, it can be demonstrated from Sherira's Letter itself that scholars of Sura occupied the Gaonate of Pumbedita. An extraordinary circumstance, to which no attention has been paid hitherto, is that Sherira notes 1 Halevy might have added Rabbi Samuel ben Hofni, for he was a grandson of the Pumbeditan Gaon Kohen-Zedek, and assuredly belonged to the Academy of Pumbedita. 40 THE GEONIM the provenance of only three of the Geonim of Pumbedita 1 . They are Rabbi Natronai, of Bagdad, Rabbi Isaiah of WBKOT, a suburb of Bagdad, and the successor of the latter, Rabbi Joseph of vb or vhv. It is, of course, inconceivable that the rest of the Geonim of Pumbedita, as many as three dozen, should all have hailed from Pumbedita itself; or that Rabbi Sherira should be ignorant of their pro- venance. Rabbi Hai, for instance, reports that the Gaon of Pumbedita, Rabbi Hai ben David, had been active, before his accession to office, as judge in Bagdad 2 , and what the son knew the father could surely not have been ignorant of, and yet Rabbi Sherira does not mention the fact that Rabbi Hai ben David's home was in Bagdad. This striking peculiarity can be explained only upon the assumption that Rabbi Sherira adopted the system of mentioning the provenance of the Geonim of Pumbedita only when they were members, not of the Academy of Pumbedita itself, but of Sura an assumption that rises to the degree of certainty when we remember that Bagdad and Sura are close to each other 3 . The addition of the words " of Bagdad " to the name of a Gaon, is tantamount to calling him a member of the Academy of Sura. It turns out, too, that not only Rabbi Natronai, of Bagdad, and Rabbi Isaiah, of nxta 4 , are to be reckoned among the 1 Of course, 1 do not take into consideration the Geonim who were in active life before or about 689. Rabbi Sherira himself was not always prepared to give unexceptionable information regarding this early Geonic time, and therefore he would take good care to add any detail he might happen to know. The characterisation of the Gaon Rabbi Manasseh ben Joseph as ro'py '3 ':n p Nin rwaiai is unintelligible ; probably the passage is corrupt. 2 Ibn Gajat, ir'c, I, 63. 3 The distance between these two places can be determined with a fair degree of accuracy. Al-Kasr, a suburb of Bagdad, the original home of the Exilarch David ben Zakkai, was six miles from Sura, according to other readings seven, and even ten miles, the variations being based upon the resemblance of the letters 1*1"' to one another. ' Comp. Prof. Nflldeke in J. Q. B., XVII, 760, note 3. * Wallerstein's text even has nuab NJifOT N'cnm xnc mn THE GAONATE 4! scholars of Sura, but even Rabbi Joseph, of *&{?, which, as we learn from Talmudic references, is situated close to Sura l . As for the supposition ventured above, that the suc- cessor of Kabbi Natronai was his brother Rabbi Abraham Kahana, the proof can be adduced, that he is the sole and only Gaon of Pumbedita, in the period after 689, whose name is not linked with his father's. The natural ex- planation is that, being the successor to his brother, the father's name appeared in connexion with his predecessor's, and hence there was no need to repeat it. We should, therefore, be justified in putting Rabbi Abraham Kahana among the members of Sura who occupied the Gaonate of Pumbedita. For my part, I should he inclined to classify Rabbi Paltoi in the same way, for the reason that he refers (Miiller, p. 88) to a custom in ^33 hv mn rvo, the venerable old synagogue which Rab had founded in Sura, and the scholars hailing from Sura were the only ones who made reference to this institution 2 . 1 Berliner, Beitrdge zur Geographic und Ethnographic Babyloniens, 33, note I, is of the opinion that ^u? must be looked for in the vicinity of Sura or Pumbedita. But Baba Batra, 1723, shows, as the Tosafists noticed, that it was in the neighbourhood of Rabbi Huna's place of residence, that is, Sura. In other passages, too, it occurs only in connexion with Rabbi Huna's presence in Sura. Comp. Bezah, 25 b ; Baba Mezia, 63 b, does not indicate, as Rashi thinks, that Rabbah and Rabbi Joseph lived close to >! 5. Their dwelling-place was Pumbedita, which may have been a day's journey from 'To. The real meaning of the passage is that great traffic in wheat was carried on there, therefore it was denominated a wheat centre. If Rabbi Sherira, 30, 12, speaks of Rabbi Nahman's having been in nnm T\bc, he means that after the destruction of Nehardea he first repaired to Ttte, and then betook himself to Maho/a in the vicinity of Pumbedita. Keeping in mind the well-known tendency of the Babylonians to eliminate the letters n and n, the spelling '"HC for Tibc need not astonish us ; comp. Funk, Juden in Babylonien, 155, 160. 2 Rapoport, in p'ro -py, 142, has the proper explanation of the ex- pression so frequently used by the Geonim, boaaw irm rva, or briefly i:'m rva, an explanation that I had myself hit upon independently of Rapoport, and communicated to Professor Alexander Marx, who indorses it in his Untersuchungen, &c., n. It was only later, while engaged in the present investigation, that I discovered it in the j'to "py, to 42 THE GEONIM If it is at all proper to constitute the appointment of members of one Academy to the Gaonate of the other as which I here give credit for it. Rapoport points out that the academy and synagogue of Rab were so called in the Talmud, Megillah, 293. Halevy (p. 105) has managed to misunderstand Rapoport's words entirely. He had no idea of asserting that in the Geonic time taiatzj i*a meant the Sura Academy, seeing that it occurs almost always in con- nexion with the rrmu vro. What Rapoport did say is, that in Talmudic times the expression was applied to the academy and the synagogue of Rab, but later only to Rab's synagogue. The change has a good reason. To replace the academy erected by Rab, his disciple Rabbi Hisda (Rabbi Sherira, Letter, 30, 16) built a new and apparently a larger structure somewhere near it. With the disuse of the old building for academic purposes, the old name ivn n'a ceased to be employed for the Sura Academy. On the other hand, the building erected by Rab was used as a synagogue (Baba Batra, sb) until the time of Rab Ashi (according to some, Mar bar Ashi ; comp. Rabbinovicz, ad loc.), and the name taoatj irn iva was retained for it, even after Rab Ashi rebuilt it. It is this synagogue that continued to be called "main irn rva down to and in the time of the Geonim. The fact that it had been remodelled by Rab Ashi justifies Nahmanides (quoted by v>"tr\, end of Rosh ha-Shanah) in saying of the Geonim that " they prayed in his [Rab Ashi's] synagogue." Halevy (II, 594) maintains that the yi '31 Nrrana rebuilt by Rab Ashi was not the synagogue of Rab in Sura, but a place of worship frequented by the scholars of N'cno MHO. But though he is right in taking Sura and N'DITO sna to be two separate places, as was proved long before him by Hirschensohn, moan MTD, s.v., and by Berliner, Beitrage, &c., 45, yet there is no doubt that each of the two names was sometimes applied indiscriminately to both places together. The epithet im applied to Rab in the Geonic time occurs in the Genizah fragment published in the J. Q. R., XVIII, 403, in Harkavy (253), and in the MS. of Ibn Hofni's "Introduction." Halevy's conjecture, that -urn rm was the Exilarch's synagogue at Bagdad, fails to recommend itself for various reasons. It is true the Exilarchs had their private synagogue ; comp. the report in Ibn Yerga, 42. But in the first place, the Exilarchs are never called irn, and in the second place, the synagogue in Bagdad, in which the Geonim worshipped and preached on the vhjm rac, had a name of its own, nbtc: ~u 'aT unurna, as Rabbi Sherira tells us explicitly, 38, 6. If it is argued that Rabbi Sherira is here speaking of a single definite time when the Geonim worshipped in this synagogue, then the proper inference from the passage is that the Exilarchs had no synagogue set apart as theirs, else it would have to be explained why they did not worship in it on this occasion. Rapoport calls attention to the fact that the Sura Geonim are the only ones who speak of the synagogue "mas? '~\ % and I shall attempt to give an approximately complete enumeration of THE GAONATE 43 a standard of superiority, we now have further evidence in favour of the pre-eminence of Sura in the five names of scholars of Sura who acted as Geonim in Pumbedita, as against the two from Pumbedita who officiated similarly in Sura, aside from the fact that the appointment of the the passages in which it is mentioned : '"fi , go, Rabbi Natronai = n*c, 55 ; j*n, 125, Rabbi Zemah, this being Rabbi Zemah ben Hayyiin of Sura, not Rabbi Zemah ben Paltoi of Pumbedita, for he quotes the Sura Geonim Rabbi Jacob and Rabbi Hanina. The same Rabbi Zemah is the author of the Responsum in E*n, 187, where a certain usage of i*a aa'tD is referred to. By many Poskim it is ascribed to Rabbi Zemah ben Paltoi. However, it can be proved that it is the Suran Rabbi Zemah. While the Suran Geonim Rabbi Natronai and Rabbi Amram agree with Rabbi Zemah, Rabbi Hai (Ibn Gajat, j*c, II, 109, and others) states that he had never seen, in any synagogue, the custom described by Rabbi Zemah. The difference of opinion can be explained only by the fact that the custom of Pumbedita varied from that in Sura in this as in so many other respects. To continue our enumeration : n*?, 220, Rabbi Natronai, who shares with the Sura Gaon Sar Shalom the peculiarity of using the expression oftener than others, comp. Albargeloni, CTiyn 'D, 172, 173, 174, 249, 281, 289; 'Aruk, s.v. -u; brViir, 50 = J?*TC, 25 a, according to the readings of MSS. S and O ; also bn'ac, 49, where the Responsum quoted is by Rabbi Natronai ; see below, p. 192. The passages listed by Marx, Untersuchungen, &c., from the Seder Rob Amram probably go back to these two Geonim also. The Responsum given in 6. S., p. 91, where aa'tt -Ta occurs, in all probability owns Sar Shalom as author, the next Responsum but one being attributed to him elsewhere, as I remark in G. S., p. 90. The Responsum on p. 119, which mentions 1*3, is surely by Rabbi Natronai. In OI'DJ, 122, the text should probably read, not jm ba? nvo'33 vuai, but with Albargeloni, I.e., 281, irn to ncian rvaai. In n*c, 287, near the end, the text is altogether corrupt : the words rrop wrn: -pi "mate wn ION om are unintelligible. Perhaps what we have here is an extract from a Responsum by a European or African disciple of Rabbi Hai, who calls his teacher iran baaatc. The words in CTID, 46 b, bottom, are also to be traced back to the Responsum by Rabbi Zemah ben Hayyiin just mentioned, in which the use of baaaw Va is spoken of. The decision cited in bn'ac, 156 (= M'2n, 83), in the name of Rashi is found in DTIE, 47 b, end, and in D*n, 187, whence also the aa*TD Va in "?n*3C and <:n, Rabbi Natronai being the author. This array of material should suffice to convince the inquirer that i:'n rva must have been a synagogue in Sura, and that in turn should suffice to identify it with the iraT rva of the Talmud, the synagogue of Rab. Comp. Marx, Untersuchungen, &c., 10-12. 44 THE GEONIM scholars of Pumbedita to Sura may probably be ascribed to the autocratical interference of an Exilarch l . In his eagerness to carry through consistently his theory of the pre-eminence of Pumbedita as compared with Sura, Halevy actually turns a scientific somersault. Only by violent means could he arrive at the desired result of reversing the true relation between the two Academies. He maintained, for instance (p. 159), that the precedence accorded the Gaon of Sura at the " reception Sabbath " of the Exilarch, of which the sources tell us, is due to the circumstance that the seat of the Exilarch was near Sura, and it was natural that first place should be ceded the Gaon of Sura in his own judicature. But unluckily Halevy himself quotes a passage (p. 154) from Sherira in which the fact is stated that the address at one of the receptions of the Exilarch in Bagdad TO"i rtan was de- livered by the Gaon of Pumbedita, and if the heads of the Sura Academy could lay claim to precedence anywhere, it was surely in Bagdad, which, as Halevy himself remarks, is situated in the immediate neighbourhood of Sura. In point of fact, the passage in Sherira from which Halevy draws support for his theory is indisputable evidence in favour of the superiority of Sura. Sherira maintains (33, 13) that the regulation, originating in the time of Rab Ashi, according to which the Exilarch held his reception at Sura, whither the Gaon of Pumbedita had to betake himself, was abolished during the Exilarchate of David ben Judah. The reason was, as Graetz correctly remarks, that the Mohammedan government no longer put its powerful assistance at the disposal of the Exilarchate. From this time on, therefore, if the Exilarchs desired to keep in touch with the Academy of Pumbedita, they had no choice but to betake themselves in person to Pumbedita and arrange for reception ceremonies there. But this statement is contradicted by two other pas- sages, one in Nathan ha-Babli's account, and one in 1 Comp. the words of Rabbi Sherira. 36, bottom, and 37, 5. THE GAONATE 45 Sherira's Letter itself. Nathan ha-Babli says that so late as his own time the two Geonim waited upon the Exilarch at his reception, which took place in the residence of the Exilarch, a suburb of Bagdad. Sherira, again, mentions the fact (38, 6), that Eabbi Abraham and Rabbi Joseph, Geonim of Pumbedita, went to Bagdad to wait upon the Exilarch. These contradictions can be harmonised. The preroga- tive enjoyed by the Exilarch, of summoning the Geonim of Pumbedita to Sura for the reception, was at the same time a prerogative of the Sura Academy. Thus the interests of the Exilarchate, in aiming to abrogate the institution, coincided with those of the Pumbedita Gaonate. As the first step towards their end the Exilarchs trans- ferred their reception to Bagdad, their residence. The Geonim of Pumbedita were only too well pleased with the change, and hastened to pay their respects to the Exilarch at Bagdad. The Geonim of Sura, on the other hand, hung back for a while, appealing to their time- honoured right, which required the presence of the Exilarch at Sura. This throws light upon Sherira's passage mentioning the address delivered by the Gaon of Pumbedita on the occasion of the Exilarch' s gala day. The chiefs of the Sura Academy simply absented themselves, and the privi- lege of delivering the address naturally devolved upon the Gaon present, the Gaon of Pumbedita. In the course of time, in the measure in which the Academy at Pumbedita gained in strength, and at the same time the Exilarchate declined, the Geonim of Pumbedita also be- came derelict, and did not appear to attest their allegiance to the Exilarchs. Interested in describing only the begin- ning and the end of the development of the relations between the Gaonate and the Exilarchate, Sherira had no intention of speaking about anything except the old institution of the Exilarch's reception at Sura and the late custom prevailing in his youth, when the Exilarchs 46 THE GEONIM came to Pumbedita. These questions of etiquette naturally were determined by the relation of the Exilarch to the Geonim at a given time, and a still more important consideration by the influence which the Exilarch could bring to bear upon the government. Some years after the reception of the Exilarch is known to have taken place at Bagdad, we find again a reference to an Exilarch who restores the old prerogative to Sura 1 . It may, of course, not be overlooked that at that moment the Gaon of Sura was Sar Shalom, a son-in-law of the Exilarch, whose predilection for Sura thus appears most natural 2 . In his reference to the homage done the Exilarchs by the Geonim, Nathan ha-Babli probably had conditions in mind as they existed at the time of the Exilarch David ben Judah, who, to judge from our data about him, was a man likely to exact as a right the consideration due to the Exilarchs, if need be by resort to the help of the state. Under him, doubtless, the Geonim found it expedient to pay their respects to the Exilarch, if not annually, at least now and then, for the sake of peace. THE TITLE GAON ORIGINALLY THE PBEROGATIVE OP SURA. It appears, then, that Sherira, so far from maintaining that Pumbedita had precedence over Sura, can be cited as a witness for the correctness of Nathan ha-Babli's state- ment of the reverse. All that is necessary is to read the text critically. The assertion made by Nathan, that the title Gaon originally appertained to the chief of the Sura Academy alone, is corroborated strikingly by the following Re- sponsum, unique in its way, sent by the Sura chief to 1 j"n, 4 ; the author is Sar Shalom ; comp. Tur, Orah Hayyim, 566, and MS. Sulzberger of the Seder Rob Amram in Marx, Untersuchungen, &c., 16. a Comp. Rabbi Hai's Responsum in the appendix to Rabbi Sherira's Letter, ed. Mayence, p. 63. The Responsum was known to the author of the Tur, as appears from Tur, Hoshen Mishpat, 7. THE GAONATE 47 the Pumbedita chief. The mere fact that the Gaon of Sura transmits a decision to the Gaon of Pumbedita, suffices to demonstrate the superiority of the former as compared with the latter. Every remnant of doubt must be banished by the official superscription. The Responsum in question has been preserved in DTia (28 a), in the MS. of the n^atn l and in JJIir "I1K (I, ii4b). It contains the decision of the Sura Gaon, Rabbi Jacob ben Mordecai (801-815), addressed to the Academy of Pumbedita, presided over by Rabbi Joseph ben Shila, with the attestation to the signature of the Gaon on the part of the Sura scholars in these words 2 : Trial n^ono xncn Nrawi *nan wonpf) pawn toro nta> warn JMIOT ma tmaviD e>n sjw an nn jun NSH " This document [of Rabbi Jacob] was seen by us, the scholars of the Academy at Mehassia, and it is intended for the court of justice of the chief of the Academy, Rabbi Joseph ben Rabbi Shila." This official superscription confirms the statement made by Nathan, that the Gaon of Sura did not address the head of the Academy at Pumbedita, but the Academy itself, and when he men- tioned the head of the Academy, he did not call him Gaon 3 . Accordingly, it is highly probable that Rabbi Samuel Resh Kalla, whose pupil, Rabbi Aha, was the author of the Sheeltot, is none other than the Rabbi Samuel, the chief of the Academy at Pumbedita, whose successor Rabbi Aha would have become if the Exilarch had not hindered it. Sherira was in the habit of conferring the title Gaon by 1 Comp. the extract from the n*'ito in iron, supplem. to the Heb. periodical -men, II, no. n, p. 18. I am indebted to Dr. A. Marx for this reference. 2 The text given is based upon a combination of the three sources mentioned in the text, all of which contain many errors. 8 The question was doubtless addressed by the head of the Academy of Pumbedita, Rabbi Joseph ben Shila, to the head of the Academy at Sura. Mere courtesy, then, required that the reply should at least recognise the existence of the questioner by mentioning his name. The case in Harkavy, 276-7, does not come in the same class. 4o THE GEONIM courtesy not only upon the chiefs of the Pumbedita Academy, but also upon Amoraim 1 who were at the head of schools. He applies the same title to Rabbi Samuel, though his disciple Rabbi Aha and other sources properly call him Resh Kalla, the title originally belonging to the heads of the Pumbedita Academy. That he actually was at the head of the Academy at Pumbedita appears par- ticularly from the passage in p'V, lyb, 7, reporting a case in law which had been submitted during several Kallas to Rabbi Samuel, who never gave a decisive answer. But if the case was so important that the questioners urged a decision, why was not the opinion of the Gaon solicited 1 To say that the difficulty was brought before Rabbi Samuel during the Gaonate of Rabbi Natronai ben Nehemiah, with whom the scholars of Pumbedita had a feud, and whom they therefore ignored, is an evasion dictated by embarrassment. In the first place, one would expect the question to be put to the Gaon of Sura in such an emer- gency, and secondly, knowing as we do from Sherira, that the scholars of Pumbedita took refuge at Sura during the Gaonate of Rabbi Natronai, it would be very sur- prising if the Resh Kalla, instead of joining them, stayed behind in Pumbedita. A further verification of the fact that this Samuel Resh Kalla was the actual head of the Pumbedita Academy is found in the report in H*n, 84 a, which tells that a certain case was decided by Rabbi Jehudai, the head of the Sura Academy, in common with Rabbi Samuel. The case, which deals with the validity of a marriage between Samaritans and Jews, being very important, the opinion of both Academies was desired. There is one difficulty to be over- come, for, according to Sherira, Rabbi Jehudai attained to the Gaonate some few years after the death of Rabbi Samuel. But Rapoport (note 24 on }n3 '~\ 'Tin) points out that the dates in this passage of Sherira' s Letter require 1 The Midrash Temurah even has the superscription xypy 'an cVc? '3iw THE GAONATE 49 such correction as would bring the beginning of Rabbi Jehudai's Gaonate earlier. It is interesting to note the modification which this passage, as cited in 1X1 TTD^n, 83, has suffered. The names of the authorities are reversed as compared with the order in J"n, and it is the correct order, for Rabbi Samuel doubtless was older than Rabbi Jehudai, who attained to office only shortly before the death of Rabbi Samuel. As for the identity of Rabbi Samuel, the head of the Academy at Pumbedita, with the Rabbi Samuel who was the teacher of Rabbi Aha, it can be demonstrated from data in Sherira's Letter. The first is there called bxiB> 2"i -in m no "Q (35, 2, below). The last word eluded every attempt at explanation, and there was nothing to do but cross it off. Now, we know from statements made by the author of the Sheeltot, that his teacher, Rabbi Samuel, came from the neighbourhood of Sura, from a place situated on the river po (see Briill, Jahrbucher, II, 149 a reference not regarded by Berliner, Beitrdge zur Geographic und Ethnographie Babyloniens, 3, s. v.). Accordingly, npw calls neither for elision nor emendation. It simply means that Rabbi Samuel came from Diakara, a town close to Bagdad and Sura. As Rapoport has shown in his )^O Tiy, 33, it is called NTpn TPK in the Talmud, and by the classic writers Diakara, which cor- responds exactly to Rabbi Sherira's contracted form np'N*r. Thus we have not only succeeded in finding the teacher of Rabbi Aha in Sherira's Letter, but at the same time we learn from it that he was a scholar of Sura, one of those presiding over the Academy at Pumbedita. As was proved above, Sherira is in the habit of recording the Suran origin of Geonirn of Pumbedita. Moreover, it is very probable that Rabbi Huna Alluf or Resh Kalla, for the two titles are identical with each other (comp. G. S., p. 242) who is mentioned in :Tn, 8 b, is the Rab Huna designated by Sherira as the chief of the Pumbedita Academy at the beginning of the seventh century. The 50 THE GEONIM passage in a"n, 34 a, should be corrected according to 2"n ed. Hildesheimer, 170, to read Win 31 "iT. It refers to the chief of the Pumbedita Academy, whom the author of J"n properly calls Alluf or Resh Kalla, but never Gaon. It appears that the head of the Pumbedita Academy, Rabbi Judah, who was in office soon after this Rabbi Samuel, is identical with the Rabbi Judah who is men- tioned in j"n, aid (ed. Hildesheimer, 131), and who, though president of the Pumbedita Academy in this early Geonic period, bears, not the title Gaon, but the title Resh Kalla, or its equivalent Alluf. The addition of llpa irum to his name does not mean that he was Resh Kalla in lips 'J, but that he hailed from that town, and was active in Pumbedita. As the scholars of "Ppa 'J at the time of the Geonim belonged to the Sura Academy four Geonim Sura came thence Rabbi Judah is found to be another of the Surans appointed to the presidency of Pumbedita. Harkavy, however (Samuel ben Hofni, note 124), goes astray in holding Rabbi Haninai, N33T wn, mentioned in "w, 3 a, 17, to be identical with the Gaon Rabbi Haninai, who does not bear the title, because at the time of Bostanai, with whom Rabbi Haninai was contemporary, the title did not yet exist. The passage cited refers to a dispute among the descendants of the Exi larch. It was altogether proper that such a case should go before "the chief judge," N331 'i, of the Exilarch (comp. G. $., p. 318, note a, and above, p. la), and not before the Geonim. Interesting as these scattered indications are, yet we have no need of them in order to establish the supremacy of Sura. The whole of Geonic literature bears irrefutable testimony to it. Up to the second third of the ninth century, the Responsa literature contains not a single Responsum by a Gaon of Pumbedita 1 , while the activity 1 Graetz, V 3 , 400, ascribes the Eesponsum in y*TB, 24 b, 10, to Rabbi Natronai ben Nehemiah, the Gaon of Pumbedita. His hypothesis that moi is simply a slip for M'-im is doubtless correct, and corroborated by the MS. reading, but the inference is by no means inevitable that THE GAONATE 51 of the Geonim of Sura began as early as the eighth century. The first Gaon of Pumbedita from whose hand we possess Responsa in numbers is Rabbi Paltoi, and the first three years of his Gaonate coincide with an interregnum in the Sura Gaonate 1 . But even the Responsa originating in Pumbedita after the time of Paltoi cannot compare with the output of Sura, either in point of quantity or quality. The Responsa bearing the names of Kohen-Zedek, Sar Shalom, Natronai, Amram, Nahshon, Zemah, Hilai, Saadia all Geonim of Sura practically form the Geonic Responsa literature until Rabbi Sherira and Rabbi Hai appear upon the scene. When the extinction of the Gaonate was immi- nent, the Geonim of Pumbedita stepped into the foreground by reason of the dissolution of the Academy at Sura. The assertion that the communities of Africa addressed their questions to the Geonim of Pumbedita, and those of Spain theirs to the Geonim of Sura, is incorrect in both its parts. Natronai, Zemah, Saadia, and even Samuel 2 , the last Gaon the Responsum was written at the time of the false Messiah ':nc, as little as Emden's zeal against Sabbatians argues his contemporaneity with Sabbatai Zebi. The authorship of Rabbi Natronai ben Hilai is confirmed by the fact that Responsum 9 in y*r, 243, is by the same* Gaon as no. 10, and in the former a plain reference is made to the Karaites. Accordingly, Natronai ben Nehemiah, who lived long before Anan, cannot be the author. Notice also the linguistic peculiarity that the Responsum is introduced with the expression Vura, a habit of the Sura Gaon Rabbi Natronai ben Hilai. Comp. oVo:, 32; y*c, 21 b, 22; and y"ic, 15 a, bis, which belong to Rabbi Natronai ben Hilai beyond the peradventure of a doubt. 1 Comp. Pardes, aid, where Rabbi Paltoi is described as rvnve' VTO *ro VIT:. " On Natronai and the scholars of Kairwan, comp. above, p. 32, note 7. Of Rabbi Zemah ben Hayyim we have not alone his correspon - dence with the scholars of Kairwan relative to Eldad, but also his Halakic Responsum addressed to the same in DTID, 21 a. The corre- spondence of Rabbi Saadia with the scholars of Kairwan is to be found in y"ir, i8b-iga, referred to above, p. 32. Even Rabbi Dosa, the son of Rabbi Saadia, corresponded with the scholars of Kairwan ; comp. noto rftnp, 72. The correspondence of Rabbi Samuel ben Hofni with the scholars of Kairwan is published in the J. Q. R., XVIII, 402. The scholars of nro with whom R. Nahshon used to correspond (Pardes, 26 d) are probably the scholars of Kairwan. E 2 52 THE GEONIM of Sura, were consulted by the African Jews, and, on the other hand, Paltoi and his son Zemah, of Pumbedita, re- ceived inquiries from Spain l . The fertility of Sura, manifested in the Responsa litera- ture, was no less noticeable in other departments. The works of the Geonic period originated there rather than in Pumbedita. Not counting the works of Hai, whose literary activity falls in a time in which the Sura Academy had gone out of existence, the only production by a Gaon of Pumbedita preserved for us is the lexicographical work of Rabbi Zemah ben Paltoi 2 . The authoritative works all originated in Sura. The author of the 3"n 3 , and Rabbi Amram and Rabbi Saadia, all occupied the Gaonate of Sura. Rabbi Amram compiled his Seder in compliance with a request addressed to him by Spanish communities, and Rabbi Saadia his order of prayers in compliance with a request addressed to him by Egyptian communities, showing that in so important a matter as the fixing of the liturgy, the communities of the Diaspora desired to have the advice of the Sura Academy alone. THE ORIGIN OF THE GAONATE UNDER THE MOHAMMEDAN RULERS. Returning for a brief resume of the results of our inquiry into Nathan's account, we find that Rabbi Samuel ha-Nagid derives his data about the Academies from Rabbi Nathan, and a source that was considered authoritative by Samuel surely deserves our confidence, too. Further, we have seen that Nathan's report has nothing to do with the Amoraic Academies ; it deals exclusively with those of the Geonic period, and by no means can the origin of the latter, as was demonstrated in detail, be relegated to the Talmudic 1 Comp. J. Q. R., XVIII, 401-2, 770. 2 And even this is doubtful ; comp. below, pp. 159-60. 3 Whoever may be designated as the author of the j'n, it is certain he must have belonged to the Sura Academy. Comp. Epstein, by a*n 'D. THE GAONATE 53 time, seeing that the older epoch knew nothing of a well- organized institution like the Gaonate, vested with great power and unquestioned authority. At the same time, our in- vestigation has completely corroborated Nathan's statement that at first there was but one Gaon, the Gaon of the Academy at Sura. Hence the transition from the schools of the time of the Amoraim and Saboraim to the Academies of the Geonic period requires an explanation that concerns itself with more than the merely Jewish conditions prevailing in Babylonia. It is in some way connected with the political situation. It must be conceded that we possess no direct historical information naming the Gaonate as an institution of the early Califate, but no other political change took place during the centuries following the redaction of the Talmud capable of producing an institution of the character of the Gaonate. The supposition made by Graetz (V 3 , 895-6), that the Gaonate arose under Ali (657), remains the only plausible hypothesis, the more so if one remembers what Sherira says regarding the kind reception which Ali accorded a great Jewish scholar, Rabbi Isaac, of Firuz- Shabor. Graetz, however, can hardly be right when he supposes that this Rabbi Isaac obtained special privileges for Sura. It is, as Halevy says if Rabbi Isaac had been inclined to be partial, his bias would have been in favour of his alma mater at Pumbedita, to which Firuz-Shabor belonged. It seems rather that what the spiritual leaders of the people secured from the new rulers was the per- mission to call into being, by the side of the Exilarchate, a religious authority with definite powers and competence. If this was so, it was natural that the chief of the old and venerable Academy at Sura should be placed at the head of the new board. In the course of time, as the Academy at Pumbedita developed more and more, its chief in the same measure gained in importance. But the parity of the two Academies reached the stage of an accomplished fact only in the time of Kohen-Zedek, when it is probable that Sura happened to be without a Gaon. 54 THE GEONIM This assumption as to the origin of the Gaonate explains at the same time the frequent occasions for friction between the Exilarchs and the Geonim of Sura until the year 689, though they disappeared for ever after that crucial time. It was natural that the Exilarchate should not accept so powerful a rival as the Gaonate of Sura without manifesting some resistance. It required almost two generations for the Exilarchs to forget their former undivided power. But scarcely had the reconciliation of the Exilarchs and the Geonim of Sura taken place when the rise of the Academy at Pumbedita gave occasion for new difficulties. From the time of Mar Yanka (719), who had been installed as Gaon at Pumbedita contrary to the wish of the Academy, until the equally arbitrary appointment of Rabbi Isaac (833)> there elapsed more than a century, during which the Pumbeditans had much to endure at the hand of the Exilarchs. The Gaonate of Sura was recognised by the State, and therefore the Exilarchate was forced to respect its rights ; while the Academy at Pumbedita possessed no privileges reinforcing its claims, and was exposed to wanton interference on the part of the Exilarchs. Finally, in 830, when the Calif Maimun decreed that ten members of a religious body sufficed for the election of a chief for themselves, the disputes between Pumbedita and the Exilarchate were silenced for ever. After this ordinance was in effect, the Gaonate of Pumbedita took and maintained its place by the side of the Gaonate of Sura as an equal power. Thenceforth, neither the Academies nor the Exilarchate could count upon the exclusive support of the government ; it was a matter of chance which gained its ear, and their differences had to be adjusted privately. These circumstances explain the fact remarked above, that Rabbi Paltoi (842) was the first of the Geonim of Pumbe- dita who issued decisions to outside communities. As long as the Gaonate of Sura was, beside the Exilarchate, the only Jewish authority recognised by the State, foreign Jews addressed their questions to the Geonim of Sura. THE GAONATE 55 After the rescript of Maimun, it depended primarily upon the learning of the Gaon in the one place or the other whether the Academy of Sura or that at Pumbedita was given the preference. NATHAN HA-BABLI'S ACCOUNT OF UKBA. We have again come round to our starting-point, and I venture to think that a satisfactory conclusion has been reached concerning the remarkable relation sub- sisting between the Exilarchs and the two Academies. Before leaving the subject, however, it would be advisable to give close consideration to the last controversy between the Academy of Pumbedita and the Exilarch. Of this controversy we have two widely divergent reports. At the end of his Letter, Sherira informs us that a quarrel broke out between two factions after the death of his grandfather Judah, in the year 917. One party favoured Mebasser 1 ; the other, with the Exilarch 1 Steinschneider, Arabische Literatur, 70, believes the name to be a translation of the Arabic Mubashshir, which is not very convincing to me. Rather I should take it to be a '1:2 for Elijah, whose appellative in Jewish literature is Mebasser, "Proclaimer of Good Tidings," without further mark of identification. In the synagogue at Aleppo there is an inscription dated 834, in honour of -ino 11 jn: 11 'to (Adler, Jews in Many Lands, 161), probably the earliest mention of the name known. In a letter dated 1029, also coming from Aleppo (D'TCW *i:a, III, i6a), there occurs a imo p nD' ; likewise in a letter of the same year, written in Egypt, a Din: 'a icao and a icio '2 rpv are mentioned (.7. Q. S., XIX, 254). In the J. Q. R., 1. c., 727, occur the following : 'V? p ITCUO, p mmn iirio, and rfro p TDTD, all from the middle of the eleventh century. That an appellative of Elijah's should be used as the name of a person is not strange ; the widespread name Emanuel is an epithet of the Messiah, as are also Zemah, the name of three of the Geonim, and in common use down to our own day, and Sar Shalom (Isa. ix. 5), which is known to have been borne by others besides the prominent Sura Gaon, as, for instance, Sar Shalom ben Joseph, the signer of a contract in Fostat in 750 (J. Q. R., XVII, 428), and the Chief Rabbi of Persia at the time when Benjamin of Tudela visited the land. Comp. also Harkavy, Saadia, 225, bottom. A propos of names in the Geonic time, is the name of the Gaon wn, identical with N'Tn used by French Jews, recorded in Gross, Gallia Judaica, 149 ? 56 THE GEONIM David at its head, favoured Kohen-Zedek, as Gaon of Pumbedita. Five years later a truce was concluded, the Exilarch gave up his opposition to Kabbi Mebasser. Nevertheless, Kohen-Zedek persisted, supported by a number of influential men, who remained loyal to him. Finally, after the death of Rabbi Mebasser, in 926, Kohen- Zedek was acknowledged Gaon by all, and he occupied the position for ten years, until his death. At first sight the account of the occurrence given by Nathan ha-Babli seems far different. He has this to say: Between the Exilarch Ukba and the Gaon Kohen- Zedek a dispute broke out on account of the revenues derived from the community of Khorasan. Ukba appro- priated them, though the moneys belonged to the Academy of Pumbedita. The Sultan, urged by the most influential of the Jews, banished the Exilarch, but he reinstated him after a year's exile, and then banished him again, this time irrevocably. Ukba emigrated to Africa. The Exilarchate, having been left vacant for a period of four or five years, the people demanded the appointment of David ben Zakkai. Their candidate was endorsed by Rabbi Amram ben Solomon, the Gaon of Sura. But Kohen- Zedek could only be prevailed upon to acknowledge the new Exilarch after a period of three years. Now, it would be possible to reconcile the differences between Sherira's account and Nathan's as they affect the relation between Kohen-Zedek and the Exilarch. As the facts are, it would not be impossible to assume that a whilom enemy, once reconciled, is transformed into a friend. But the difficulty lies elsewhere. The chrono- logical contradictions between the two sources are so numerous that Graetz's way of escape does not help the honest inquirer. Graetz accepts Nathan's account in respect to the facts of the case, and he places trust in Sherira's chronological data. Halevy justly argues against a method that is arbitrary and unscientific, and carries with it the implication that an authority like Sherira tells THE GAONATE 57 a confused and unreliable tale of events happening in his own lifetime. Halevy himself, who represents Nathan as an ignoramus living after the extinction of the Gaonate, and patching his report together from older sources which he failed to understand correctly, is even further removed from the truth than Graetz. It appears now that it is not sufficient for us to deal with a detail. The question that takes precedence is Nathan's credibility and trustworthiness. It therefore behoves us to analyse Halevy's presentation of the matter. The controversy, Halevy maintains, was not between Ukba and Kohen-Zedek, the Gaon of Pumbedita, but between Ukba and the Kohen-Zedek who was Gaon of Sura (845). But Nathan, according to Halevy, knew nothing about the older Kohen-Zedek, and he confused him with the younger man, the Gaon of Pumbedita of the same name, and, as he was aware that at some time a dispute had occurred between the Academy of Pumbedita and the Exilarch David, he constituted Kohen-Zedek the opponent of David, although Sherira informs us that the opposite was the case. As a consequence of the quarrel between Ukba and the Sura Academy, of many years' duration, Amram was appointed Gaon by the Exilarch, in opposition to the incumbent Natronai (#53-6). The celebrated Gaon Amram bar Shashna 1 , the author of the Seder, Halevy holds, is no 1 The great difficulty lies in this, that, according to Kabbi Sherira's Letter, Rabbi Amram had himself proclaimed as Gaon during the lifetime of Rabbi Natronai, while, to judge by the y"-\D, the relation between the two must have been very cordial. Not only does Rab Amram speak of Rabbi Natronai with great respect (comp. particularly his words in Marx, Untersuchungen, &c., 2), but he also quotes his Responsa on every page of his Seder. Indeed, the number of Responsa by Rabbi Natronai in the y"^o is larger than those quoted from all the other Geonim taken together. Halevy's hypothesis, so far from doing away with the difficulty, rather increases it. For if Rab Amram, as Halevy maintains, was put up as Gaon in opposition to Rabbi Natronai, during the quarrel between the Sura Academy and the Exilarch Mar Ukba, then Rab Amram was disloyal not only to Rabbi Natronai, but to the Academy as well ! This forces upon me the conjecture that the passage in question in the Letter 58 THE GEONIM other than Aim-am ben Solomon, who continued to preside over the Academy at Sura, according to Nathan's state- ment, even during the interval between the deposing of Ukba and the installation of David. The latter was generally accepted as Exilarch about 875, shortly after the death of Ami-am, and he remained in office for more than half a century. Furthermore, Halevy says, Nathan labours under a misapprehension when he states that Hai ben Kiyyumi l was the predecessor of Saadia in the Gaonate. The simple explanation is that he had heard of a Gaon of Sura named Hai, Hai ben Nahshon, and he confounded him with the celebrated Hai ben Sherira, the last Gaon of Pumbedita, and at once he was ready to make the latter Gaon of Sura, and endowed him with a father of another name. So far Halevy. For the present, we shall put aside the question as to the time and trustworthiness of Nathan, and shall confine ourselves to the consideration of Halevy's theory. by Rabbi Sherira is corrupt. I would propose the following reading : rrrn rvmi njy> DTM? nb pw rbo mn pn >opi "And before this time [before Rab Amram became Gaon], the Gaon [Rabbi NatronaiJ waived the honour due to him from Rab Amram, and the latter therefore omitted to pay his respects to him." It must be remembered that 'Tjjl* y'lc, " to show respect," is used in the Talmud, as, for instance, Baba Batra, ngb, and in j"n, 54, by Rabbi Natronai, in the sense of " yielding precedence ." Furthermore rp:n rvn: is the reverse of rrnp ivr, which Rabbi Sherira uses, 28, 5; 41, 4, to express the recognition given to a Gaon, in that the members of the Academy, including even the most prominent scholars, attended the lectures of the Gaon occasionally. Attention should be called to the fact that in this passage IT"? ybs cannot be translated by "he opposed him." For this Sherira would have used rvbr 1 , as in 41, 4. There remains only to add that the words oiny Y'm JW3 WITE: '-ft TTDH, quoted by Rabbi Aaron, of Lunel, in n"-itf, I, i8a, from Nahmanides, are to be corrected so as to read Ten -p moy '-\ TIDTI fiNa 'N:TRM 'i, as appears from Nahmanides, on Hullin, 24, who quotes Rabbi Natronai's Responsum given in J>"ID, na. A MS. of the rViN in the Sulzberger Collection of the Jewish Theological Seminary of America, contains not only the corrupt text in the edition, but C2"on instead of p"tn besides ! 1 Perhaps nothing but another way of writing wp.- THE GAONATE 59 In the first place, it is settled that David ben Judah was Exilarch in 833. Sherira and other sources 1 are unanimous on this point. After him, and before Ukba ruled, there were two Exilarchs, Natronai and Hisdai, the son of Na- tronai 2 . According to Halevy, the rule of these two Exilarchs together could not have exceeded twelve years, lor in 845, the date of the Suran Gaon Kohen-Zedek, he says, Ukba was in the thick of a conflict with the Sura Academy. Considered by itself, this brief period is not a probability, but the assumption is stamped as an impossibility by the fact that we meet with the Exilarch Hisdai as an active participant in affairs as late as the Gaonates of Natronai and Amram 3 . This disposes of the possibility of a dispute between Ukba and Kohen- Zedek of Sura. From the premise set up by Halevy, that the quarrel between the Exilarch and the Academy was caused by the revenues from Khorasan, appropriated without warrant by Ukba, it follows, he says, that the encroachments of the Exilarch brought him into conflict with Sura, and not with Pumbedita, as Khorasan is in the neighbourhood of Bagdad, the judicature of Sura. An elementary atlas might have taught Halevy that Khorasan lies only about 800 miles to the east of Bagdad! As a matter of fact, the case is precisely the reverse of its statement by Halevy. Originally, the sources of 1 Comp. Graetz, Geschichte, V 3 , 389. 1 Dukes in Ben Chananjah, IV, 141-2, from a MS. Responsum by Rabbi Zemah ben Solomon, chief judge of the court of the Exilarch Hisdai. 3 Ibn <;;im;i. in Graetz, Jubelschrifl, 17, names Rabbi Nathan ben Hananiah (comp. above, p. 32) of Kairwan as the correspondent of Rabbi Natronai, and he is the same Rabbi Hananiah to whom was addressed the Responsum, mentioned above, by Rabbi Zemah under the Exilarchate of Hisdai. As the sons of Rabbi Nathan were contemporaries of Rabbi Saadia (928) (comp. above, p. 32), he could not have nourished before the-Gaonate of Rabbi Natronai (850), and the letter of Rabbi Zemah must date from the period during which the Sura Gaonate was vacant, probably between Rabbi Malka and Rabbi Hai ben Nahshon, about 888; comp. Sherira, 39, 17. 60 THE GEONIM revenue assigned to the Exilarchate and to the Sura Gaonate were limited to Babylonia and the nearest Persian provinces. The Academy at Pumbedita, which attained to equality with Sura at a comparatively late day, had to content itself with revenues gathered in the more remote provinces. The only possible inference then is that Khorasan, situated at a considerable distance, belonged to the parish of Pumbedita. The peculiarity of Halevy's method is again illustrated by his opinion that Nathan confuses the Gaon Hai ben Nahshon with the celebrated Hai ben Sherira and then calls him Hai ben Kiyyumi. But how is it conceivable that that ignoramus Nathan, who mixed up the Geonim of Sura with the Geonim of Pumbedita, who had not the slightest knowledge of the happenings in the Academies, nor of the relation of the Academies to the Exilarchate how is it conceivable that he should have hit upon so obscure a name as Kiyyumi, he who was not even acquainted with Sherira "? In the earlier portion of this Introduction certain facts were set forth testifying to the credibility and trustworthi- ness of Nathan. We shall now pursue this subject further. The introductory words of Nathan's account, " What he himself partly saw and what he partly heard in Baby- lonia, relative to the Exilarch Ukba," are a good recom- mendation for the author. A gossip or a vagrant scribe would not have used this circumspect clause. His exactitude in the description of the vicinity of Bagdad displays itself particularly in the Arabic version, as Dr. Friedlaender shows in the above-mentioned article. A writer who is acquainted with the name of a mistress of the Persian king in whose honour a fountain had been erected centuries before, does not impress one as likely not to know the leaders of his nation at his own time, at least by name. As to what Nathan's time was there can be no doubt. In the Arabic version of his report he speaks of Natira, " the father of Sahl and Ishak," showing that he lived after THE GAONATE 6l the death of Natira, but contemporaneously with the sons of Natira. Accordingly, he had not been an eye-witness of the dispute between Ukba and the Academies, in which Natira was the chief figure, or he was too young at the time to carry personal recollections of it away with him. On the other hand, not only was he an eye-witness of the quarrel of Saadia with the Exilarch David, he was actually present when David entered upon his office in 920. Nathan's minute description of the ceremonies at the installation of an Exilarch he goes so far as to give in detail the exact height and width of the throne used by the Exilarch on the occasion admit of no doubt as to his having been present and seen such a celebration, and it could have been only at the induction of David into office, as Nathan expressly calls him the last of the Exilarchs. This offers us, not only a terminus a quo, but also a terminus ad quem. A Genizah fragment, published by Dr. Cowley in the J. Q. R., XVIII, 402, gives the information that the Exilarchate, vacant since the death of David, was filled again in 953. Nathan therefore must have written his account before 953. As, on the other hand, he mentions Aaron Sargado as Gaon of Pumbedita 1 , who entered upon office in 943, Nathan's account must have been composed between 943 and 953. 1 Halevy, 276, doubts the identification between nnc 'a ibs mentioned by Nathan and Rabbi Aaron ben Joseph, Gaon of Pumbedita, though all of seven years before the publication of Halevy's book, Harkavy had published, in D': < TO*O I > jvoi, V, the polemics of Rabbi Aaron against Rabbi Saadia, whence the identity of the two appears unmistakably ! The name Fjta = 3^3 (comp. J. Q. B., XI, 127) occurs in so early a document as one dated 750, J. Q. R., XVII, 428. From the fact that Rabbi Nathan knew no Exilarch after David ben Zakkai, it follows that his account actually ends with the passage on Sargado. By homoeoteleuton the passage on the Gaonate of Rabbi Hananiah dropped out at the end of the report. The reading should be : mvr '-\ p rv:3n VVTN -|boi ICEJI .... j'jQ'i "raD3i nsnoi c':c 'n . The various texts of Rabbi Sherira's Letter also show signs that passages have been dropped from it in this way. Frequently the names of the Geonim and the length of their official term are missing, due to the fact that the sentences between two mrai were overlooked by the copyist ; comp., for instance, ed. 62 THE GEONIM The question as to who deserves more confidence, Sherira or Nathan, may therefore not be decided, as Graetz does, in favour of the latter, on the ground of his having been closer in time to the occurrences described, for, as now appears, they were contemporaries. One must agree with Halevy, who insists that a Gaon, son and grandson of Geonim to boot, must invite greater confidence than an unknown writer. But if the two accounts are read with a critical eye it will appear that they mutually complement, and in no wise contradict each other. THE LAST CONFLICT BETWEEN THE EXILARCHATE AND THE POMBEDITA GAONATE. The controversy between Kohen-Zedek and Ukba broke out, according to the Arabic version l of Nathan's account, in the fourth year of Kohen-Zedek's Gaonate. If we / remember that even according to Sherira he was appointed^ as Gaon by/the Exilarch in the year 918, then the year 922 would have to be designated as the beginning of the dispute. A point to be noted is this, that Sherira makes Kohen-Zedek to be put into office by the Exilarch David, while, according to Nathan, Ukba was Exilarch at the time. However, the Sherira text is very doubtful in this portion. Most of the editions mention David's name three times in connexion with the Gaonate of Kohen- Zedek, but Wallerstein has it only once 2 . Moreover, this Wallerstein, 20-1. Therefore, the omission of Rabbi Hananiah's Gaonate in Nathan's narrative proves nothing derogatory to the authenticity of the narrator, as Halevy holds (275-6), but only to the correctness of our text. In Harkavy, 215, Rabbenu Hai is described as the son of Rabbi Hananiah, which, naturally, is due to homoeoteleuton. The words between irsiiN and i:'3iiN dropped out. If Rabbi Sherira's text regarding the length of Rabbi Zemah ben Kafnoi's term of office is correct, then we should read Dnrin n*i n:o in Nathan. 1 The Hebrew version has the fortieth year, which is absolutely out of the question. 2 I am indebted to Dr. Alexander Marx for the information that the Vienna MS. of Rabbi Sherira's Letter agrees with Wallerstein. THE GAONATE 63 passage in Sherira's Letter offers a great difficulty in the nature of the facts set down. The Academy, it says, appointed Rabbi Mebasser the successor to Rabbi Judah, while Kohen-Zedek was the choice of the Exilarch, and the conflict between the Academy and the Exilarch lasted five years (923). Finally, the Exilarch recognised the Gaon chosen by the Academy. But Sherira goes on and says that Kohen-Zedek, with his adherents, persisted in their schism until the death of Rabbi Mebasser, in the year 926. One would search vainly for a similar occurrence during the whole course of the Geonic time an individual op- posing the choice of both the Academy and the Exilarch. If Kohen-Zedek, as Sherira is supposed to say, was put up by the Exilarch as Gaon against the will of the Academy, then it would seem inevitable that the victory of the Academy over the Exilarch, when he finally confirmed the choice of the Academy, would cut the ground from under the feet of Kohen-Zedek. How account for the continued opposition by Kohen-Zedek ? In several other respects the occurrence is unique. It is the only case in which the Academy emerged triumphant from a contest with the Exilarch about an appointment to the Gaonate. In all other cases the Exilarch maintained the upper hand. And yet it cannot be said from what we know about him that David was a weakling. A man who was able to hold his own in opposition to Saadia and all the prominent men connected with Saadia who had influence at the court of the Calif, should meekly declare himself overcome by Rabbi Mebasser! It now behoves us to view Sherira's statements in the light afforded by the facts reported by Nathan. From an incidental remark of Nathan's we learn that Kohen- Zedek was related to Ukba, and we even learn that this relationship was the reason why he opposed the appoint- ment of David later on as Exilarch. This supplies the motive for a quarrel between Ukba and the Pumbedita 04 THE GEONIM Academy he urged the appointment of a relative, Kohen- Zedek, while the Academy installed as its chief Rabbi Mebasser, whose father had occupied the Gaonate. Then Ukba sought to make the most of the schism in the Academy, and seized upon the revenues from Khorasan, in the hope that there was no need to apprehend obstacles on the part of " his " Gaon. But it turned out to be a case of reckoning without one's host. Kohen-Zedek was too conscientious and honest to sanction such high-handed measures. Some Jews of influence at the court of the Calif managed to cause the banishment of Ukba, and the Exilarchate remained vacant some years. But blood is thicker than water, and with Kohen-Zedek the feelings of kinship were further stimulated by the recollection of the fact that he owed his position as Gaon to this relative of his who was deprived of his office. Therefore, he could not make up his mind to acknowledge David as Exilarch. He, and along with him probably a large number of distinguished men, hoped it would prove possible to induce the Calif to revoke the edict of banishment issued against Ukba. But David had no sooner been installed as Exilarch by one part of Jewry than he hastened to conclude peace with the Academy at Pumbedita and acknowledge the Gaon Rabbi Mebasser chosen by it. This explains what Sherira says, that the reconciliation between the Academy and/the Exilarch took place in 923. David lost no time in making amends to the best of his powers for the unwarranted interferences of his predecessor. But the peace thus concluded exerted no influence upon Kohen-Zedek and his followers. They refused to recognise David as Exilarch, and persisted in their opposition to him and Mebasser. According to Nathan, this opposition of Kohen-Zedek ceased only three years later, in 926. But from Sherira we learn that this was the year of Rabbi Mebasser's death, when all parties acknowledged Kohen- Zedek as Gaon. THE GAONATE 65 Here Sherira furnishes us with the motive for the reconciliation between Kohen-Zedek and David, of which Nathan gives us no hint, and which he seeks in a miracle in the real sense of the word 1 . But it is unnecessarv tf to impose a tax upon our credulity. Kohen-Zedek no longer had any reason for opposing David. His position as Gaon was now assured. And to bring about complete unanimity between Sherira and Nathan we have but to cross oif the little word nn in Sherira's Letter, 40, 18. The text then reads : TwaK xraTiDn pirn snata mm . . . pnx fro m nc& nvnp jwai . . . pw n^ao an no!> ninpi m no oy K*BO nn KB^B> vnjn 3^n r\yy WK iy sn:i^a mm nB>3Q " There was a dispute. The scholars of the Academy, held their meeting and chose Rab Mebasser as Gaon, while the Exilarch [=Ukba] named Kohen-Zedek as Gaon. The dispute lasted until Ellulof the year 233 [ = 922], when the Exilarch David concluded peace with Rabbi Mebasser." There is another possibility that the beginning of this passage is to be read K'B>3 nn nil, "the uncle of the Exilarch David." Sherira describes Ukba, the deposed Exilarch, as the uncle of David, of whom he had spoken shortly before, and to whom he had to refer again at once. As the last of the Exilarchs and the opponent of Saadia, he could suppose that his name was well known to his readers a supposition that would not hold good of Ukba. But the copyists, considering in in as dittography, either omitted the first nvi, as in Wallerstein, or inserted it in the last sentence, before tops 2 . From the beginning of the Ukba controversy until the recognition of David as Exilarch on the part of Kohen- Zedek, about eight years elapsed according to Nathan, the 1 We may safely assume that the blind T 1 : played an important part in allaying the quarrel between the Exilarch and the Gaon, even if we are not credulous enough to accept the miracle. 8 It is, however, highly probable that Rabbi Sherira at first spoke only of xnc: ( = Ukba), and afterwards, in connexion with the reconciliation with the Academy, properly mentioned nnr:n TIT, and then the 7n of the second passage was added to the NX*: of the first. I F 66 THE GEONIM same number of years being occupied, according to Sherira, by the dispute between Rabbi Mebasser and Kohen-Zedek. The only disparity between the two accounts is that, according to Nathan, Kohen-Zedek had been Gaon in 918 for more than four years, while according to Sherira it would be impossible, as it was only in that year that his grandfather Rabbi Judah died, and his death was the occasion for the dispute about the succession. There can be no doubt that the two sources are not in disagreement. We are evidently troubled by a copyist's error. We must put the date of Rabbi Judah's death one year earlier in Sherira, and we must read rut? 13, " about a year," in Nathan (78, 7, below), which was misread as [riJB>J '3, the 1 being taken for a stroke over the . This by reading jnnx for d^yniN, became pJB> 'l in the Arabic version. This assumption is further supported by the variant reading "Tn instead of f"H, for the year of Rabbi Judah's accession, and as all agree in naming eleven years as the duration of his incumbency, f"3*l results as the year of his death, and not n"31. In that case, Kohen-Zedek would have been in office about a year in rTai. THE PREDECESSOR OF SAADIA. Another difference, at first blush essential, between the two sources, concerns the Gaonate of Sura. According to Sherira, it was filled during the eight years we are now interested in by Rabbi Yom-Tob ben Rabbi Jacob. Nathan, however, names Rabbi Amram ben Solomon as the Gaon at Sura during the same period. The explanation made by Halevy of this portion of Nathan's account we repu- diated at an earlier stage. The difference between Sherira and Nathan can be reconciled only by assuming that the Gaon went by two names. There is a precedent for this. Rabbi Yom-Tob had a celebrated predecessor in the presi- dency of the Sura Academy, who also bore the name Yom-Tob, and after his entrance into office changed it. THE GAONATE 67 I refer to Rabbi Tabyomi (=Yorn-Tob), the son of Rab Ashi, who was called Mar as chief of the Academy. It is peculiar that Halevy should oppose the identification of Rabbi Yom-Tob with Rabbi Amram on the ground that, although Jews occasionally have two names, a Hebrew and a non-Hebrew, it has never happened that the same man bore two different Hebrew names. Is it conceivable that an historian of the Geonim should write thus, failing to recall that a celebrated Gaon of Sura is called Rabbi Moses in some sources, and Rabbi rvenE'D in others ? Or is a name with the ending rp less Hebrew than 3lt3 DV? One of the oldest of the Geonim of Sura, Rabbi Shashna, had the name rwiB> engraved on his official seal. So Sherira reports. In connexion with this, it is worth noting that Sherira shortens the name of the Sura Gaon Sar Shalom to Shalom. It is not surprising, then, that he should be tempted to put so long a name as Yom-Tob Amram through the same process of abbreviation, by lopping off the first half. In a much later time the case of Immanuel of Rome forms an interesting parallel to the one under consideration in the Geonic time. In the introduction to his commentary on Proverbs he calls his father Jacob, though elsewhere he appears only as Shelomoh, just as the father of our Sura Gaon is Jacob to Sherira and Solomon to Nathan. The probability is that he owned both names, nodes' apJJ', a combination not infrequently met with in later times 1 . There is still another Gaon whose father's name undergoes a trans- formation in different sources. Rabbi Paltoi is introduced as the son of Abaye by Sherira and other authorities, while the author of the Bpiri ^3K>, 420, calls his father Jacob. 1 An example in modern times is the "Lissa Rav,' 1 who calls his father nxro and also nico apr. The latter may have received his second name by means of ctrn 'i:\c, in consequence of some severe illness, though it would be rather extraordinary that it should be Jacob, the same name as his son's, an unusual occurrence among the Ashkenazim. P 2 68 THE GEONIM The only problem left unsolved in Nathan's narrative is his statement that the successor of Rabbi Amram ben Solomon and the predecessor of Saadia, in the Gaonate of Sura, was Hai ben Kiyyumi, whom he describes as " the first of his generation," and as occupying the Gaonate for twenty years, until his death. As a period of twenty years is out of the question here, and as 3 and 3 are letters easily confounded, Graetz proposes to read 2 instead of 3, so giving Hai ben Kiyyumi two years as president of the Academy instead of twenty. The objection made by Halevy to this emendation of Graetz cannot be taken seriously. " How," exclaims Halevy, " is it possible to read 3 in this passage ? How could the writer [Nathan] have been betrayed into the error of calling one 'the first of his generation ' who officiated only two years ? Can a man become the first of his generation within two years ? " It is difficult to maintain one's gravity with such reasoning. Does Halevy suppose any one would think of suggesting that Rabbi Hai was called to the Gaonate as an infant in arms? Nathan remarks that Hai received his exalted office as the first, the most distinguished, scholar of his time, and what more natural than such a remark ? Whether Rabbi Hai, a contemporary of Rabbi Saadia, deserved the title Tnn B>N1 cannot be determined after the lapse of time, but Nathan surely had as good a right to apply it to Rabbi Hai as many a modern author of Rabbinical works has to call two and sometimes three of his endorsers, on one and the same page, r6ian -03 $>3 t?tn. For the rest, this Hai apparently was not an insig- nificant personage. Saadia did not consider it beneath his dignity to quote him. Rabbi Isaac, of Vienna, in his yi"iT TIN, I, 197 a, top, cites an explanation with the words pw n n oe>3 'B jiw nnyo 3-11. As both Rabbi Hai ben David and Rabbi Hai ben Nahshon were not living at Saadia's arrival in Babylonia, it could have been no one but this Hai, who, according to Nathan, died shortly before the appointment of Saadia, and, as THE GAONATE 69 we know now 1 , Saadia lived in Babylonia for a time before he was chosen Gaon. In this period he must have made the acquaintance of Hai ben Kiyyumi, who accordingly does not owe his existence to the ignorance of Nathan, as Halevy would have us believe. It is easy to surmise why this Hai is not mentioned by Sherira, if one but scrutinises the words used by Nathan. The remark introducing him, "he was the first of his generation," yields the desired explanation. After the death of Amram ben Solomon, or, to call him by the name Sherira uses, Yom-Tob ben Jacob, Sura possessed no dominant personality worthy to act as his successor in the Gaonate. Rabbi Hai was "the greatest scholar of his circle," and as such he presided over the Academy, if not as Gaon, at least as the leading spirit. It was on his death that the Exilarch was forced to entrust the office to the alien Saadia. That is the meaning of the sentence *niD nTt?> jrw pr im3 nn B*n rrn mm. Sherira, who enumerates only the Geonirn, had no occasion to mention Rabbi Hai ben Kiyyumi, who was not a Gaon. He was content to dispose of the couple of years of hrs activity as vice-Gaon as the time when the life at Sura was at its lowest ebb. THE CHRONOLOGY OP THE GEONIM. We have reached the end of our investigation, which has resulted in a brilliant vindication of Nathan. We might stop here, except that it is proper to acknowledge the fact that the dates used here for the terms of the office of the Geonim were taken from the table contributed by A. Epstein to the Jewish Encyclopedia, s.v. "Gaon," 1 This follows from the letters in the Ben-Me'ir Controversy, the correspondence relating to which can now be examined in its entirety in Sokolow, barn 'c, 19-189. It is noteworthy that while Rabbi Sherira leaves the impression that Rabbi Saadia was called from Egypt to the Gaonate, Rabbi Nathan properly represents Rabbi Saadia as being in Babylonia when the call came to him. 70 THE GEONIM though I was well aware that, in spite of the extreme care taken in compiling it, it must remain inaccurate in some details, because it is based mainly on Sherira's Letter, of which we are not yet fortunate enough to possess an unexceptionable text, and Sherira himself is not blameless of errors and inaccuracies, especially in connexion with the older chronology. How careful one should be in such matters is illustrated by the following: In a long inquiry, extending over several pages (pp. 240-41, 248), Halevy endeavours to prove that Rabbi Zemah ben Paltoi occupied his office, not nineteen (B' S ), but nine ('3) years. Halevy's trouble was in vain. The great- grandson of this Gaon, Rabbi Hezekiah ben Samuel, SMH "no, writes in 953 l , in explicit words, that Rabbi Paltoi and his son Rabbi Zemah officiated "about forty years." From this there can be but one inference, that Zemah was in office at least nineteen years, which, added to the sixteen years of his father's incumbency, amounts to thirty-five, the " about forty years " of his great-grandson. 2 In the discussion of the point whether Rabbi Samuel Resh, Kalla, the great-grandfather of Rabbi Sherira, is identical with Rabbi Samuel Resh Kalla, the teacher of Rabbi Aha of Shabha, Halevy seems to find no particular difficulty in the fact that the latter flourished about the middle of the eighth century, while Rabbi Judah, the son of the other Rabbi Samuel, died as late as 918, for Halevy implies that this Rabbi Judah attained to the age of one hundred and thirty years. Sherira reports that the secretary to the Gaon, Rabbi Joseph (814) was pao '2N 'ONi 1j'2X "ON, which, according to Halevy's interpretation, means that Rabbi Judah, who died in 918, occupied, in 814, the high office of secretary to the Academy, and as it is not likely that so important a position Sherira tells us that the secretary to Rabbi Joseph managed the whole business of the Academy would be entrusted to a man 1 J.Q.R., XVIII, 401 ; on the writer of the letter comp. above, p. 7, n. i. 2 Comp. Kiddushin, 12 a : CON? aiip. THE G AGNATE 71 under twenty-five, we must fix the year of his birth at about 790. It is superfluous to defend so serious an historian as Sherira against the charge of imbecility involved in attributing such statements to him. The sentence quoted means nothing but this, that " the grandfather of the Gaon, who was my grandfather, was the secretary to Rabbi Joseph V Accordingly, not Rabbi Judah, but Rabbi Judah's grandfather, and the father of Rabbi Samuel Resh Kalla, was the secretary to Rabbi Joseph, and this fits the dates naturally, without the wrench of a miracle. Rabbi Judah, who died in 918, was probably born about the middle of the ninth century, and his grandfather was a personage of importance as early as 814. The Geonic period is thus the poorer by two miracles : neither Rabbi Samuel nor his son Rabbi Judah lived beyond the age of Moses. But their descendant Sherira is the gainer in his reputation for truthfulness. Accordingly, when Rabbi Sherira speaks of the Gaon Rabbi Abba ben Ami (869) as i>NVDB> n no *?v 1:2 p, we may not, in imitation of Halevy, impute to him the absurdity of meaning that he is a grandson of Rabbi Samuel, who acted as Gaon in 733. Sherira designates him as a " descendant " of this Gaon 2 . 1 Rabbi Sherira did not care to say fiNan ':pi UN UNI, because his maternal great-grandfather, Rabbi Zemah, had also been a Gaon, and the expression ':pi might have been applied to him. Also in the letter in J. Q. R., 1. c., 'IN UN UN is used for a similar reason. 8 Com p. also Rabbi Sherira, 36, 4, below, TOTDN to m '33, naturally not grandchildren, but descendants. Halevy should not have permitted himself to forget the Halakah : eras en nn D':a ':i. II. THE HALAKIC LITERATURE OF THE GEONIM. HALAKAH THE MAIN FEATUEE OF GEONIC LITERATURE. ALL the literary products of the Geonim bear the marks / of a transition period. The nihia OD^n can equally well be considered/Ian epilogue to the Talmud as y a precursor of Maimonides' Yad.J In an appraisal of the literary achieve- ments of the Geonim, the double character of the influence at work in their day must be borne in mind. On the one hand, it was the time in/which the[jtext of the Talmudlwas / -^^ fixed, and thec_Targumim and Midrashim received their final redaction,) and, on the other hand, a beginning was made in; the study of the Hebrew language, in Jewish philosophy, and in various other branches of literature and science that attained to full development in a later period, the so-called Rabbinic period. However, though poetry and philology, Targum and Midrash, mysticism and philosophy, were all represented in the time of the Geonim, the Geonic literature par excellence is after all; Halakic in/character and purport. / Rabbi Saadia is one of the fathers of Bible exegesis and Hebrew grammar, and/he may with propriety be called the earliest Jewish philosopher Philo was a Jew and a philosopher, but hardly a Jewish philosopher. But Saadia's many-sided effectiveness cannot be put to/the account of the Geonim. If he was a notable grammarian, a pioneer / philosopher, an original exegete, it was not because;he wasr a Gaon, but/in spite of having been a Gaon. Even after the decay of the Palestinian Academies, it was in the Holy Land that the study of the Bible and the cultivation of THE HALAKIC LITERATURE 73 the Haggadah were carried on zealously ] . The Masorah is a product of Palestine in the time we are considering, the greater number of the later Midrashim originated, there, and there also we must look for the beginnings of the Piyyut and of neo-Hebraic poetry. But when we come to the field of the Halakah, we must turn to Babylonia, whose Jews occupy the leading place as Halakists. The rivalry of old standing between the Palestinian and the Babylonian scholars was decided by the work of the Geo- nim once for all time in favour of the eastern centre. The Babylonian Amoraim created a Talmud ; the Geonim made of it " The Talmud." Even the Palestinians acknow- ledged its authoritativeness 2 . The historical importance of the Geonim may be summed up in this expression : They transformed a textbook into a code, and their literary activity was limited almost exclusively to the exposition and codifying of the Talmud. THE IMPULSE TO GEONIC LITERARY ACTIVITY. It is difficult to determine the date from which to reckon the beginnings of Geonic literature. The works preserved to us originated as late as the second half of the eighth century. But it is more than probable that written notes of the older Geonim, as well as their oral teachings and traditions, were embodied in the works of their successors 3 . For instance, the important decision given in 3*n, 108 a (ed. Hildesheimer, 442), relative to the wording of a docu- 1 The greater number of the so-called rroep 'DO are, it is true, Palestinian, but only their final redaction falls within the Geonic time. The works proper belong to the Tannaitic-Amoraic period. The onrc 'co, pub- lished by Schonblum in his c'nnc: onBD nave, Lemberg, 1877, is likewise pre-Geonic in its main contents. Rabbenu Hai, Vcr, II, 40, and s'rr, 189, quotes a Halakah as a D^EID rvnbrn wrvu, which is found literally in onEO 'CD . 0*1210 'on alone is a Palestinian Halakic work of the Geonim period, but the author was familiar, not only with the Babylonian Talmud, but also with the Babylonian customs of his day. He must have spent some time in a Babylonian Academy as a student. 2 Comp. above, p. 4, n. x. s Comp. y'a, 46 : 7om TDTBE = Vocw, II, 53. 74 THE GEONIM inent manumitting a slave, is cited literally by Hai, but not from this source. He introduces it with these words l : na 'nut? onno n^aca nrw nnx D^ijrxn innna nr nan an " Thus wrote the former scholars, each in his secret roll, in which they recorded, for their own use, many teachings originating with the authorities of remotest times, who lived before Rabbi Jehudai." Another passage in 3"n,96b (ed. Hildesheimer, 387-8), is quoted by Rabbi Sherira, but again not from this source 2 . He says: rwoin nnan 'tniao jaam NBTVB pan D'pa "The scholars have the following explanation [of this passage] as a tradition of the Saboraim, who lived after the redaction of the Talmud." A third passage in a*n, ai a, is quoted thence by Rabbi Hai, but he adds 3 : ir6 wna wan pin nnan Nnsian " The great men who lived after the Saboraim gave this explanation." What Rabbi Hai tells us regarding "secret rolls," for the private use of their owners, may help us to form an idea of how Geonic literature originated and developed. When the exigencies of the time made it absolutely necessary that the Talmud be put into tangible, permanent shape, the prohibition against committing the Law to writing was still not abrogated. It was merely limited in its application to all productions except the Talmud : it alone was exempt. However, here and there a disciple of the early Geonim transgressed the regulation and indulged himself to the extent of keeping a " secret roll " for his own private use, and recording there the dicta of his teachers which he desired to safeguard against oblivion. Therein the disciples of the Geonim followed the example of their Talmudic predecessors. But of actual literary 1 Albargeloni, mirren 'c, 126. 2 Halevy, 180, did not remember that this passage occurs in :"n 8 Rabbenu Nissim, on Skabbat, 12 a ; comp. Halevy, 181. THE HALAKIC LITERATURE 75 activity there was none. The impulse to produce in the real sense was supplied later, when the Geonim became the leaders of the Diaspora, and they were addressed by Jewish communities, remote and outlying as well as near by, for decisions on practical questions and for explanations of difficult Talmud passages. But the Gaonate as an institution vested with authority dates, as we have seen, from the second third of the seventh century. Before its first hundred years of institutional activity had elapsed, necessity, having first limited the application of the com- mand against committing the Law to writing, gradually abrogated it entirely 1 . The Responsa are more than the beginning of Geonic literature. They are at the same time its most important department. The phrase current in Rabbinic literature, " the Geonim say," or " the Geonim write," means one thing only, " this is to be found in a Geonic Responsum." But as their Responsa possess value collectively, in relation to the period as a whole, rather than individually, as indicative of the mental calibre of one or another author, it seems desirable, before dealing with the Responsa, to consider the Halakic-Talmudic productions of the period. RABBI AHA, OF SHABHA. The oldest work of the Geonic time are the Sheeltot "Discussions 2 ," by Rabbi Aha, of Shabha. Of the author nothing is known except that he left Babylonia about the middle of the eighth century, and settled in Palestine. 1 Comp. below, pp. 97-8 and 119-20. 2 That mrtro means not "questions," but rather " discussions," was first maintained by Muller, Briefe und Respotisen, 31, note 62, and this view is justified in detail by Mendelsohn in R. A. J., XXXII, 56 et seq. The latter makes no mention of Muller. As to the relative age of the Slieeltot and the a*n, see below, pp. 98 and 106. In beginning the discus- sion of the Halakic literature of the Geonim with the Sheeltot, I follow the accepted order. My own opinion is, as I show further on, that the nucleus of the Halakot Geddot goes back to an earlier age than the Shefltot. 76 THE GEONIM There was a reason for his emigration. In filling the Gaonate of Pumbedita the Exilarch had passed him by, disregarding his claims upon the office, paramount claims by reason of his position and his scholarship. Rabbi David of Estella, in the Provence, who lived at the beginning of the thirteenth century, speaks of works written by the Gaon Rabbi Shashna. If his statement rests upon a valid tradition l Estella confesses that he himself was acquainted with no works by this Gaon except Responsa we should have to remove the initial date of Geonic literary activity to about a century earlier than accepted facts have hitherto warranted, for the Gaon of Sura, Rabbi Shashna, also called Rabbi Mesharshia ben Tahlifa, occupied his office before 689. Unluckily, we cannot put implicit trust in Estella's assertions, as is shown by the other information he gives us about Rabbi Shashna. He describes him as "the Gaon ordained during the lifetime of Rabbi Aha, of Shabha, who was passed over at the appointment." What probably happened was that Estella wrote that wnoJ received the Gaonate instead of Rabbi Aha, and then he confounded this Natronai with the celebrated Gaon Natronai ben Hilai, the author of a number of Responsa and supposed author of a Halakic compendium 2 . In addition, a copyist twisted wniM into WVV. The next statement made by Estella, that Rabbi Aha lived after Rabbi Simon N1"p 3 , he derived from Rabbi Menahem Meiri 4 , who in turn took it from the chronicle of Rabbi Abraham Ibn Daud. RaBeD, who had a very corrupt text of Rabbi Sherira's Letter before him, may have based his statement upon the passage about Rabbi Samuel, 33, 2, below. The unusual name, "IE 3~i ID, together with the 1 A Kabbalistic author of the fourteenth century mentions a *CTDTD '",, Z.H. B., XII, 51. Is it a fictitious name? 2 Comp. below, p. 119. 3 The origin as well as the pronunciation of this name is very doubtful. With Kahira it certainly has nothing to do. * Me'iri's statements about the Geonim are full of errors, as proved below, p. 89. THE HALAKIC LITERATURE 77 unusual geographical designation npsi, which, as we have seen 1 , was misunderstood even in modern times, was " emended " to read KTpn 'B> 11 -ovni. This supposed passage of Sherira's is translated into Hebrew by RaBeD, who, after mentioning the Gaonate of Rabbi Samuel ben Mari, as he calls him, adds the words nT taicc? 'n 'oval jnp pyot? 'n iTn (63, 6). Accordingly, there is no good reason for removing R-abbi Aha from the place conceded to him as the earliest Halakic author after the close of the Talmud. But if the time of Rabbi Aha remains as before, the scene of his literary activity is open to question. Palestine and Babylonia each urges its claim upon the Sheeltot. Though the work is based exclusively upon the Babylonian Talmud, and the Palestinian Talmud is absolutely ignored in it, yet it is certain that Rabbi Aha did not compose his book until after he had settled in Palestine, whither he went when the Exilarch, for personal reasons, installed Rabbi Natronai, the secretary of Rabbi Aha, as Gaon of Pumbedita. Halevy is no less convinced (pp. 132, 211-13) of Rabbi Aha's having written his work before leaving Babylonia than he is of his having drawn upon the Palestinian Talmud in writing it, in the use of which source, he maintains, Rabbi Aha was like all the Geonim they all knew it 2 . I hope to treat 1 Comp. above, p. 49. 2 Halevy's remark on Rab Amram's relation to the Yerushalmi is characteristic. In "j*j , 58, we have Rab Amram's Responsum addressed to the scholars of Barcelona, who were led to speak of a YerusJialmi passage in their question, because its relation to the Babli was not quite clear to them. Rab Amram writes : " And the dictum of the Yentshalmi similar to this [of the Babli] which you quote, is not known to us." Ergo, reasons Halevy, it can be seen that the Yerushalmi was disseminated everywhere ! If this passage proves anything, it is an endorsement of Rapoport, Frankel, and Schorr, against whom Halevy directs his polemics. Their view is that the Babylonian Geonim did not know the Yerushalmi, but it was studied by the scholars of the aiyo , that is, of Spain and especially North Africa. Also Halevy ignores the fact that this Responsum is not really by Rab Amram, but by Rabbenu Hai, to whom it is ascribed in n'c, 119, by Albargeloni, D'nyn 'c, 212, and by Nahmanides, rrcrrto, Pesahim, X, 3. 78 THE GEONIM elsewhere of the relation of the Geonim to the Yerushalmi in detail. Here I shall confine myself to the discussion of this one point, whether or not it was used in the Sheeltot 1 . THE SHEELTOT AND THE YERUSHALMI. Halevy believes he has found two quotations from the Yerushalmi in the Sheeltot, enough to decide the question in his mind. But a superficial examination of the passages suffices to show that resort to the Yerushalmi is precluded. In Yer. Bezah, I, 60 a, the inference is made from the three superfluous words, nii> Kin . . . ^, in Exod. xii. 16, that, although the preparation of food is permitted on holidays, it is forbidden to reap, grind, and bolt. Each superfluous word points to a prohibited form of work. The passage in the Sheeltot, I, 158-9, supposed to correspond to the Bezah passage, reads : " Even work necessary for the preparation of food is permitted only if it is of a sort habitually done on the same day, such as slaughtering, baking, and cooking, but grinding and bolting, which can be done before the holiday, may not be done thereon, for the Scriptures (Exod. xii. 1 6) excluded them, saying, 'that only,' cooking, baking, and the like, may be done 2 ." While the Yerushalmi specifies three definite kinds of work excluded by the use of three superfluous words in the Scriptures, Rabbi Aha deduces a principle, applicable to all work connected with the preparation of food. This principle he finds implied in the 113^, " that only," of the Scriptures, excluding all kinds of work which as a rule are performed days before the food is prepared for the table in the restricted sense. So fundamental is this dif- ference between the Sheeltot and the Yerushalmi, that even if it were impossible to trace Rabbi Aha's real source, 1 On the relation of the Sheeltot to the Yerushalmi, see the articles by Dr. Poznanski and Dr. Kaminka, in the Hebrew periodical cipn, I, which appeared while this book was going through the press. 2 Comp. also Sheelta, CVII, 143. THE HALAKIC LITERATURE 79 we might still be sure that he was not deriving his support from the Yerushalmi. Fortunately, we are now able to assert that his source was the Mekilta de Rabbi Shiirne'on, 17, where his statement is found verbatim. Dr. Hoffmann, the learned editor of the Mekilta, would probably not have attempted the correction of the text according to the Yerushalmi if he had had the passage in the Sheeltot in mind. This Mekilta, designated by Rabbi Hai Gaon in Harkavy, 107, as 21 '31 nQ'D, in contradis- tinction to the Mekilta of Rabbi Ishmael, which he calls "the Palestinian," was naturally well known to the Babylonian Rabbi Aha, and as he not infrequently made use of the other Halakic Midrashim, his resort to the Mekilta de Rabbi Skimeon in the passage under ex- amination calls for no remark. Of course, there is no intention of denying that a close connexion exists between the Mekilta passage and Rabbi Hezekiah's dictum in the Yerushalmi. Rabbi Hezekiah modified an old Halakah in accordance with his own general system. The old Halakah, as given in the Mekilta, forbade all work con- nected with the preparation of food which as a rule is not done on the day on which the food is consumed. Illustrations are adduced reaping, grinding, bolting. These and such as these are not permissible, the prohibition being indicated by the word 113^ in the Scriptural passage. Rabbi Hezekiah, a consistent representative of the school of Rabbi Akiba 1 , who, took the particles IN and Nin as " exclusives," conceived the three sorts of work mentioned, not as illustrations of a general principle, but as an exhaustive enumeration of specific cases, finding a justi- fication therefor in the three Scriptural words, vob, Kin, and -]K. The other Yerushalmi quotation found by Halevy in the Sheeltot, XXIH, 69, requires mere collation of the two passages to demonstrate how untenable his con- 1 See the discriminating remark made by Epstein in nrjiaijnD, 53 et seq. 8o THE GEONIM tention is. Rabbi Aha writes: ntf ffb KB n^ ntf ai mb -IOK x^> 1^ tanin isio n^ -IK w ^n . In Fer. Nedarim, X, 42 a, we read : 5>D3 7^ 1S1O 1DNB> fpn nyi3K> |N3 pN I'M IW P "1B1K fptni . . . D1^3 "IEN X^ 7^- If Rabbi Aha had actually used the Yerushalmi, it would be inexplicable why he made so decided a change in the formula for the absolution from vows by a scholar, mnn D3n. Halevy permitted himself to be misled by a marginal note by Rabbi Isaiah Berlin on the Sheeltot, referring to the Yerushalmi passage. In reality, Rabbi Aha repro- duces the wording of the Babli Nedarim, Jjb, where 1^> tan "jb 1210 is given as the usual formula for ^>jn man. The attempts made by Reifmann, in the Bet-Talmud, III, 52-3, to prove Rabbi Aha's use of the Yerushalmi, are by far more serious and painstaking. Nevertheless, his conclusions are hasty. Scrutiny reveals that not one of the five passages adduced by Reifmann, in support of his opinion that the Sheeltot drew upon the Yerushalmi, can be said with certainty to have been taken by Rabbi Aha from the Palestinian Talmud. His words in I, 2, of the Sheeltot, regarding Sabbath garments, agree literally with Pesikta R., XXIII, H5b, and not with Yer. Pedh, VIII, 21 b, top, an agreement to which Friedmann in his notes on the Pesikta called attention 1 . It is therefore more probable that Rabbi Aha used either the Pesikta or one of the sources of the Pesikta, than that he used the Yeru- shalmi. Weiss's statement, 25, note 6, that the Pesikta is younger than the Sheeltot, is not a serious objection. What- ever may be its age in its present form, no one entertains a doubt that a very considerable portion of the Pesiktot is as old as the Talmud. The opinion of Rabbi Aha (XL VII, 146), that the reason 1 Comp. also Buber, Bet Talmud, III, 210, who entertains the same opinion as Friedmann, though he does not name him. However, this passage in the Sheelta does not seem to have belonged to the work in its original form. It is missing in most of the MSS., as may be seen in the first instalment of Dr. Kaminka's Sheeliot, Vienna, 1908. THE HALAKIC LITERATURE 8 1 for keeping the Day of Atonement only one day, is that a two days' fast might endanger life, has its parallel, not in the Yerushalmi alone, Hallah, I, 57 c, but also in the Babli Rosh ha-Shanah, 21 a, where Rabbi Nahman ex- claims against the Palestinian who would have had him fast a second day, " Death will be his (euphemism for 'my') end!" Jeremiah xvii. 22, is cited by both the Yervdutlmi, at the beginning of Shabbat, and the Babli, Bezah, 12 a, as the basis for the prohibition of carrying burdens on the Sabbath. Hence its use for the same purpose in the Sheeltot, XIL, 156, proves nothing conclusive as to Rabbi Aha's use of the Yerushalmi. The explanation given by the Sheeltot 1 , LV, 186, of the Babli Baba Batra, 165 a, coincides with the view of the Yer. Gittin, IX, 50 c. Nevertheless, Rabbi Aha's words are not a quotation from the YerusJtalmi, but merely an explanation, his explanation, of the Babli passage. That the formula for pen ^oa given by Rabbi Aha, LXXIV, 26-7, is not derived from Yer. Pesahim, II, 28 d, Reifmann might have deduced from the language. Not only is it Hebrew in the Yerushalmi and Aramaic in the Sheeltot, but the Aramaic is Babylonian and not Palestinian. Instead of rrovfn .... NT>en, the Palestinians would have said nwon .... Ny^en. It is interesting to note, by the way, that in the rituals the formulas vary between .... NTOH iTjv:n and rrrwn .... xjwn. The Palestinian wording of the formula and the Babylonian have come down to us side by side. It should also be noted that the Yerushalmi cites the formula on the authority of the Babylonian teacher Rab. Its use by Babylonian Jews can, therefore, be presupposed without assuming that they had to derive it from a source foreign to them. Comp. Ratner, T^IX, ad loc. Besides these seven passages enumerated by Reifmann and Halevy, I would call attention to two more, which, 1 Reifmann, in his essay in the Bet Talmud, III, 53, did not know that the Tur Hoshen Mishpat, 51, meant this Yerushalmi passage. I G 82 THE GEONIM at first sight, would seem to confirm the opinion that Rabbi Aha used the Yerushalmi for his Sheeltot. But a closer examination disposes of them as of the others. In contents the sentence in LXXIII, 25 *, . . . N1H jy "pi tt6, comes pretty close to the Yerushalmi statement in Hallah, II, 58 d, top. And yet it need not be supposed that Rabbi Aha did not derive his view from the Babli Shabbat, 76 b. The Haggadistic reason for the four cups of wine formu- lated by Rabbi Aha, LXXVII, 36, is found in the Yer. Pesahim, X, 57 c, top, but also in Genesis R., LXXXVIII. As Rabbi Aha's use of the Haggadic Midrashim in other parts of his work is not open to doubt, the probabilities are in favour of his having drawn upon the Midrash rather than the Yerushalmi as his source a likelihood that is strengthened by the fact, that for centuries after Rabbi Aha it was still customary to quote Haggadic passages from the Midrashim, even when they occurred in the Talmudim 2 . Moreover, Rabbi Aha's book, as a whole, is planned after the model of the Haggadic Midrashim on the Pentateuch, which would argue a natural preference for the Genesis Rabba as compared with the Yerushalmi. If, as to the last passage, it must be conceded that our data do not permit us to go beyond the mere supposition that Rabbi Aha drew his Haggadot from sources other than the Yerushalmi, there can yet be no doubt that the legend which he relates about Artaban and Rabbi, CXLV, 114, is not taken from the Yerushalmi Pedh, I, i5d, bottom, but from a Haggadic source, and a Babylonian Haggadic source at that. The passage ij^ DJ33 TO occurs neither in the Yerushalmi, 1. c., nor in the parallel passage in Genesis R., LXXXV, end. In contents it reminds one strikingly of the Babylonian legend about the healing of 1 The words Ncno MITT "irmb mean " to mix the chaff with the grain again.'' 2 Bashi, for instance, in his commentary on the Pentateuch, frequently quotes Genesis R. and other Midrashim, though he might have found the same passages in the Yerushalmi. THE HALAKIC LITERATURE 83 the princess by Rabbi Simon ben Yohai ', and linguistically it betrays Babylonian origin by the use of nit?, "she-devil 2 ." The Palestinians knew no female demons, and certainly not the word applied to them by Rabbi Aha. The reference to Ezra x. 8, as the Scriptural basis for the excommunicating power of the court, in the Sheeltot, CXXX, 45, Rabbi Aha did not derive, as might at first .sight be supposed, from Yer. Moed Katan, III, 81 d. His text in the Babli Moed Ifatan, i6a, doubtless read n^wn lino bn^ wm sran pnnroi f^oi instead of j^oi ins nw ann pennon. The whole detailed discussion of court procedure in the Sheeltot is taken literally from the passage in the Babli, and it would be difficult to suggest a reason for Rabbi Aha's resorting to the Yerushalmi for a single point, especially as he completely ignores the only new legal aspect presented in the Yeruskalmi 3 . The assumption here made cannot be objected to as forced, because we know that Rabbi Aha's text of the Babli frequently varies from ours, and in the passage under consideration, where our text is manifestly corrupt, the reading offered by him is an essential improvement 4 . 1 Briefly in the Talmud, Meilah, 17 b ; in detail in :*n, ed. Hildesheimer, 601-4 ; and Jellinek, Bet ha-Midrash, VI, 128-30. About a Genizah fragment of this legend, see a note by the present writer in the Z. E. B., XI, 127. 8 The Biblical mo was translated by "sedan-chair" in Palestine, and by " she-devil '' in Babylonia ; Gittin,68&. The sources enumerated in note a (with the exception of the Talmud reference) also use rmr in the sense of "she-devil.' 1 The popular belief in Babylonia could not get away from Lilith and the she-devils akin to her. Another noteworthy expression is lac, occurring in this passage of the Sheeltot, but in this sense not used in the Yerushalmi. 3 The Yerushalmi speaks of excommunication for a person who does not obey a summons to court within three days. The Babli and Rabbi Aha say nothing about the term. 4 According to our text, the same used by Rashi, mn serves as proof for cin, which contains curses, but that curses may, in certain circum- stances, be employed the Talmud derives from Neh. xiii. 25, where Db^Ni is used ! Hence there can be no doubt that the correct reading is : G 2 84 THE GEONIM It must be admitted that circumspect care is required in dealing with the Talmud text of Rabbi Aha. The pas- sage in the Sheeltot, IIV, 177, on ro-a u"> spn, is a striking illustration. In form it is much closer to Yer. Berakot, I, 2 d, than to the corresponding Babli text, Berakot, 42 a. In his learned scholia D^emi jvx nariK, 1 1 , ad loc., Ratner does not hesitate to attribute it to the Yerushalmi as Rabbi Aha's source, and yet it can readily be demonstrated, from the words of the Sheeltot, that it goes back to the Babli. In the first place, the dictum regarding the washing of the hands is attributed to the same Amora, Rabbi Hayyah bar Ashi, in the Babli as in the Sheeltot, while in the Yerushalmi, Rabbi Zeira cites it in the name of Rabbi Abba bar Jeremiah, and these latter personages appeared in the Yerushalmi text of the Geonim, as can be seen from the citations in Ratner. But there is a more important difference, the radical difference between the conception of the Babli and the conception of the Yeru- shalmi. According to the Babli, the Halakah ordains that the washing of hands must be followed at once by the saying of grace after meals, while the Yerushalmi holds that another subject is dealt with, the washing of the hands before the meal, to be followed directly by the bene- diction prescribed for it. We are here not interested in determining which of the two is the correct conception 1 . Rabbi Aha, however, does not leave us in doubt as to his viw 'JXD rrbn HION '11 D'npi JMQ tab mini TKTOOT JNO tab jrnnsn }~n murp TON .... rnr>, and not only was this the reading known to Rabbi Aha, but it was also that of the anonymous Gaon in 01*03, 217. What the Talmud wanted to derive from the verse is that the great excommunication, Din, forbids all intercourse with the excommunicated. As for the power of the court to decree excommunication, that the Talmud derived from Ezra x. 8, as may be seen from Rabbi Aha's text. Comp. also Rabbi Hananel on this passage, the text of which, as he had it, also deviates from ours. 1 The attempts to harmonise the contradictory statements in the Yerushalmi and the Bdbli on this point are futile, in spite of the fact that Rabbi Elijah Wilna countenances them in his commentary on the Orah Hayyim, 166, 2. THE HALAKIC LITERATURE 85 opinion in the matter. It accords so entirely with the view of the Babli, that every possibility is precluded of tracing his citation back to the Yerushalmi. It is true, the Scriptural passages justifying the Halakah are enumer- ated only in the Yerushalmi, which might suggest the idea that, though Rabbi Aha espouses the view of the Babli, he yet resorts to the Yerushalmi for proofs. But this suggestion may be considered disposed of by the fact that the MS. Paris of the Babli contains the Scriptural passages in the Babli. There is thus no reason why Rabbi Aha should have had to resort to the Yerushalmi. A valuable passage for the present investigation is offered by the Sheeltot, XCVI, 104-5. A- case is there discussed which was submitted by Samuel to his friend Rab, but which is not mentioned in the Talmud. In his t?SJn ^J?3, d. Berlin, zd, the RaBeD comments upon Rabbi Aha's statement with the words, " I do not know where he found it." But the RaBeD's father-in-law, Rabbi Abraham of Narbonne, in his EshJcol, I, 117, gives the Yerushalmi as Rabbi Aha's source. Whether or not the author of the Eshkol had in mind Yer. Ketubot, II, 26 c, which contains a statement similar to that in the Sheeltot, cannot be main- tained with any degree of certainty. He may have used, as is frequently done by the old authors 1 , ^em 11 to designate some apocryphal source or other. However this may be, that Rabbi Aha did not use the Yerushalmi passage in Ketubot admits of no doubt. His presentation of the case is very much more detailed than that of the Yerushalmi, and the peculiarity of Rabbi Aha, so far from being a tendency to elaborate a passage, is to condense the Tal- mudic sources. There is a positive and clinching proof, besides, to show that his source was a Babylonian and not 1 To this peculiarity Rapoport drew attention in his biography of Rabbenu Nissim, note 39, and in recent times such so-called Yerushalmi quotations were collected by Buber, Epstein, and Wolf Rabbinowitz, and published in Luncz's D'telT, VII. Rabbi Aaron, of Lunel, n*-\N, II, 179, calls our Tamid t Yerushalmi Tumid ; comp. also below, p. 157. 86 THE GEONIM a Palestinian work. The final phrase, jwr pjniK rw -&n T makes it plain ; this expression occurs nowhere but in the Babylonian Talmud 1 . Another proof of the Babylonian origin is afforded by the proverb cited, NBn K3*TO ir^n nTDVn rcb, also a Babylonian locution. Moreover, it appears from a comparison of this passage with "vntni, II, 145-6, that our text of kheSheeltot has been considerably shortened 2 ; the author of the We-Hizhir had the complete text before him, and as he has it, it could not have been taken from Ter. Ketubot, which is by far not so full of details. It is not an impossible supposition that Rabbi Aha's text of Babli Ketubot, 22 a-b, contained his whole statement, while but a few words have been preserved in our Talmud editions. PLAN AND PUKPOSE OF THE SHEELTOT. In spite of all the results attained above, it would still be an over-hasty conclusion to infer that Rabbi Aha wrote his work in the years of his life in Babylonia. Internal and external reasons alike militate against this assumption. There are, in the first place, a number of linguistic peculiarities in the Sheeltot, which clearly betray the Palestinian origin of the work. With a Babylonian like Rabbi Aha, who handled the dialect of his native land with extraordinary skill, they can be explained only as marks left upon his style by the Palestinian Aramaic of his later abode 3 . Here are some of the idiosyncrasies on 1 Ketubot, 22 b, and six other passages, marked in the margin of the Talmud. * The application of this proverb becomes intelligible only in the form it has in the Tmm ; Brull (JahrMcfier, II, 149-50), who, contrary to his usual habit, has treated this question of Rabbi Aha's use of the Yerushalmi in a very superficial way, decides in the affirmative, essentially on the basis of this passage. 3 If Rabbi Aha actually delivered lectures in Palestine, which seems very probable, the influence of the Palestinian Aramaic is all the more to be expected. THE HALAKIC LITERATURE 87 which the assertion just made is based. Rabbi Aha uses ' NJ/n NJV:n indiscriminately for Mishnah and Baraita, while the Babylonian Talmud is unfailing in drawing a sharp distinction between prwno, the Mishnah, and NJroriD, a Baraita. In this respect, Rabbi Aha follows the habit of the Terushalmi, which conveys both concepts by NrwnD. The interrogative pronoun NTTI, an exclusively Palestinian expression, is frequently used by Rabbi Aha. Similarly, the introductory formula of many of the Sheeltot, NB^, peculiar to our author, is of Palestinian derivation. In the Babylonian dialect the only permissible forms would be or Nsb^ 2 . The other formula used by our author, D"I2, is also Palestinian ; in the Aramaic of Babylonia, D~i3 is not used at all, and the connotation given to "pY by Rabbi Aha also corresponds to its Palestinian rather than its Babylonian meaning 3 . In connexion with this linguistic analysis, it must be borne in mind that Pales- tinian forms of speech were current in official and legal documents. With the customs and regulations which the Babylonian Jews imported from their Palestinian brethren, they borrowed also the language garb in which they were clothed in their original home. From the lexicographical point of view, the Targum Onkelos is the Aramaic of the Babylonian dialect, but its grammatical structure stands the most rigid tests imposed by a correct Palestinian Aramaic. The formulas prescribed by the Babli 4 for legal 1 The passages are enumerated by Reifmann, 1. c., though he failed to notice that they betrayed Palestinian influence. On this difference between the Babli and th Yerushalmi, comp. Lewy, Ueber . . . Mischna des Abba Saul, 4, note 2, and the article on "Baraita'' by the present writer, in the Jewish Encyclopedia. 3 The root F|"J disappeared from the Babylonian, with the exception of WETIN, which may be a Palestinian terminus technicus. Instead of it, nV is used, to which, of course, f]"rQ belongs, not, as Kohut, s. v., says, to rpx. He also reads mc'nto, deriving it from the Aphel, though the other form N:cbv assures the reading HJD^N from the Kal. * Rabbi Aha uses -pjt in the sense of "doubtful", as the Yerushalmi does. The use of the word in the Babli is very different. 4 Comp., for instance, Gittin, 85 b, and what is said upon the passage in G. S., p. 166. 88 THE GEONIM documents are likewise in the Aramaic dialect of Palestine, and it would not be unnatural to find that the turns of speech used in the Academies and in the houses of prayer were Palestinian. As for the formula THV D"Q, Nathan says explicitly (84, 12) that it was used by the Geonirn in their lectures. In view thereof, it is very suggestive that Nathan himself offers us the Babylonian form, K^iD^, while Rabbi Aha uses Nn^Nt?, the Palestinian form x . These internal proofs of the Palestinian origin of the Sheeltot are strengthened by reasons of an external nature. The most important Halakic product of the Geonic time, the Sheeltot are yet not mentioned by a single one of the Geonim, excepting only the last of them, Rabbi Hai. The latter has only one reference to Rabbi Aha's work, to be found in Harkavy's Collection, 191. But of Rabbi Hai we know 2 that he was in correspondence with Pales- tinian scholars, and it is natural to conjecture that the Sheeltot were brought to his notice through his intercourse with them. Even in the post- Geonic time, the scholars who make use of Rabbi Aha's work are those in particular of whom we know in other ways that Palestinian sources were accessible to them 3 . So far as I am aware, Alfasi never mentions the Sheeltot in his compendium 4 , while his younger contemporary in France, Rashi, attributes great importance to them 5 . Also, the Italian Nathan, the author 1 On this peculiar use of srfow, comp. above, p. 75, n. 2. 2 Harkavy, 29. 3 If the -vrnm was not composed in Palestine, at least it was written under Palestinian influence. Comp. Epstein," .R. fi.J., XLVI, 201, and Barnstein, in Sokolow's bivrt f c, 49. Concerning the relation of the Sheeltot to We-Hishir, see Parties, 22 a, where the text stands in need of emendation. R. Kalonymos of Lucca quotes the Sheeltot, comp. p"j, 133. 4 The benediction for yran bTO2, in Alfasi, Pesahim, I, i, is not derived from the Sheeltot. but from a Geonic Responsum, and the passage in j'Vcn 'n, 15, ed. Wilna, is a gloss. 5 Rashi copies complete sentences from the Sheeltot, and always calls the author fW3 ; comp. the Sheeltot passages cited by Rashi, in Zunz's biography of Rashi ; also the quotation from the MS. of DTID in Azulai : :"mr, s. v. NHN. THE IIALAKIC LITERATURE 89 of the 'Aruk, mentions the Sheeltot several times. Now it is well known that the Italian and the Franco-German Jews early maintained relations with Palestinian scholars, and this would explain their knowledge of the Sheeltot. We are now called upon to deal with a curious com- bination of circumstances a work composed in Palestine ignores the Yerushalmi, though its author has the oppor- tunity of citing it on every one of his pages. The explanation must be sought in the nature of the author's aim when he set himself the task of writing the book. In the introduction to his work, rrvron JV3, reprinted in Neubauer, Mediaeval Jewish Chronicles, II, 225, Rabbi Menahem Me'iri has the following to say upon this subject : " We have a trustworthy tradition that Rabbi Aha had a son who refused to devote himself to study, and for him he wrote the Sheeltot, that in reading the Pentateuch portion each week, he might at the same time be forced to familiarise himself with certain Halakic pieces." In spite of all the reverence due to so great a scholar as Me'iri, it is still difficult not to indulge in doubts of the trustworthiness of his tradition. We are expected to believe that the first work of importance after the close of the Talmud owed its existence to the laziness of an unruly boy. In general, Mei'ri's account of the Geonim is a mixture of distorted and inaccurate statements 1 , and this fact relieves us of the necessity of dealing seriously with his legend, which, besides, is denied by the plan and style of the Sheeltot. First as to the plan of the book. In the editions 2 we 1 Rabbi Nahshoii is put before Rabbi Moses, Rabbi Hai ben David officiates as the successor of Rabbi Saadia, while Kohen-Zedek and Rab Amram are called his successors ! This specimen should suffice to put a proper valuation upon Mei'ri's Geonic traditions. 3 First edition, Venice, 1546, to which the other editions go back, with the exception of ed. Wilna, for which the learned editor and commentator. Rabbi Naphtali Zebi Berlin, used manuscript material. The bibliography on the She&tot will be found rather complete attached to the present writer's article, " Aha of Shabha," in the Jewish Encyclopedia. 90 THE GEONIM have of it, it contains 1 7 1 l Sheeltot, arranged according to the weekly pericopes of the Pentateuch. Each Sheelta consists of five elements, unfortunately not always present in our printed edition. We shall take as an illustration the first Sheelta, which probably has reached us com- paratively intact. It begins thus : " Sheelta : The house of Israel is in duty bound to rest on the Sabbath day, for when the Holy One, blessed be his Name, created the world, he created it in six days, he rested on the seventh day, which he blessed and sanctified." This is the intro- duction to the first division of the Sheelta, which consists of a number of Halakot from the Talmud relating to the rest of the Sabbath day and its sanctification. Then follows the second division, beginning with the words : sjbn^ DN TIX Dia, "But this thou must learn," which introduce two Halakic questions whether a fast should be broken simultaneously with the entering of the Sabbath, as fasting on the Sabbath is forbidden, and whether the prohibition against running on the Sabbath includes run- ning to the synagogue or the house of learning. The arguments for and against having been stated briefly, the third part comes, introduced by the formula m"pl rpB> T"O ?K"1B" n*3 n>y tisbib W31 HE'D H' *?y KJlTOflDl NTI'IIK $> 3HH " Blessed be the Name of the Holy One who hath given us the Torah and the laws, by the hand of our teacher Moses, in order to instruct his people, the house of Israel." But instead of giving a decisive reply to the questions propounded, the third division consists of Halakic and Haggadic pieces taken from the Talmud Babli, and from the Midrashim, all of them such as bring out the significance of the Sabbath. After this rather lengthy portion, in the nature of a digression, the fourth division presents the answer to the two questions, introduced by the words : " And regarding the questions which you put to me," Nr6w i"Jyi> KPWT. The questions and arguments are recapitu- 1 There are two ways of counting the Sheeltot, I follow that of ed.Wilna. THE HALAKIC LITERATURE 9! lated, and on the basis of the statements of the Talmud, a conclusion is reached. The final division is a Deraahah, of which the text has preserved only the superscription l , and nothing besides. While the other four parts are still more or less distinguishable in many of the Sheeltot, the fifth part, the Derashah, has disappeared in absolutely every instance, and even of the superscriptions only twenty-nine have come down to us 2 . In an article by the present writer, on Rabbi Aha, of Shabha, Jewish Encyclopedia, I, pp. 278-80, the conjecture was hazarded that these Derashot were talks consisting of Halakic and Haggadic material, and that the Sheeltot as we now have them were abstracts of these lectures, giving the beginning of them and the end. It now appeal's that this conjecture requires considerable modification, by reason of the new light shed upon the subject by the Genizah fragment published in G. $., pp. 354-62, which constitute the Derashah attached to Sheeita, XLIII, and pp. 365-9, the Derashah of the next Sheeita, show the character of the fifth, the concluding division of each of the Sheeltot. They are neither more nor less than literal extracts from the Babylonian Talmud, occasionally somewhat shortened, the choice of the parts of the Talmud being influenced 1 The superscription is iarr:c otpo, the fourth section of the treatise Pesahim. The beginning (50 b) deals with travelling on Friday, a subject akin to the one discussed by Rabbi Aha in this Sheeita. Reifmann, 1. c., thinks that uniffi oipo has reference to Yer. Moi ! d Katan, III, 82 d, which is out of the question. 8 Comp. the list in Reifmann, I.e. In G. S., p. 366, a marginal note by a scribe or a reader gives the order of the succession of the parts of a Sheeita agreeing with that of the editions. The probability is, however, that originally the Derashah came in the fourth place, with the introductory word ~pa. For reasons given further on it was later moved to the end of the Sheeita, and then dropped entirely. This surmise is corroborated by G. S,, p. 364, 1. 5, where -pa is followed by the heading "Derashah" together with the theme of the Derashati, though the Derashah itself is at the end, in p. 365, line 9 et seq. If I am correctly informed, the order here described as original with the Shefttot is met with in MSS. of the Sheeltot. 92 THE GEONIM by their connexion with the subject treated in a given Sheelta. The Derashah on Sheelta, XLIII, pp. 354-62, is composed of extracts from the fifth section of the treatise Baba Mezia, containing the Talmudic laws of usury, which are discussed in the Sheelta. A similar analysis holds good of the other Derashah given l . This being their character, it is now plain why the copyists omitted the Derashot. They conveyed absolutely nothing new, either in form or in content, and in later times there was no reason for rewriting what could be found in the Talmud copies. The important aspect of the Derashot is that through them light is thrown upon the purpose intended to be served by Rabbi Aha with his book. The Sheeltot have the purpose of introducing the Babylonian Talmud to the Palestinians. At the time of Rabbi Aha, we may be sure that copies of the Talmud were not too plentiful, therefore it was his aim to extract verbatim a considerable portion of it 2 , especially the practical material, and group it about the Biblical laws as they succeed each other in the Scriptures. To make his collection available for practical, pedagogic ends, Rabbi Aha, considerate of Palestinian taste, provided each section of his compendium with a lecture consisting of Halakah and Haggadah, in which a comprehensive summing up was made of one or more of the points treated ramblingly and minutely in the Derashah. From of old, the Haggadists in Palestine applied the Yelamdenu Midrash for their purposes. Their method was to take a Halakah as their starting-point, and then pass over to their real subject. Rabbi Aha followed their example to the extent that he did not exclude the Hag- gadah from his lectures, but in his scheme it occupied the same place that the Halakah had in the scheme of the 1 In this DerasJiah there are even extracts from the Mishnah. Probably they were followed by the Talmud passages applying to them. a If the Derashah reproduced in G. S., pp. 354-62, is a proper criterion as to the length of the Derashot, Rabbi Aha extracted about one-fifth of the whole Talmud ! THE HALAKIC LITERATURE 93 Palestinian Haggadists 1 . The Haggadah was his starting- point, his real subject was the Halakah. To the Haggadists he owed also the arrangement of the material according to the weekly lesson from the Pentateuch, which had never before him been attempted by a Babylonian, nor was there one to attempt it after him 2 . In Babylonia, the home of the Halakah, a plan on this basis would have been entirely unnatural, in view of the fact that the first book of the Pentateuch is purely narrative, as are also large portions of the second, fourth, and fifth, and therefore altogether unsuitable as a basis for legal discussions. Palestine, on the other hand, was the home of Haggadistic interpretation, for which the Pentateuch was chosen with instinctive dis- cernment. Kabbi Aha shows a fine sense for the peculiarity of his new surroundings, when he accepts, for Halakic purposes, the model furnished him by the Haggadists. But docile as he was, he could not prevent himself from betraying his Babylonian origin. Instead of using as the basis of his work the triennial cycle of Pentateuch pericopes adopted in the Holy Land, he held to the annual cycle of his native country 3 . In general, Rabbi Aha remained more or less consciously under the dominance of Babylonian customs during his sojourn in Palestine. His predilection appears notably in the fact that he did not attach his discussion upon the importance of the study of the Torah to the Biblical law 1 Graetz, Geschichte, V s , 162, has completely reversed the true relation of Rabbi Aha to the Haggadic Midrashim, when he maintains that the Sheeltot served as a model for the later Haggadic collections, by which he means the Tanhuma Midrashim. 8 Of all the Midrashim, the #"~\ ^f\s may be designated as Babylonian, and although it is essentially a Haggadic elaboration of the narratives in the first book of the Bible, it still is not arranged according to the Pentateuch lessons. 3 Doubtless the influence of the Babylonians must have made itself felt in this respect in the time of Rabbi Aha, and probably there were "Babylonian synagogues" in Palestine, such as had the one-year cycle of Pentateuch lessons. On the influence of the Babylonian rituals in Palestine see G. S., p. 58. 94 THE GEONIM in Deut. vi. 7, in the section pnnxi. Instead, he displayed great ingenuity in working it into the pericope called "p "p. The reason is very simple. The " reception Sabbath " of the Exilarch in Babylonia coincided with Sabbath *]b 'p. The Geonim, or rather the Geonim of Sura, were in the habit of utilising this occasion, which attracted people from all parts, for a lecture, and naturally enough the study of the Torah was a favourite theme 1 . And it was this custom of his native land Rabbi Aha had in mind when he used the Sheelta on "ft *]*? for a disquisition on lic^n mm. How completely the Geonic and post-Geonic develop- ment of Halakic literature was moulded by Babylonia, is shown by the fact that there is but a single work patterned after the Sheeltot, the book We-Hizhir, the be- ginnings of which are probably to be placed in the tenth century. All that we know about the author is that he stood under strong Palestinian influences 2 . Not only is the We-Hizhir constructed on the same formal plan as the Sheeltot, but it embodied copious excerpts from Rabbi Aria's work, a circumstance which makes it most valuable for us, inasmuch as its text of the Sheeltot frequently differs from ours 3 . The text upon which our editions are based has suffered additions and abbreviations as well. In G. S., p. 353 et seq., below, Genizah fragments of pieces of the Sheeltot missing in the printed text have been repro- 1 Comp. above, p. 5, n. i. 2 Comp. above, p. 88, n. 3. 3 On this com p. Rapoport, jn: '~\ nnrin, note 4, and Addition i, also Reifmann, 1. c. Our SJieeltot are defective in arrangement, too. For instance, there can be no doubt that the Sheelta CXXIII on c':no 7012 belongs to the pericope NIIT: and not to frn^yrra , as the editions have it. Mahzor Vitry, 394, and Rashi's Siddw (Buber's Introduction to misn 'r, 84) quote this Sheelta properly as belonging to NIT: . Hurwitz, the editor of the Mahzor, and Buber both went astray, therefore, when they were of opinion that the Sheeltot passage in question was missing in our editions. On Sheeltot quotations in the 'Aruk, comp. Buber's letter addressed tn Kohut, in the latter's introduction to the -piy. THE HALAKIC LITERATURE 95 duced from the Taylor-Schechter collection at Cambridge 1 . On the other hand, the Halakot of Rabbi Aha, which are mentioned by Maimonides in his introduction to the Mishnah, are not a lost book, but the Sheeltot under another name. The Halakot Pesukot of Rabbi Aha, sup- posed to be mentioned by Rabbi Moses of Coucy in his 3"iDD, Commandement 50, is a printer's error as old as the second edition of 1488. The first edition, before 1480, reads properly 'XTp instead of RABBI JEHUDAI THE EARLIEST HALAKIC WRITER IN GEONIC TIMES. " Since many years until this day there was none like unto Rab Jehudai, for he was great in knowledge of the Bible, the Mishnah, the Midrash, the Tosafot, the Haggadot and in the practical law. It was his habit never to say anything he had not heard from his teacher. He was great in holiness and purity, in piety and humility, he was zealous in the fulfilment of all commands. He sacrificed himself for the sake of God 3 , and he drew men near to the Torah and to obedience to the law, and none after him was like unto him ..... Rabbi Jehudai once said, Ye have never submitted a matter to me, and I 1 There is no telling whether all these Sheeltot fragments belong to the original work of Rabbi Aha, or are later productions modelled after his work. The She&ta on the Day of Atonement, pp. 373-81, shows so many verbal agreements with the j"n that it cannot but have made use of the latter. 2 The first to call attention to this alleged Halakot of Rabbi Aha was S. Bloch, in his Hebrew translation of Zunz's biography of Rashi. Reifmann, 1. c., mentions it likewise, without referring to Bloch. Comp. also below, p. 100, n. i. * The expression Dvaicb r^sr ns YDNO is usually applied to martyrs who sacrifice life in the service of God, but the preceding word rrm shows plainly that there was no idea of conveying the notion of Rabbi Jehudai's having died a martyr's death. Rapoport's assumption, ion DID , VI, 243, that Rabbi Jehudai died a martyr, is refuted by this fragment ; comp. also Weiss, 41, n. 17. 96 THE GEONIM decided it, but that I had a proof from the Talmud for my decision, and from the practice of my teacher, who would have it from his teacher. I never rendered a decision wherefor I had only a proof from the Talmud, and not from the practice of my teacher, or wherefor I had a proof only from the practice of my teacher, and not from the Talmud." This characterisation of Rabbi Jehudai, quoted in G. S., pp. 52-3, by a younger contemporary of the great Gaon, shows how high an opinion his own time had of his ability and achievements. The centuries following his death felt the same appreciation of his mental powers. He was called the "light of the world," and a number of other epithets betokening honour and reverence x . An anony- mous author, probably a Gaon of Punibedita, flourishing about the beginning of the ninth century, could find no more effectual way of investing what he wrote with authority than by the plea that " all I have written unto you I did not write out of my own learning and wisdom, but it rests upon what I have derived, in theory and in practice, from my teacher Rabbah, the disciple of Rabbi Jehudai Gaon, may the memory of our teacher be unto a blessing and unto life in the future world 2 ." 1 Comp., for instance, S"ID, 45 a. Rabbi Sherira, Yi, 43, observes that Eabbi Jehudai granted no absolution for oaths, and as a consequence the scholars of the generations succeeding him opposed the exercise of myntEJ mnn, since they would not arrogate to themselves greater authority than Rabbi Jehudai assumed ! On his aversion to absolving from oaths and vows see Nahmanides, Nedarim, end. Comp. also the Geonic Responsum in c 'Ittur, II, 2 a, where the authority of Rabbi Jehudai is given high praise. The epithet "light of the world" was probably applied to him in contrast to his blindness, while that of Rabbenu Gershom, " light of the Diaspora," is derived from Huttin, 59 b. 2 Ha-Goren, IV, 71. Harkavy's attempt to fasten this fragment, published by him, upon Rabbi Hilai, the father of Rabbi Natronai, is not successful. The strict interdict against fasting on roisj res? con- tained in this fragment contradicts the view of Rabbi Natronai (comp. G.S., p. 261, and the sources cited there in connexion with Responsum 10), and it is not conceivable that the latter would have ignored his father's position completely. Rather is it probable that the author of THE HALAKIC LITERATURE 97 Rabbi Jehudai's learning alone could not have secured these extraordinary honours for him. The impartial his- torian is forced to confess that in respect to scholarship he was outstripped by more than one of his successors. Not to mention Rabbi Saadia, whose genius was so many- sided that he became the pioneer on a number of fields of Jewish science, Rabbi Jehudai's achievements even upon the limited field of the Talmud cannot be compared with those of Sar Shalom and Natronai, to specify only a couple of the older Geonim. The Responsa by Rabbi Jehudai, if they go beyond a curt affirmative or negative, offer at best a brief reference to a Talmud passage, without further comment. Nothing of the depth of a Sar Shalom or the great erudition of a Rabbi Natronai. Indeed, the pane- gyrist quoted above recounts it as one of his distinctions that Rabbi Jehudai never said anything for which he could not find endorsement in the Talmud or in religious practice. Accordingly, Rabbi Jehudai's importance must be sought in some concrete deed which made him a commanding figure in the eyes of his contemporaries and his successors. And for a deed of this calibre we need not search far or long. The words of Rabbi Hai quoted above 1 , in which he speaks of the " secret rolls," wherein the " authorities of remotest times," " who lived before Rabbi Jehudai" were wont to record traditions " for their own use," suggest the solution. Rabbi Jehudai is the earliest author, at least the earliest Halakic author, of the Geonic time. He was the fragment was a Pumbeditan, and his teacher, ruKi, of whom Harkavy says that no mention is made of him otherwise, was the Gaon of Pumbedita, Rab Abba ben Rabbi Dudai, the nephew of Rabbi Jehudai. It is work of supererogation to prove the identity of the names nn , nam , and NIN '-\ ; however, even the versions of Rabbi Sherira's Letter, 39, have rm and NIN at for the same name. It only remains to add that the prohibition against fasting on nrra rue goes back to Rabbi Jehudai ; comp. Miiller, Handschriftiiche Jehudai Gaon zugewiesene Lehrsatze, 1 1 and 1 8. 1 Comp. above, p. 74. I H 98 THE GEONIM the first to put Halakic matter down in writing for general use, and it is from this point of view that he may and should be regarded as a pioneer. The objection will be raised that in the previous section Rabba Aha, of Shabha, a contemporary of Rabbi Jehudai, was presented as an author of a Halakic work. It is altogether probable that this contemporary of Rabbi Jehudai was stimulated to take up his pen when the latter, with all the authority of a Gaon, abrogated the prohibition against the writing down of the Halakah. The assump- tion, in itself highly probable, that so important a change emanated from a Gaon invested with dignity and power rather than a private individual, finds corroboration in the chronological data marshalled in the first part of this Introduction. It was shown above, p. 48, that the Gaon of Pumbedita, Rabbi Samuel, was still alive when Rabbi Jehudai entered upon the Gaonate of Sura. Furthermore, we know that Rabbi Aha wrote his Sheeltot after his removal to Palestine, and this event did not take place until after the death of Rabbi Samuel. But at bottom the Sheeltot do not affect the present point. In Palestine the prohibition against the writing down of Halakah had ceased to be enforced with rigour back in the Talmudic time 1 . So that even if the Sheeltot had not remained unknown in Babylonia, being a Palestinian product, they still would have had no influence upon the question of Halakic authorship in Babylonia. 1 Comp. Temurah, 14 a ; the beginnings of the practice of writing down the Halakah are probably to be sought in the srwoo WTUN, the written communications sent from Palestine to Babylonia. The sharp condemna- tion by Rabbi Johanan of the practice of writing down the Halakah, Temurah, 1. c. , is not found in the Yerushcdmi, while there is, in Ter. Berakot, V, 9 a, an endorsement of Haggadic writings by Rabbi Johanan. Comp. Brull, Jahrbucher, II, 5. THE HALAKIC LITERATURE 99 CONFLICTING TRADITIONS ABOUT THE AUTHOR OP THE HALAKOT GEDOLOT. Rabbi Jehudai's priority as an Halakic author is contested by another, by Rabbi Simon {O^p 1 . The most important Halakic compendium of the Geonic period, the Halakot Gedolot, is ascribed by some old authors to Rabbi Jehudai, but others name Rabbi Simon as the author. Rabbi Abraham Ibn Daud maintains plainly that Rabbi Jehudai's Halakot Kezubot are an abstract of the Halakot Gedolot of Rabbi Simon. Halevy emends (pp. 200-13) the text of the RaBeD so that he finds the exact reverse to be the case, that it was Rabbi Simon who based his work upon Rabbi Jehudai's. It is Halevy's theory that Rabbi Jehudai wrote a Halakic compendium long before he became Gaon, and it served as the source from which his younger and less important contemporaries, Rabbi Aha, the author of the Sheeltot, and Rabbi Simon, the author of the Halakot Gedolot, drew their material. The assumption is highly improbable to repeat what was said above that the first step toward a fixation of the Halakah in writing in Babylonia proceeded from a private individual, but if it were an acceptable assumption, the priority of Rabbi Simon would be established, for the RaBeD puts the time of his activity a generation earlier than Rabbi Jehudai, and no emendation can dispose of that statement. But there is no room for doubt as to the incorrectness of the RaBeD's statement about Rabbi Simon. It clearly rests upon a misunderstanding, and it is vain to try to harmonise it with other reports of a reliable nature 2 . Rabbi Hai, as appears from his words quoted above 3 , 1 The most important literature dealing with a'rt is recorded by Epstein in his a*n IDD **? T2tro. 2 How RaBeD reached this view of his, comp. above, pp. 76-7, and Epstein, 1. c., 51. 3 Comp. above, p. 74. H 2 100 THE GEONIM assuredly considers Rabbi Jehudai the earliest author of the Geonic period, and bearing this Responsum of Rabbi Hai in mind, another passage of his, in p"a, 87, , . , , pyB> 'i KllfT an "if pn'oyox Dp vh, admits of no meaning except this : Rabbi Simon N"Vp, the compiler of the Halakot Gedolot, misunderstood the opinion of Rabbi Jehudai. Rabbi Hai's last quoted statement propounds another problem, the solution of which is extremely difficult. In this Responsum and elsewhere, Rabbi Hai clearly says that the author of 3"n was Rabbi Simon NT 11 ?, and not Rabbi Jehudai, wherein he argues with the scholars of Spain and the Provence, and is in opposition to those of France and Germany. The latter 1 name Rabbi Jehudai as the author of 3'n. In his enlightening essay upon the subject, Epstein does not hesitate to characterise the tradition of Franco-German authorities regarding the author of 3"n as an outright error. However, he makes no attempt to elicit the cause of the error. It could not have been caused by confounding J"n with the nipioa ni3/n ascribed to Rabbi 1 The older Italian scholars, as, for instance, Rabbi Isaiah di Trani the Elder, agree with the Franco-German scholars, while the younger ones seem to have wavered. Kabbi Zedekiah ben Abraham, the author of the bn'ic, in most passages calls Rabbi Jehudai the author of the :*n, yet there are places in which Rabbi Simon NT'p appears as such. Though Rabbi Abraham ben Nathan, the author of the Manhig, studied in Northern France, he wrote his work in Spain, hence he usually speaks of Rabbi Simon as the author of the 3*n, but, again, in some passages, he was dominated by the French tradition. Among the Spanish- Proven9al authors, too, there is a tendency to variation. In y't?, 14 a, Rabbi David nn p (Alfasi quotes him in n*3, 301) speaks of Rabbi Jehudai as the author of j"n, and Rabbi Isaac, the author of the 'Ittur, though he almost always considers Rabbi Simon as the author, says in one passage (II, 48 d) . . . . nobn bri ro an D'DI, which should most probably be read '3:n 'ya mv n, since the passage quoted occurs in both versions of the a*n, at the beginning of HDD, but it does not occur in the SheSltot. The same slip of the pen, making sn of mv, was shown above, p. 95, to have occurred in :*QD . There is the possibility, however, that the ' Ittur had this passage in SheSlta LXXIII and LXXIX. The description of Rabbi Aha as the 'frr 'ya was demonstrated above, p. 95, to occur in Maimonides. THE HALAKIC LITERATURE IOI Jehudai. They knew the latter work as well as the former, and the widely varying character of the two books would suggest separate authors rather than the same. Halevy, applying the Talmudic maxim, D"n DTi^K nan "km "hx, to historical data, can see no contradiction between the two opinions. He holds that the Franco-German authors had made Rabbi Jehudai the author of J"n, because they knew that for this work of his Rabbi Simon fcO"p had made constant use of the niplDS nia^n of Rabbi Jehudai. They therefore did not hesitate to describe Rabbi Jehudai as the author in the real sense. Apart from the improbability of this conjecture, which imputes to scholars of the eleventh and twelfth centuries the practice of changing the name of the author attached to a given book, on the ground of literary criticism, this alleged historical criticism was far from doing honour to the penetration of the critics. The Halakot Pesukot, it is true, are freely made use of in the present form of the Halakot Gedolot, but these two Halakot collections are so radically different in their under- lying plans, that there would be as much justification for ascribing the same author to them as for ascribing the Halakot Gedolot to Rabbi Aha, of Shabha, whose Sheeltot, too, have been drawn upon considerably therefor. Now, if it were simply a matter of choosing between Rabbi Hai's statement and the statement of European scholars, we should not have to hesitate long. The Baby- lonian Rabbi Hai, the Gaon of Pumbedita, was assuredly better informed about the author of important Halakot collections made in the Geonic time than the authorities of Germany and France living at a distance from the time and the scene of the activity of the Babylonian Halakists. However, we are in possession of a Geonic tradition very much older than Rabbi Hai's, and it tells us, in unmis- takable words, that Rabbi Jehudai is the author of the Halakot Gedolot. In a Responsum in G. S., pp. 85-6, a decision occurring in the Halakot Gedolot is repudiated on the ground that it lacks authenticity, and the view is 102 THE GEONIM expressed that it did not emanate from the author of the 3"n, but rather from Rabbi Jacob, the Gaon of Sura. If it is taken into consideration that even the last of the Geonim, Rabbi Hai and Rabbi Samuel ben Hofni, express their opinion on Rabbi Simon N'V'P plainly, indi- cating that they do not regard him as an authority 1 , the Responsum referred to would become altogether un- intelligible on the assumption that its writer looked upon Rabbi Simon as the author of 3"n. Instead of undermining the authority of the decision disputed by him, he would confirm it by attributing it to so eminent a person as Rabbi Jacob, Gaon of Sura. The Responsum conveys sense only if we assume that its writer considered Rabbi Jehudai Gaon as the author of the J"n. Now a decision emanating from him had unassailable authority in the eyes of the Geonim, and therefore the writer of the Re- sponsum adds that the moot passage had originated, not with Rabbi Jehudai, but with a disciple 2 of his, Rabbi Jacob, and the view of this Gaon he did not accept as of unquestioned authority. The writer of the Responsum under examination is not 1 Comp., for example, Rabbi Hai's rather incisive observation on Rabbi Simon in p"j, 87, and Rabbi Samuel ben Hofni's patronising words in Harkavy, 146. It may be noted, by the way, that Epstein, 1. c.. overlooked this quotation from the 3*n by Rabbi Samuel. 2 Rabbi Jacob referred to oral instructions given by Rabbi Jehudai in his presence (I"IN, I, H4b; DTiE, 2&&; and below, p. 31), as is indicated particularly by the words 'NTirv 'i TO uioi pi. The end of the Responsum by Rabbi Jacob in I"IN reads : 31 'CO 'n mo NIWD NIT< 'smrr (in me the text is corrupt), whence the inference seems to be that the teacher of Rabbi Jacob was not Rabbi Jehudai himself, but one of the pupils of the latter, perhaps Rabbi Hanina. As the death of Rabbi Jacob occurred forty years after Rabbi Jehudai's, it is possible for him to have heard Rabbi Jehudai dispense instruction, without having been a pupil of his in the true sense of the word. Comp. also ;*rr, 125, which gives the impression that Rabbi Jacob was a disciple of Rabbi Hanina. In the MS. of the n">}Ki, mentioned above, p. 47, the parallel passage reads : i:"npi mrb pi ... rmo JIN: ['::n = ] 'N ! r:n 10 nm ;iJ 'NTirr n 'BO. Accordingly, it is Rabbi Haninai, and not Rabbi Jacob, who referred to personal instructions received from Rabbi Jehudai. THE HALAKIC LITERATURE 103 mentioned, but it seems highly probable that it was issued by Rabbi Natronai ben Hilai, who elsewhere, too, accuses Rabbi Jacob of seeking to give a view of his own undue weight through the protection of Rabbi Jehudai's name 1 . Also, the expression N^in pan is frequently used by Rabbi Natronai. At all events, the rather cavalier way in which a view of Rabbi Jacob's is rejected, indicates that the author of the Responsum must be a Gaon not too far removed from Rabbi Jacob in time 2 . JEHUDAI GAON AUTHOR OF THE ORIGINAL HALAKOT GEDOLOT. Another circumstance adds to the difficulty of determining who the author of the Halakot Gedolot is. We have two widely varying versions of the book, and it is a serious task to establish which of the two, if either, is the original form. This is not the place to discuss in detail the rela- tion existing between these two versions ; one point, however, requires immediate consideration. One version, which will be designated as J" n I, mentions no authorities younger than Rabbi Jehudai Gaon 3 , while a"n II refers to 1 Comp. G. S., p. 31. ' Comp. R. Natronai's Responsum in G. : < :n mi [N:no] NSTO " And this [section] was expounded by the head of the Academy and Gaon Rabbi Hanina in the Academy." It would seem that Rabbi Hanina was disposed to give his students compilations of Haggadic material and Halakic as well. It must be admitted, however, that nobi may refer to Rabbi Samuel, and not to Rabbi Haninai. Who D"D '~\ is, mentioned in both versions of the j"n, cannot be made out. The father of the Pumbeditan Gaon Rabbi Zemah is called CB in a MS. of the Letter, instead of v:E3, but this must be merely a slip of the pen, as Rabbi Nathan also has 'x:E3. 1 Probably the reading should be 'in'p instead of 'ovp in z"n II, 548. The person meant is the Gaon of Sura (about 832), not the Gaon of Pumbedita (ab. 906), the father of Mebasser, as no Pumbeditan Geonim are mentioned in a"n with the exception of Rabbi Paltoi and his son Zemah. Responsa by a Rabbi Kimoi are to be found in the anonymous Halakic compendium published in J. Q. R., IX, 669-81, and he is pro- bably the same as our Rabbi '107. It is proper, however, to call attention to the fact that Rabbi Nathan calls the father of Rabbi Saadia's predecessor as Gaon of Sura 'ovp, and not nc'p. About jun p ipy fm \ in ;"n II, 230, we know absolutely nothing. Is it possible that he may be Rabbi Jacob of Nehar Pakod, who was Gaon of Sura about 715 ? His decision against the use of phylacteries on n'mn is in agreement with Rabbi Shashna (n'tc, 266), who officiated as Gaon of Sura about one generation earlier. At all events, the name yyi, in its Aramaic form wan, occurs at this time ; comp. above, p. 17, n. i. I am very suspicious about the genuine- ness of the end of the Responsum in n*TD, 1. c. It is missing in n"c, 155, and in I'IDD, I, 47, it forms part of a Responsum by Rabbi Moses. We can hardly be said to know Rabbi Shashna's view on nn"im pVBn. 2 This rule, of course, has its exceptions. Rabbi Isaac of Vienna also used the rroED'N to a'm. On the other hand, Albargeloni seems to have known i"n I, as was observed by Halberstam in his introduction to the m's' 'D cno, 12. Comp. above, p. 100, n. i. THE HALAKIC LITERATURE 105 and they assigned it to Rabbi Jehudai as its author, on the strength of a well-founded tradition. This work was recast about 900. by Rabbi Simon, who made many additions thereto, by reason of which additions the work acquired such popularity that it superseded the original of the great Rabbi Jehudai. Now, when Rabbi Sherira and Rabbi Hai desire to speak of Rabbi Jehudai 's work, they designate it specifically as Wirv 'n iro^n in contrast to the i"n par excellence, which circulated a century after Rabbi Simon in the form given to it by him. This "improved" version fell into the hands of the Hispano-Proven^al scholars, who properly referred to the J"n as the work of Rabbi Simon, inasmuch as they did not know its older form. Again, the anonymous writer of the Responsum in G. S., pp. 85, 86, who lived before 900, knew none but the first version, and there was no need for him to name the author, Rabbi Jehudai, explicitly. In his time no Halakot Gedolot existed except those of Rabbi Jehudai. The words of Rabbi Hai 1 , wvr 101 rnhlJ nttbm, are therefore not to be emended to read 'jnirr no ttchmi ni^na ni^m, as suggested by Epstein, but "id is to be changed to ~io"i. Rabbi Hai refers to the various readings in the a"n of Rabbi Jehudai, without concerning himself about those of Rabbi Simon, to which he attributed no particular importance. It must be admitted that Rabbi Hai cites 2 a view from the Halakot of Rabbi Jehudai which is in contradiction to a*n I. But this can hardly be brought up as an objection to the above explanation, if we consider that as early as the time of the Geonim the text of a*n had been badly tampered with 3 . We are probably dealing with a correction of 3"n I in accordance with 3"n II, a process not by any manner of means unique*. Though 1 Quoted in DTI QTin, 233, 119. * o'c, II, 66. 8 Comp. Epstein, 1. c. 4 Of the many proofs that might be brought forward, a couple follow : a*OD, Prohibition 138, cited from a*n II, which we have in a*n I, 134 d, while Commandment 63 he cites from ;*n I, with us contained in :*n 106 THE GEONIM Rabbi Simon fell far short of enjoying the respect paid his predecessor, Rabbi Jehudai, his work was used to a much larger extent than the shorter compendium of Rabbi Jehudai, who even had to submit to improvements after Rabbi Simon. A much more serious objection might be advanced, based upon the presence of Sheeltot quotations in the J"n. It is to the last degree improbable that Rabbi Jehudai would regard the work of his contemporary Rabbi Aha, whose activity, besides, displayed itself in Palestine, as of sufficient importance to be excerpted by him. But on closer examination this objection to the explanation given develops into a supporting argument. It was mentioned above that down to Rabbi Hai the Sheeltot were not mentioned by any Gaon, which makes the frequent quotations from them in the 3"n all the more remarkable. Another point to be noted is that Rabbi Aha, the author of the Sheeltot, is mentioned by name four times in :"n, but his opinions are each time intro- duced with the word "ICK. whether they are statements of his appearing in the Sheeltot, or such as are not taken thence. An interpretation of these facts would properly permit us to infer that the author of the 3"n was per- sonally acquainted with Rabbi Aha, and was told one thing and another by him in conversation, but his work, the Sheeltot, written in Palestine, was not known to Rabbi Jehudai, who may have written his own Halakic collection earlier than Rabbi Aha wrote his. Hence the Sheeltot quotations, which on their face are passages from the book reproduced literally, cannot have been put in by Rabbi Jehudai himself. The same explanation applies to them as to the fairly numerous decisions of Rabbi II, 528. The rf~\ bsr a'rt quoted by French authors was j"n II, as appears from Tosafot, ffullin, 46 b, catchword 'oaiN , yet it was not identical with our text of the second version. For example, the J"OD quotes passages from the n'n of n*i, to be found neither in 3*n I nor II. Comp. also Freimann, We-Hizhir, II, 82-3. THE HALAKIC LITERATURE IOJ Jehudai himself that are to be found in the 3*n doubtless a pupil of Rabbi Jehudai inserted, in appropriate places in his work, opinions of the master known from other sources 1 . In the same way he enriched it with intro- ductions taken from the Sheeltot. It is not impossible that this same disciple may have sat at the feet of Rabbi Aha, too, while the latter still lived at Babylonia. Accordingly, the development of the 3"n must have proceeded as follows: About the middle of the eighth century Rabbi Jehudai composed a Halakic compendium, which he named nii>na maSi 2 . This work of his was provided with additions by a pupil. The additions were mainly of two sorts, introductions 3 , taken from the Sheeltot, to comprehensive sections of the work ; and extracts from Responsa by Rabbi Jehudai, together with other of his oral and written decisions. The result was the work which came to the hands of the Franco-German scholars. This same work of Rabbi Jehudai's, with the additions and introductions inserted by his pupil, formed the foundation upon which Rabbi Simon N"T"P, in about 900, built up a remodelled work, known to the last of the Geonim and to the Hispano-Proven^al Jews as the " Halakot Gedolot of Rabbi Simon NT"p>." Originally, it is fair to assume, the latter book circulated under its full title, pyot? 'n jpnt? mbnj nia/n "the Halakot Gedolot [of Rabbi Jehudai, of course, there being no other in existence] arranged [in Hebrew, the same as composed 4 ] 1 The Mishnah, the work of Rabbi, and also the Seder Rab Amram contain teachings by their authors, who are mentioned by name, and as this does not invalidate their claims of authorship, so the frequent occurrence of Rabbi Jehudai's name in the j'rt testifies for his authorship rather than against it. In the last case, the author's blindness is an additional consideration. Many a sentence dictated by him directly may have been set down by his pupils with the introductory words, ' Rabbi Jehudai says." The title was probably derived from the Talmud, Shebu'ot, 458. ' Most of the SheSltot quotations are of this kind. * On the various uses of jpn, comp. Zunz, Gtsammelte Schrtften, III, 51, and below, p. 161. 108 THE GEONIM by Kabbi Simon." Later, familiar use wore the title down to the Halakot Gedolot of Rabbi Simon, and the name of the real author dropped into oblivion. LATER AMPLIFICATIONS OF THE HALAKOT GEDOLOT. Besides these two principal forms of the J^n, there were, of course, various texts of each, as was bound to happen with books consulted and studied as industriously as these. It was equally inevitable that they should suffer additions and omissions. Aside from the Spanish jfn, which, it will be recalled, is identical with i"n II, and, according to my opinion, corresponds to the version of Rabbi Simon, we find references in some of the old authors to a 3"n from Palestine and also a J"n from Babylonia 1 . In view thereof one is hardly justified in making categoric statements regarding the origin and author of either, on the basis of nothing more than the two printed texts of the 3*n . On pp. 383-97, in the G. S., will be found some Genizah fragments in the Taylor-Schechter Collection which agree neither with j"n I nor with J"n II. I would refer the reader particularly to p. 397, which will be seen to differ from the printed texts (io8b ; ed. Hildesheimer, 443) containing the expressions 'i:n plio JTN1. Again, in some other Genizah fragments 2 Sheeltot quotations are not met with. These 1 The author of the i"w, I, n6a, introduces a quotation with the words taan nro:o a"n, but the sentence thus introduced is to be found neither in a"rr I, nor in a"n II. The same author speaks of bxc j"na nrta 'UJNTi rorur JTI D'Jiwi owrc: (a similar description of 3*rt2 occurs in p*Vn, par. 243, 49 d, to which my attention has been called by Dr. Marx), but his meaning is not quite clear. It is possible that rnVru rvo'TTO here does not mean a work at all, but only "in important decisions." The author of the 'Ittur, II, 22 c, refers to '"NO 1N2C NT'p c*-n nwVrr ! Comp. G. S., pp. 400-1, which fragment, as is explained 1. c., p. 352, is of Palestinian origin. 3 I have in my possession, from the Taylor-Schechter Collection, a copy of a few badly damaged leaves of the a*n, which contain the section on Kliddush. The section begins : to imai 'npb 'ten DV nn 1121 : rfmm \riTp .... pn, and accordingly has not the Sheeltot quotations which are to be found in :"n I and II. THE HALAKIC LITERATURE 109 variations seem to offer strong corroboration of the view expressed above, that the original form of the a^n did not contain the Sheeltot quotations. Likewise, the Genizah fragments present an arrangement of the material departing essentially from that which we are familiar with in the printed versions J . In defining the relation of the Sheeltot to the Halakot Gedolot, an important circumstance must not be overlooked. Doubtless Rabbi Aha must have embodied a number of Halakot and Talmudic explanations, formulated in the Saboraic and early Geonic times in his work, in their literal wording. Such use of a common source would account for some of the passages that agree literatim et verbatim in the two books. As we saw above 2 , the last of the Geonim cite teachings and explanations, in the form of oral traditions, from the Saboraic and the early Geonic period, identical word for word with sentences in the 3'n. How much more may we expect to find such literal accord between contemporaries like Rabbi Jehudai and Rabbi Aha. They may have been disciples of the same teachers, and certainly were members of the same academy. Another class of Sheeltot quotations in the J'n can readily be shown to be later additions. The passage in the a*n on the insertion of ffoan y^ in the prayers on the Sabbath of Hanukkah is a case in point. The section cnn B>K"rt raen Kin snvn which occurs in both versions of the 3*n (25 c ; ed. Hildesheimer, 85) is a repetition of Sheelta XXVI, 85, but the following section pBD1M -paroi demonstrates that the author of the 3"n differs essentially from Rabbi Aha in his view of this liturgical regulation. Rabbi Aha holds that on the Sabbath of Hanukkah, by D'oan is to be inserted both in the 'Amidah and in the grace after meals; the author of the a^n insists upon the 1 Comp. the fragment reproduced below, p. 382. I have noticed in other Genizah fragments, besides, an order essentially different from the printed versions. 1 Comp. pp. 73-4, above. 110 THE GEONIM former only. This difference of opinion did not escape the notice of Kabbi Jehudai's pupil. He added to the work of his master the passage in the Sheeltot bearing upon the question, but that Rabbi Jehudai's opinion might not be contravened, he omitted Kabbi Aha's final sentence. He could not avoid stating the same Halakah in two forms, conveying the same content and differing only in their verbal terms. Side by side with each other, we have Rabbi Aha's view and Rabbi Jehudai's, on the insertion of D^wn hy on the Sabbath of HanukJcah. There are also a number of other elements whiqh, like the quotations from the Sheeltot, do not belong to the original component parts of the 3"n. Even when they occur in both versions, they are still to be looked upon as additions. At the end of the section on rv% there are three Halakot of liturgical content totally unconnected with what precedes enough to make one suspicious of their right to be considered an integral part of the real 3"n. The last of the doubtful Halakot is irrefutable evi- dence of the spuriousness of all three. It teaches that Kaddish and Baraku may be recited with but six worshippers present. The author of Masseket Soferim, X, 8, informs us that as late as his time, several centuries after Jehudai, the Babylonians insisted upon the presence of ten men, while the Palestinians contented themselves with six 1 . The only proper inference is that this passage in the 2"n was interpolated at a late time, probably after the date of Masseket Soferim, a Palestinian work cited by no Babylonian author of the Geonic period 2 . The other two Halakot are taken from the Seder Rob Amram 3 (26 a 1 The text of Mas. Soferim bears various interpretations. The conception presented in 3"n agrees with Kabbenu Tarn's ; comp. Miiller on this passage. That none of the old authors referred to the passage in j*n, may also be adduced as a proof of its spuriousness. 2 Rabbenu Hai quotes Masseket Seforim, not Masseket Soferim. Comp. above, p. 73, n. i. 3 Epstein, 1. c., mentions neither of these two quotations from the 2*n in the y"ic. THE HALAKIC LITERATURE III and 31 a). As to the first of them, it is questionable whether its form in the Seder, as we have it, is the original form. The words Wl^ *fyffOn in the Seder are very likely to be a later addition, because Albargeloni, in his DTiyn ISD, 178, says that he did not find them in a Geonic Responsum in which this Halakah was quoted. As the words in question were in the J"n used by Albargeloni 1 , as he tells us, we are obviously dealing with a comparatively old addition. The sentences and short paragraphs which we have been discussing and characterising as additions to the :Trr do not exhaust the series of interpolations to which the book was subjected. As the versions before us are constituted, there must be parts, of considerable size, not in the original plan of the book. But in order to recognise them as interjected members, it is necessary to understand clearly the underlying plan and construction of the first Halakic compendium of the post-Talmudic time. PLAN AND PURPOSE OF THE HALAKOT GEDOLOT. At the time of the Geonim the Talmud; was not only the authoritative source for/religious practices, but also the work the study of/ which constituted the chief task of a Jewish scholar. The vast accumulation of material in it, anc^its discursive manner of presenting the subject-matter, made both its practical use and theoretic investigation tasks of huge difficulty. The Karaitic schism dating from the time of Rabbi Jehudai demanded inexorably a codification of the religious laws affecting practical conduct 2 . The 1 The editor of the D'nyn 'c observes that the quotation is not to be found in our a'n ! ' l Decided anti-Karaitic tendencies manifest themselves in Rabbi Jehudai, especially in his Responsa. The most detailed of his decisions is that on the importance of p'Dn in n"ir, 153, and it is obviously directed against the Karaites, who would have nothing to do with phylacteries. Also, I entertain no doubt as to the auti-Karaitic purpose of the famous decision by Rabbi Jehudai regarding the use of DTI otD for 112 THE GEONIM scholar and the educated layman alike had to be given the possibility of readily distinguishing the true from the false, the " traditional law " from/the law of the Karaites. This goal could be reached in one of two ways. Either the Talmud had to be shortened and reshaped, so/ as to bring it within the capacity of the average scholar, or the Talmudic Halakot had to be grouped anew. These two tendencies 1 in the code literature, whose classic repre- sentatives in a later generation were Alfasi and Maimonides, respectively, existed in the Geonic time. By the side of the Geonic Halakot Gedolot there were the Geonic Halakot Pesukot or Kezubot. It cannot be supposed, therefore, that it was lack of creative ability that forced Rabbi Jehudai to shorten the Talmud, instead of systematising it anew. We could not have expected him to produce so artistic a work as the Yad of Maimonides, but it would not have transcended his powers to systematize the Halakot in their rudimentary form, as we have them systematized in the Halakot Kezubot. Rather it seems that the author of the j"n had good reasons for keeping to the arrangement of the Talmud. His work was intended to serve two purposes at once it was to be a guide for the student desirous of acquainting himself with the Talmud, and also it was to enable the scholar to decide a case submitted to him, according to law, without having to wade through the three thousand folio pages of the Talmud. Taking into consideration that it was a first attempt at these two tasks, one cannot but admit that the 3"n was a brilliant achievement. a nil, which caused such great embarrassment later. The Karaites denied totally the obligatory character of nu nVlc. Likewise, his decision in E"n, 103, on a rror who has married again without rrs'^n, is anti-Karaitic, as appears from a comparison with 'Anan's book of laws, 170. The old view is found also in a Responsum in y"c, 2 a, 10, which is not in a corrupt state, as Miiller, Mafteah, 69, note 25, thinks. It represents the old Halakah. 1 Comp. the art. "Law, Codification of the," by the present writer in the Jewish Encyclopedia. THE HALAKIC LITERATURE 113 Rabbi Jehudai's method was the following : In the first place he set about and he succeeded in excluding from his work almost all Haggadic elements. For religious practice the Haggadah had no value, and as a number of Haggadic Midrashim were at the disposal of the student, he needed no guide to this department of literature. The exclusion of the Haggadah at once produced a considerable reduction in the bulk of the material. Still keeping practical needs in mind, the author excluded also the material which no longer had application to the religious practice of his time and of the Diaspora \ The whole of the Order Kodashim z excepting the treatise Hullin alone, was not included in the j'n , nor was the treatise Hagigah of the second Order, and the treatise Sotah of the third Order. This abbreviated Talmud was condensed still more by the exclusion of the discussions as far as possible. Only the results derived from the argumentation are stated. In this way it became possible for Rabbi Jehudai to accomplish the feat, for instance, of compressing the eleven folios constituting the first chapter of the first Talmudic treatise, Berakot. into a single folio. It marks a big step forward in the direction of an independent, systematic presentation of the Talmudic material, that Rabbi Jehudai succeeded in his attempt to collect certain portions from their places here and there in the Talmud and group them together according to content. In one and the same treatise the Talmud expounds the prescription for the Sabbath lights and the prescription for the Hanukkah candles, connecting with the latter also the treatment of the Hanukkah liturgy. The same treatise contains, besides, the laws of circumcision, being introduced there incidentally to the special case of this ceremony 1 Of the Order Zera'im, he incorporated, beside D'xbD, rrnr, rfcn, which had practical bearing, also HUE, probably because in ancient times the command of Peah was executed by the pious even in Babylonia, though meant to apply only to Palestine. Comp. the Responsum in G. S., p. aaa, and the remarks introductory to it, pp. 217-18. 3 On the later additions comp. below, pp. 115-16. I I 114 THE performed on the Sabbath. The author of the 3"n has dealt with these various subjects systematically. Whatever the Talmud has to say on Hanukkah he put together under the separate and independent heading roi^n JTDpn, and whatever it has to say on circumcision went in the class of n^D mabn . A still more striking illustration of his fresh attitude is afforded by his gathering together what the Talmud has to say upon the subject of proselytes, and joining it to n^D mairi , in view of the fact that circumcision is the conditio sine qua lion for admission to Judaism. Bold as he was in these attempts of his at systematic grouping, he yet, as is natural, could not give up entirely his dependence upon the Talmud. For instance, the two subjects men- tioned, mun 'n and n^D 'n, he inserted after rats', only because the Talmud deals with them in the treatise Shabbat. The aim of the J"n, to attain to an organic system according to which to present the Halakot, is well exem- plified in the consecutive sections on the intermediate days of the festivals, on mourning, ritual defilement, the priestly blessing, synagogue ordinances, Tefillin, Mezuzot, and Zizzit. This apparent mixture of heterogeneous ele- ments is in reality a connected series. In arranging the order of the first two he followed the example of the Mishnah and the Talmud, in which they come together for the reason that the degree of abstinence from work imposed upon mourners (during the first seven days after a death) is the same as the degree imposed upon all during the intermediate days of a festival, Passover or Tabernacles. The author of the a"n logically followed up these laws for mourners by the prescriptions important for a priest in mourning. They set forth in what circumstances a priest is permitted to defile himself upon a corpse. Interested in these laws of the priest, he took occasion to speak also of the priestly blessing at the public service. These two sets of laws, on defilement and the priestly blessing, dispose of all the duties and privileges of a priest in the Diaspora and after the destruction of the Temple. But outside of THE HALAKIC LITERATURE 115 the priestly blessing, the only other element of the liturgy requiring a communal public service, is the reading from the Scriptures. The natural order, therefore, is to proceed with the regulations for reading from the Torah, the character and make-up of the scrolls, and the ordinances for the synagogue, the place at which the law is read. In effect, the scroll is identical with the Mezuzak and the Te/Ulin, so far as the rules for making them go, and in view of the holy character of the three. The sections on the two latter subjects therefore follow of themselves upon the one dealing with the mm 'D, and the next, the section on Zizzit, joins that on Tefllin without a break, both being the paraphernalia connected with the Morning Prayer. If we were to stop and analyse the whole of the a"n in the foregoing way, we should find that its author conformed as far as possible to the order of the Talmud. His procedure was novel and independent only in that he brought together, under single comprehensive headings, small portions dealing with a given subject that are scattered in many treatises. An examination of the plan of the a'n shows that the sections on mion n^yo roroo ninna DTQT could not possibly have been arranged by the author himself. They contain nothing that was of importance for the religious practice of his time, and to such portions of the Talmud Rabbi Jehudai, as we have seen, paid no attention in his book. And granted that he may have changed his system when he reached the treatises enumerated, we should still be called upon to account for the fact that he reduced the j 20 folios of the treatise Zebahim to a half-folio *. While 1 And even this half- folio, superscribed DTUI roD^n, contains a big piece from Middot and the whole of the fifth section of the Mishnah Zebahim, an unusual element in the j*n, which is in the habit of giving extracts from the Talmud, but not from the Mishnah. This fifth section of the Mishnah Zebahim formed a part of the prayer-book even in the Geonic time (see 0. S., p. 116, and R. Saadia's Commentary on Berakot, aaa), and was probably appended to the s'n by the copyists for practical purposes. I 2 Il6 THE GEONIM elsewhere Rabbi Jehudai excludes all Haggadic material on principle, his n^yo 'n consists of a single legend taken from the Talmudic treatise of the same name nothing else! Temurah is in pretty much the same case, and if we except the comparatively small portions dealing with matters of practical importance, which in other parts of the a*n are presented under the headings TW, nm, p^an, mj?, the no folios of Menahot are reduced to a half-folio! Moreover, the variations between these sections of Kodashim in the two versions of the :Tn are of so radical a nature that they can hardly be supposed to be of common origin. Though I am not in a position to give a plausible explana- tion of how these sections slipped into the 2"n , yet the proofs demonstrating their spuriousness are too convincing to admit of any doubt. To the questionable sections enumerated above we must also add the last section, 1SDH nia^n , a hodge-podge which in its present form cannot have originated with the author of the 3'n. My supposition is that it is a composite of two independent sections, which in some way were badly mixed up with each other. The one probably bore the super- scription as at present, IBOn JYGPn, the other 'iSD 'n = onao nota, " The Section on the [Biblical and Rabbinical] Writings." A copyist must have read the second as a single word, and, besides, confused the single letters 1 and "l, so that the second superscription became identical with the first, and was dropped. Rabbi Jehudai's work, which had to submit to these numerous interpolations, changes, and extensions, had to serve, besides, as the basis of two other books, retaining his name as author, viz., the 1X1 ni3?n l , called also 1 Although a great deal in it is not in our present texts of the a*n, this does not prove that other works were drawn upon for it. As was remarked before, the a*n as we have it now is anything but complete. It is curious that Epstein should maintain that the passage on i:\vcm in wi nobri, 18, and j?"nc, 45 a, is not quoted from a*n, but from the D*rt of Rabbi Jehudai ; it occurs literally in j"n II, 148, and also in a'rt I, 37 c, though in the latter place it is in shortened form, with '131 ; THE HALAKIC LITERATURE 117 , which has been edited by Schlossberg (Versailles, 1886), after an Oxford MS., and nuixp mabn, which has been published by Horowitz in the first part of his jmin D^1t?N"i ^ after a Parma MS. (Frankfort-on-the-Main, 1881). The former, the IN") rvo^n , is nothing more than a shortened Hebrew translation of parts of the a"n (so far as known, the first translation ever made from any language into Hebrew), while the latter, the nmvp 'n, is an attempt to give a resume of the J^n , by omitting the Talmudic elements. According to a statement made by Rabbi Hai 1 , this resume of the J*n and others of similar character were not compiled until fully a century after Rabbi Jehudai's time, and then outside of Babylonia. He therefore warns students to be very cautious in using these abstracts of the J"n. CODIFICATION NOT FAVOURED. A century after Rabbi Jehudai, Rabbi Paltoi (died 858), / the Gaon of Pumbedita, was asked what". was more advisable to study, the Talmud or/ the Halakot taken from it and systematically grouped. His answer was, that they who devote themselves to the study of the Halakot only do not act properly, yea, it is forbidden to do it, for they diminish Abudraham, 142, also quotes it from the j*n. That p"n and c"n respectively are based on 3"n, and not the latter on the former, is proved by the fact that the old authorities speak of nrnsp ;*n and nipiDE 2*n , meaning that the rraisp and mpiDD are taken from the :*n, Epstein, 1. c., 64, quotes rnnsp j"n from Mordecai, Shebu'at, 788, and emends it to ninsp mrrn, but the same expression occurs in many other places ; comp., for instance, Vo, 244, 416 ; and Pardes, i8b. On a single manuscript leaf in the Jew. Theol. Seminary, containing the passage from Mordecai referred to, the reading agrees with that proposed by Epstein, but it seems to be a later emendation. Comp. bn'ac, 147 : rmn 'D Vc D*nii ! 1 Comp. I'IN, II, 177 a. The enigmatic words uno fra'jpN in this Respon- sum by Rabbi Hai mean "City Secretary"; comp. in Harkavy, 86, the words of Rabbi Hai, *no IED Drabip:x, and po':pn is only another way of spelling Dio'npx, and the Responsum is cited as having been dictated by Rabbi Hai to the communal secretary. A less likely hypothesis is that Dio>:Ynp3M is to be read for MT.O po'jpN, as G. S., p. 37, which would in- dicate that the Responsum was directed to Rabbi Kalonymos, of Lucca. Il8 THE GEONIM the Torah, and in the Scriptures it is said, " It pleased the Lord, for his righteousness' sake, to magnify the law, and make it honourable"- (Isa. xlii. 21). They do still more evil ; it is they who cause the Torah to fall into oblivion. The collections of brief Halakot were not compiled for the purpose that they should become the real object of study, but for the purpose that one who has studied the whole of the Talmud, and has occupied himself with all its details, may consult the Halakot in case one or another thing seems doubtful to him, and he cannot explain it I . Rabbi Jehudai's work had a fate similar to the code of Maimonides later. Its practical advantages were so striking that the study of the Talmud was seriously menaced, and the Geonim very properly raised the voice of warning against it as an authoritative source replacing the Talmud as such. Rabbi Paltoi did not mean to deny the authority of the Halakot. He doubtless shared the universal admira- tion for their author. His aim was to make clear that the Halakot were not intended to supplant the Talmud 2 , but only to supplement it, and the above characterisation of the 3"n goes far to strengthen the position assumed by Rabbi Paltoi. During a period of nearly two centuries, the interval between Rabbi Jehudai and Rabbi Saadia, we hear of no activity in the field of the Halakah. As we have seen, the Geonim were disinclined from the work of codification. Yet it must be considered that their time and energies were absorbed in giving replies to the questions of a 1 A Responsum by Babbi Paltoi in :"n, no; in TOCN, II, 50, the question runs : mjncp rrobni picyb IN niabni pwrfj, which may be explained as asking which Halakot should be given the preference in study, the Halakot [Gedolot of Rabbi Jehudai], or the nwrap rrobn extracted from the former. The more probable meaning is that the first ni^ni stands for Talmud, the expression having been chosen under the influence of the following ma 1 *. 2 The judgment of Rabbi Paltoi on a"n is, mutatis mutandis, the same as that of the -co'vn on Maimonides' Yad ', comp. the remark in his Responsum XXXI, 9. THE HALAKIC LITERATURE 119 practical and a theoretic nature put to them replies which in part served the purposes for which one usually resorts to compendiums and reference books. What Muller says in his Mafteah, about Rabbi Natronai ben Hilai, Gaon of Sura, and a contemporary of Rabbi Paltoi, that he com- piled a series of Halakot Kezubot, cannot be proved a fact, and in view of Rabbi Paltoi's words, it is highly improbable. The " Brief Decisions " published by Horowitz in I"ETI, II, 5 et seq., after a Parma MS., are assuredly not attributable to Rabbi Natronai. They are a late compilation, without plan or system, of Geonic and old French * decisions. The Geonic portion is taken in large part from the Responsa and decisions of Rabbi Jehudai 2 . Another portion may perhaps be traceable to Rabbi Natronai's Responsa as its source 3 . As for the superscription over this conglomerate material, pw wntM 'm roen, it is, without a doubt, the invention of an untrustworthy copyist. PRAYERS FIRST PUT IN WRITING. Nevertheless, the time we are speaking of has a work to its credit which is closely akin to the Halakah, the Seder Rab Amram, originating about the middle of the ninth century. When Rabbi Jehudai ventured to set aside the old custom and permitted the writing down of the Halakah, the prayers still remained to a large extent under the ban against written transmission. A Responsum of Rabbi 1 Rabbenu Gershom is mentioned by name, p. 7. The Responsum rnya niton, 6, is by Rashi, and may be found in TmVi ncns 'n 's?n, 42, in a more correct form. Comp. Schorr, He-Haluz, XII, 97. a The brief oral decisions by Rabbi Jehudai in 7*3, 45, are most of them to be found here again. 3 The decision (p. 8) regarding a priest who left Judaism for a time is an extract from Rabbi Natronai's Responsum in a'n, 54, and D*n, 8, quoted also in biacK, I, 28. Likewise, the decision, following close upon it, regarding any renegade who returns to Judaism, goes back to Rabbi Natronai's Responsum in \"w, 24 b, 8. On the other hand, the Responsum on p. 12 regarding the sick man, contradicts the view of Rabbi Natronai as given in "jn'jc, 42 ; comp., however, 3*n, 48. 120 THE GEONIM Jehudai s informs us that the Reader at the synagogue in his time was permitted to use a prayer-book on the Day of Atonement and other fast-days. Such leniency was not extended to the festivals he was expected to recite the prayers by heart on them 1 . At a time in which the Reader was obliged to recite the prayers by heart, it goes without saying that the members of the congregation surely had no prayer-books, or at least did not use them in public. But it did not take long for the last remnants of the prohibition against the writing- down of religious works to disappear. In a Responsum, Rabbi Natronai, whose period of activity is a hundred years after Rabbi Jehudai, dis- cusses the question whether a blind man may officiate as Reader in the synagogue 2 . He decides that there is no objection to his reciting the prayers, but he may not give the lesson from the Torah, because it is imperative that the latter must be read from the scroll. This reveals that, in Rabbi Natronai's day, the general custom was for the 1 Miiller, Handschrifttiche Jehudai Gaon zugewiesene Lehrsutse, 10. Though Rabbi Jehudai was a Gaon of Sura, by education he was a Pumbeditan. Therefore it is not extraordinary for him to use the expression 0122 ~p ijn: N11D31 in his Responsum. It is interesting that opposition to the use of prayer-books should prevail as late as the time of Rabbi Ephraim, as appears from his remark in Waic, 12. The identity of this Rabbi Ephraim cannot be established with certainty. He is probably the pupil of Alfasi, and not the Rabbi Ephraim of Bonn who lived a century later. Buber, in his list of authors' names for brTatD, attributes all the passages in the book to the former Rabbi Ephraim, but there can be no doubt that the Rabbi Ephraim in 33 is the German Rabbi Ephraim, as his correspondent is the German Rabbi Joel. From c^n 'IN, I, 5!), bottom, it may be seen that no prayer-books were taken to the synagogue on week-days, though, to judge from the words of the author, this was not to be ascribed to scruples against the use of prayer-books. What Ibn Gajat says, in vfv, I, 62, regarding the recitation of the 'Abodah on the Day of Atonement, does not prove that in his time it was not written down ; it means that in some congregations it was recited only by the precentor, while the worshippers merely listened. Comp. also brt"aj, 58, TOS n'to DTU"J, which also presupposes recitation by heart. 2 Properly ascribed to Natronai in n"c, 245, and n", I, 18 a, while in I'IN, 42 a, Rabbi Jehudai appears as the author, which is not correct. The prayer-books mentioned in G. S., p. 153, belong to the time after R. Amram. THE HALAKIC LITERATURE 12T Reader to use a prayer-book, else a congregation would not have been in doubt as to the fitness of a blind man, who could recite the prayers only by heart, for the office of Reader. Of course, even after prayer-books had long been in use in Babylonia, there was no occasion for the Geonim to occupy themselves with the task of fixing the order of the prayers. With centuries of continuous development in Babylonia the conduct of the divine service lay in the hands of men who would do the right thing without the necessity of special instruction. Moreover, the judges and the other communal officials stood under the direct juris- diction of the Geonim, who would be sure to watch over the divine service and its conduct in accordance with the accepted regulations. Of the three " Orders of Prayer," it is certain that two were compiled at the request of con- gregations outside of Babylonia. Rab Amram wrote his for the Spanish congregations 1 , and Rabbi Saadia his for the Egyptian 2 , and it is altogether probable that Rabbi Hai, too, did not arrange his Seder for Babylonia 3 . The countries outside of Babylonia lacked both historical con- tinuity and a central body with acknowledged religious authority, and there were other circumstances, besides, standing in the way of securing an established order of the prayers. In spite of the high respect in which the Gaonate was held, the Jews of Europe and elsewhere were not altogether free from Palestinian influence 4 . In the depart- ment of liturgy this influence was most marked, for even after the disappearance of her Academies, Palestine still remained the home of the Piyyut and the prayers. In point of fact the chief work done by the Geonim with 1 Explicitly stated by Ibn Daud, in his rrapn 'D, and demonstrable from the Seder itself. * Comp. below, pp. 166-7. 3 For a hypothesis regarding the destination of Rabbi Hai's Seder see below, p. 175. 4 Rabbi Hai knew this very well, as is shown by his remark in Rabbi Isaiah di Trani the Elder, rnao, 42. Comp. also TT'C, II, 55, where Palestinian customs in Spain are mentioned. 122 THE GEONIM regard to the prayers was to guard the main, original prayers zealously against additions, and even so they were not wholly successful in warding off Palestinian influence 1 . Another current that threatened the stability of the order of prayers was Karaism, especially its feeble offshoots, which were close enough to Rabbinism to influence rather than repel it. The Responsuin by Rabbi Natronai, in the Seder Rab Amram, 37 b-^S a, is an interesting exemplification of Karaitic influence on the Rabbinical liturgy. The Haggadah fragment published in the J.Q. R., X, 42, with its Rabbinic and Karaitic elements, shows that this influence was so strong as to leave traces in literature. Spain and Egypt were the countries in which these 1 The many decisions of the Geonim, partly contradictory of one another, on the subject of insertions in the 'Amidah, especially on the New Year's Day and the Day of Atonement, reveal unmistakable traces of a long struggle against the Piyyut, ending finally in a compromise. In general, the investigator gains the impression that the Geonim of Sura were by far more kindly disposed toward the Piyyut than those of Pumbedita, of which a comparison between the Responsum of Rabbi Natronai in j"rt, 50, with one by Rabbi Hai in DTirn 'c, 252 (however, see 1. c., 288), affords a characteristic illustration. It is difficult to see how Weiss, 118, succeeds in discovering a predilection for Kalir in Rabbi Natronai from his Responsum. Rabbi Natronai (in j'n, 50) names two Piyyutim, nv^a yupa and mVira yaFrryi, with disapproval. The second is probably identical with ppna mbiru by Kalir in the 'Amidah for Purim in the German ritual ; and even the first, nvba yup, may be Kaliric, as Kalir seems to have written more than one Piyyut for the 'Amidah of Tisha' be-Ab. Comp. Landshut, rmiyrr mnr, s. n. As for the influence exercised by Pumbeditan tradition on Rabbi Jehudai (see above, p. 120, n. i), the fact is significant that he opposed any and every insertion in the 'Amidah, according to the information given in G. S., on p. 51. If the text of the c"to, Berakot, 34 a, and of "?n"ac, 27, is correct, the opposition to insertions extended even to yinc jn pi, which, however, can hardly be so ; it seems certain that it is an insertion made in Talmudic times. As for Egyptian conditions, it is to be noted that from rather early until comparatively recent times, both Palestinian and Babylonian synagogues flourished in Egypt, comp. J. Q. R., XVIII, n, 564 ; XIX, 460, Benjamin of Tudela, Itinerary, pp. 90-1, ed. Griinhut ; Neubauer-Cowley, Catalogue, 238, no. 16 ; and Poznanski, Z. H. B., X, 145. THE HALAKIC LITERATURE 123 currents were distinctly noticeable l , and they are the countries whence requests came to the Geonim regarding the order of the prayers. THE LITURGICAL PART OF THE SEDER RAB AMRAM. Exclusive of small sections of the prayer-book, the Seder Rab Amram is probably the first Order of Prayers issuing from the hand of a Gaon. His predecessor, Rabbi Natronai, sent to Spain a brief arrangement of the "hundred bene- dictions," published for the first time in G. S., p. 119 et seq. 2 It is possible, too, that the Gaon Kohen-Zedek, officiating shortly before Rabbi Natronai, put a Passover Haggadah together 3 . But of a complete Order of Prayers not a trace can be found until we reach Rab Amram. In its quality as the first Seder arranged by an acknow- ledged authority, Rab Amram' s enjoyed greater consideration than any work of the Geonic period. While of Rabbi Saadia's Seder only a few quotations were preserved, and they by specialists in liturgy, so that it was until recently considered a lost book, there is scarcely any work of importance belonging to the centuries between the years iooo 4 and 1500 that does not contain a reference to Rab 1 The remark by Rabbi Samuel ha-Nagid in avyn 'D, 267, throws an interesting light upon the masked Karaism infecting Spain during the Geonic time. The Gaon Rabbi Natronai learnt about 'Anan's book of laws from the Spanish Rabbi Eleazar Alluf, y*-c, 38 a. 3 Rabbi Natronai seems to have arranged also regulations for the readings from the Pentateuch ; comp. y"-\D, 29 a, and 3*n , ed. Hildesheimer, 623. s Comp. TT'C, II, 100, Marx, Uniersuchungen, &c., 5-6, and Muller in Handschrifttiche Jehudal Gaon zugewiesene Lehrsdise, 17, where may also be found the information obtained from Derenbourg, to which he refers in Mafteait, 83. Harkavy's view, in Saadia, 144, deserves to be mentioned as a curiosity of literature. He says that pis jrta and ncta 'i, in c'xrr, 1. c., are one and the same person, that is, Ibn Gajat is supposed to have called one person by two names in the same sentence ! The inn nc'NO 'CTO mentioned by Rabbi Saadia may perhaps be the maternal grandfather of Rabbi Sherira, 'isro (comp. above, p. 12, last line), of which 'cno is a variant form. 4 Rabbi Sherira, in "n'ac, in, is the oldest author who cites the z"~c. 124 THE GEONIM Amram's Seder. Though it was prepared for the Spanish Jews primarily, it was used as extensively by the Franco- German authorities as by the Hispano-Proven9al. From Rashi down to the anonymous fifteenth- century commen- tator 1 of the German prayer-book, published at Trino, 1525, the Franco-German scholars do not leave off appealing to the authority of Rab Amram. And the Hispano-Proven9al scholars of the same period, from Rabbi Isaac Ibn Gajat down to Abudraham, likewise form an unbroken chain of authors deriving their information from the Seder Rab Amram. Besides, it is probably the only Geonic work of which four complete MSS. 2 have been preserved. Of Rabbi Saadia's we have a single one, and that imperfect. This same circumstance, that Rab Amram's Seder was resorted to so zealously, carries with it a drawback. Due to it, we shall probably never know its true, original form. It was used until it was used up. To realise the whole extent of the problem thus forced upon us, we must remember that the Seder contains more than the prayers. They are accompanied by a continuous chain of important Halakot relating to the prayers. The introductory sentences of the Seder, the words of Rab Amram to Rabbi Isaac ben Simon, the addressee of the Seder Responsum, mention nothing about this Halakic exposition. His words are : " And relative to the prayers and benedictions for the whole year, concerning which thou didst make a request of me, it seemeth good to me to arrange them in order and send them to thee as they have been transmitted to us, the order of the Tannaim and Amoraiin." 1 The y"iD is quoted in the commentary on the Haggadah, with the words anas an 'nno nspa. Also in the brief observations preceding the prayers in tfxm nnno the Seder is quoted. It ceased to be quoted only after printed prayer-books became common. 2 On the MSS. com p. Marx, Untersuchungen zum Seder des Goon Rab Amram, Frankfort-on-the-Main, 1908, which reached me while this book was going through the press. In the following pages MS. S stands for the Sulzberger MS. in the Jewish Theological Seminary of America, and MS. O for the Oxford MS. THE HALAKIC LITERATURE 125 An argumentum ex silentio like this may not be pressed too hard. It is to be assumed that the Spanish congrega- tions did not ask the Gaon simply for a prayer-book. That they could have procured from any Babylonian Jew. They must have desired the valuable explanations and notes accompanying the prayers, and the Gaon, in his introduction, briefly spoke of the order of the prayers, which in his mind included the Halakot appertaining to them. Indeed, the probability is that the Spanish Jews laid more stress upon the Halakot than upon the prayers. On the whole, and certainly in all that was essential, the latter were settled everywhere according to local custom, which had too strong a hold upon the congregations to permit us to suppose for a moment that they would have given their peculiarities up for others, though the others had the high sanction of the Geonim. Furthermore, the quotations in the oldest authors that mention the Seder, Rabbi Sherira, Ibn Gajat, Rashi, and Albargeloni, are from the Halakic portions. This leaves no room for reasonable doubt that the Seder received its dual form from Rab Amram himself. The introductory words quoted above also show how untenable is the tradition reported by Azulai, in his Wa'ad la-Hakamim, s.v., which makes the Seder the work of the school of Rab Amram. This tradition probably originated in the fact that the name of Rab Amram is mentioned several times in the Halakic portions of the Seder, as are also decisions by authorities who lived after him, Rabbi Nahshon, Rabbi Zemah, Rabbi Nathan, and Rabbi Saadia 1 . If these decisions were the only alien elements in the Seder, we should wonder that a book so much used had come down to us in a comparatively unchanged form, rather than that it had received such additions. In fact, a critical examination of the Sedtr shows that it was abused to an extreme degree, and the 1 In MS. O Rabbenu Hai is also quoted. Comp. Marx, Untersvchungen, &c., ii. 126 THE GEONIM portion that suffered most is the Order of Prayers specifi- cally, rather than the Halakic explanations. In the following paragraphs proofs will be adduced and they might be increased tenfold to show that our present Seder Rob Amram has preserved a minimum of its original form, so far as the prayers themselves go. The concluding sentence of notw Tita in our Seder begins 3 |HN, while Abudraham 1 , 27, gives DW2n i?3 pan finx as the reading he finds in his copy, at the same time calling 'our form of it just quoted the custom of the " common people." The formula of minn nana, as it now appears in the Seder, assuredly did not originate with Rab Amram. As is shown by the Responsum by Rabbi Natronai, G. S., p. 116, line 3, the expression minn jnu was used in Babylonia, instead of the . , , notan of the Seder. Rabbi Natronai's wording is corroborated by 3"n, ed. Hildesheimer, 8. Rabbi Abraham ben Nathan states, in his Manhig, 9, that minn fnu was used at his time in Spain, while a century later, as we can see from Abudraham, 30, the form of the Franco-German Academies was in vogue, which is the form that agrees with our printed text of the Seder. The version used by Rabbi Aaron of Lunel showed still another deviation from the original Seder Rab Amram. It had mm nana piDyi>, instead of mm nan by, also to be ascribed to Franco-German influence 2 . The priestly blessing after minn nana can be traced back at least to the time of Rabbi Jacob, the author of the Tur ; he had it in his copy of the Seder. But the Responsum of Rabbi Natronai shows that it was not used in Babylonia. In the introductory note to the Responsum, in G. , p. no, it is demonstrated that it was a French custom, and, there- fore, is naturally missing in S and 0. 1 I quote from the edition Warsaw, 1877. 2 Comp. one, 41 c, where pcr> is denominated a Minhag of Lorraine, as compared with the custom prevailing in Spain. MS. S has correctly rrvm jm:. Comp. Marx, Untersuchungen, &c., 7. THE HALAKIC LITERATURE 127 Our text, 2 a, calls for the recital of the verses on the Sabbath sacrifices, while the Manhig, 9, indicates that the Seder provides for them also on the New Moon Day. Abudraham, 37, accuses the " common people " of having twisted niTDDI rnrQBQ, as correctly given in the Seder, into wcm vnaeo, but our text agrees with the wording used by the people. The nw in our text of the Seder forms the conclusion of the mo^n 'plDB, but we have a trustworthy tradition (D'nyn 'D), 249, that the recital of the rrvi? was unknown in the principal sj'nagogues in Babylonia as late as the time of Rabbi Natronai, the immediate predecessor of Rab Amram . From another source, R. & J., XXIII, 234, we learn that the first one to introduce the nTt? in Germany was Rabbi Moses ben Rabbi Kalonymos. All this would seem to point to the inevitable conclusion that the fiTt? in the Seder Rab Amram is not one of its original elements, a conclusion strengthened by the fact that, as is patent from the Manhig, 10 b, the m^e> did not appear in the copy of the Seder used by the author of the Manhig l . Indeed, the printed texts them- selves betray that we owe the nw to a copyist. On page 27 b, where the Sabbath prayers are recorded, the conclusion of the moTi V^DS is properly given as ... D95llOI. The omission of the passage Bnn -IIK at the end of the first Shema* Benediction cannot but be a correction made in accordance with the Seder of Rabbi Saadia. Rabbi Nahshon, the successor of Rabbi Ami-am, quotes this passage incidentally (:Tn, ed. Hildesheimer, 224), showing that he was not aware of any objection thereto, and it was recited in Babylonian synagogues still later, in the time of Rabbi Sherira (!>n*at!>, 13). There is even an explicit statement that Rabbi Saadia could not make his opinion prevail in Sura itself. This brings out an interesting point in the history of the liturgy. It may not be out of place to dwell 1 The MSS. have preserved the original text here only in part. See below, p. 144. Com p. also n*M, I, 6c, and MaJisor Romania, under rone* in the Sabbath Morning Prayer. 128 THE GEONIM upon it here. Originally the prayers connected with the Shema^ contained no reference to the future, the Messianic, redemption. Zion. the Temple, and the restoration of the house of David were prayed for only in the 'Amidah. Gradually the three benedictions preceding the 'Amidah were subjected to insertions dealing with the redemption. As we have seen, Rabbi Saadia protested, though vainly, against the presence of Knn "UK in the first Skema Bene- diction. His objection was that the Benediction in question was intended to be a prayer in praise of the majesty of God revealed in the sun and the light of day, and a prayer for redemption could not be attached to it fittingly. The Benediction following the Shema was originally a prayer of thanksgiving for deliverance from Egypt, and as is demonstrated in 0. S., p. 89, the insertions bearing upon the future redemption go back to the Geonic time, though they established themselves in opposition to Geonic authority, which was on the whole directed to the end of preserving the main, central prayers intact and unchanged. In this case, it seems their authority was here and there exercised unsuccessfully. The second of the Shema Benedictions, the ranx or D^iy rons, also contains a reference to the future redemption which must be very old, seeing that no echo of any opposition to it has come down to us. The old dispute about the opening words of the Ahabah has nothing to do with the insertion of a reference to the future redemption l . 1 The supposition put forward by Dr. Elbogen, Studitn zur Geschichte des judischen Gottesdienstes, 27, that the discussion on the opening words of the second Shema' Benediction actually turned upon the insertion of the Geullah, seems to me untenable. If his supposition were correct, what explanation could be offered for the fact that all the liturgies preserved until our time, the Ashkenazim, Sephardim, Italiani, Romania, all have the Geullah in this Benediction, though they differ as to the initial words. Furthermore, the Talmud itself, Berakot, ub, records a difference of opinion regarding the introductory words, but it is hardly possible that the insertion of the Geullah could go back to the Talmudic time. Dr. Elbogen considers it inconceivable that so petty a variation as between nn nan and cVw runs should have caused so much talk and THE HALAKIC LITERATURE 129 The fact that the shortened Yozer of the printed text is missing in the MSS. of the Seder, would by itself suggest the conjecture that it is derived from the Seder of Rabbi Saadia, even if the MS. of the latter did not contain it, and so make it a certainty. But the view that this Yozer, without a Keduahah, is the Palestinian, that is, the older form, is decidedly incorrect. The words of the Tosefta, Berakot, I, 9, 'P'p'p "paon Dy ruiy n"m, leave no room for doubt Yozer contained the Kedushah as early as the Tannaitic period, and the use of "paon in the Tosefta passage precludes the possibility of making the reference apply to the Keduskah of the 'Araidah l . "paon can only mean the recital of the Shema Benediction. The ''praying" of the 'Amidah is always called b^ancn. The reasons given by Dr. Elbogen (Studien zur Geschichte den judischen Gottesdienstes, 20) for supposing that the shorter form of the Yozer was the original form, are inadequate. He says that an analogous case is not known, of curtailing a prayer once used in a long form. In reality there are at least three parallel cases : wan, the shortened 'Amidah for private prayer, originating in the Tannaitic time ; the shortened 'Am&dah for the congregation, originating in the early Geonic time, known to us from the Eshkol (I, 55) by Rabbi Abraham ben Isaac, who quotes it from Geonic literature ; and the shortened grace after meals, which we have in three different forms, the one from the Talmudic time arranged for working men, and two later forms 2 for discussion. From the ancients he might have invited the reply p cs O3Q Kin. In their sight it was not a petty difference, not any more insignificant than the much-mooted question whether 1312 should or should not close with prsrr, about which we have varying opinions, beginning with the time of Rabbi Akiba (Berakot, III, 7), down to the last of the Amoraim (ibid., 50 a ; Yer. Berakot, VII, n c). 1 The correct interpretation of the Tosefta passage may be found with so early an authority as rp'iNi, Berakot, 8 a. 1 Besides the mspa o"na of the Polish Rabbis of the sixteenth century handed down by Rabbi Joel Sirkes, in win rva, on n*N, 192, there is a much older shortened form of the grace after meals in 'n 'mx, I, 36 d, by Rabbi Aaron of Lunel. I K 130 THE GEONIM various emergencies. The shortening of the Jiton n312 is particularly interesting, in view of the fact that the prayer was held to be Biblical, while all the others were based on Rabbinical authority only. The reason for the abridgment of the Yozer is plainly stated an individual may not recite the Kedushah. Dr. Elbogen maintains that this prohibition is a fiction pure and simple, based upon a misunderstood passage in the Talmud. Nevertheless, many of the Geonim, as well as most of the old authorities down to and including Mai- monides, were actually of the opinion that the reciting of the Kedushah by a single person was forbidden 1 , and from their point of view, whether correct or not, they were compelled to formulate an abridged Yozer. A dif- ference of opinion existed only regarding the extent to which it should be curtailed. Rabbi Saadia, following the lead of the Talmud on uj'an, retained only the frame- work of the Yozer, he omitted the numerous embellish- ments attached to it, while others of the Geonim left the Yozer itself as unabridged as possible, even when it was intended for private devotion, and omitted only the Kedushah 2 . I would venture a step further, and would assert that the Kedushah of the Yozer is the oldest form in which this prayer appears, the Kedushah in the 'Amidah being specifically Babylonian 3 . This would be the only 1 The views of the Creonim regarding this point are collocated by Dr. Biichler, in JR. K J., LIII, 220-30. Maimonides, it is alleged, changed his view ; comp. Caro, Bet Yosef, n", 59. The long discussions on this point in the old authorities leave the impression that the old view, based upon the Talmud exclusively, was opposed to the recital of the Yozer Kedushah by the individual, and the other view came into vogue only through DnciD 'co. 2 It should be borne in mind that in the olden times an individual absented himself from the lias nbcrt only if he had no time or if there was sickness, hence the aim to make the TIT nbcn as short as possible. 3 In the Midrash ha-Gadol, 1, 278, the following sentence is quoted from an unknown Midrash : ratzj 'V% raw 'Vto rwnp "> I'JN, that is, four Kedushot for each day, viz. : (i) isv 'ip ; (a) mrrcj to rrro 'ip : (3) 'ip ; and (4) nrr:n to rrroy '?, to which are added on the Sabbath THE HALAKIC LITERATURE 131 way of making clear why the Palestinians, as late as the year 800, continued to offer strenuous opposition to the ' Atnidah- Kedushah on week-days, which, as appears from G. S., p. 48 et seq., was forced upon them by the Babylonians. If it had been an old constituent part of the daily service, what other reason would suffice to explain the omission of the holiest part of the 'Amidah in Palestine? It is even questionable whether the ' Amidak-Keduskah was known to the Babylonian Talmud itself. Berakot, 21 b, is not decisive. All that may properly be inferred from this passage is that in Babylonia, and perhaps also in Palestine, the third Benediction of the 'Amidah con- tained the trisagion, though not necessarily as an inde- pendent paragraph, as we have it in our Kedushah, but as an integral part of the Benediction, somewhat like this : ^D N^D nisas *"* em? emp en? 3VD3 iB> Niui nn tpnp 1 11133 p"lNn, corresponding to the closing sentence of the third Benediction for nJBM BWI and 11S3, on which days, in view of their judicial character, the verse Isa. v. 16 is used instead of Isa. vi. 3. This would serve also to make clear Rabbi Huna's point of view. As the passage in Berakot informs us, he had no objection to an individual's reciting the 'Amidah-Kedushah in his private devotion. Rabbi Huna subscribed to the accepted principle : nB>np3B> 131 ^3 1>/ D mriQ3 NiT N!?, but he saw in the ' Amidah-fedushah only a part of the third Benediction, the DtJ>n nsjmp, in- tended for private as well as public worship. Furthermore, it should be taken into consideration that the MSS. and the old authors did not have nemp in this Talmud passage as in our text, but tJmp. Apparently, then, the the Huso/ Kedushah and the vmcn 'ip at the going out of the Sabbath. Accordingly, this Midrash did know the NTIDI 'ip for the Sabbath After- noon Service, which, as is shown in G. S., pp. 288-9, i s f Babylonian origin. The Targum Sheni, V, i, has an interesting passage bearing on the subject : pot rtn MOV baa snip .... !rr. At the time of this Targum, then, the NVIDT 'ip formed no part of the regular public service. 1 It is well known that the formula nn cnp was the old ocn ncnp, and not np nrw. K 2 132 THE GEONIM subject dealt with is not the Kedushah, but the words 'P'p BT7p in the third Benediction. The 'Amidah- Kedushah received sanction and character as an independent prayer only under the influence of the Babylonian mystics. The conception conveyed by it is the mystical idea that God receives his "crown" from Israel as from the heavenly host, when they adore him by means of the trisagion 1 . The old Kedushah contained nothing of this notion. It merely ascribed holiness to God in the words of the prophet Isaiah. It was against this mystical idea that the Palestinians during Geonic times contended inch by inch. First the Babylonians living in Palestine achieved their purpose of inserting the Kedushah in the Sabbath service, and this was far from being the only Paitanic addition made to it 2 . In the end, the Babylonian JFedushah slipped into the week-day service as well. In Geonic times the Babylonian Jews living in Palestine played pretty much the same part as the Polish Jews in Germany during the last three centuries. Fault was found with them on all sides, but after all they were "the scholars," and, do what one would, their authority compelled recognition. Now, as the ' Amidah-Kedushah is the product of the Babylonian mystics, so the Yozer- Kedushah goes back to the Palestinian mystics. Josephus (de hello Judaico, II, 8, 5) says of the Essenes : " They speak not a word about profane things before the rising of the sun. but they offer up the prayers they have received from their fathers facing the sun as if praying for its rising." Mutatis mutandis, a Yozer is nothing but the prayer at sunrise, and if the liturgy preserved for us had not had a Kedushah in the Yozer, we should logically have been compelled to assume its sometime existence there, 1 Comp. Bloch's essay on the nwra mv in Monatsschrift, XXXVII, 305. Our author goes too far when he assigns the origin of the Yozer-Kedushah to the Babylonian mystics. 2 Albargeloni, in OTiyn 'c, 251, expresses' 'his decided opposition to -|nv ton. Of course, his protest against this old insertion was vain. THE HALAKIC LITERATURE 133 In the whole of the prophetical literature there is nothing suitable for a Yozer except the glorification of the Lord by the celestial host, described by Isaiah, which we call the Kedushah 1 . Furthermore, the difference between the Palestinian and the Babylonian Kedushah calls for consideration. The Yozer-Kedushalt like the Palestinian 'Amidah-tfedushah has nothing of the "crowning of God," which is so dis- tinctly conveyed by the Babylonian 'Amidah-Kedushah. When the Palestinians, acting under compulsion by the Babylonians, accepted the ' Amidaft- Kedushah, they divested it of this mystical concept, and fitted it into the Yozer- Kedusfiah additional evidence for the independence of the two Kedueliot, for while the Babylonians know only the form with "1D3 for the 'Amidah-Kedushah, no trace of the " crown " can be discovered in the Yozer-KedusJmh, as, furthermore, the Palestinians have only Bnpi or IB'HpJ for the 'Amidah-Keduakah 2 . The above exposition can lead to but one conclusion, that the Yozer-Kedushah is pre-Geonic and Palestinian, and as a consequence the short Yozer in the Seder is exactly what it is said to be, an abridgment for private worship, and not the original Palestinian Yozer. It is nevertheless indisputable that the short Yozer is not properly to be accounted an original constituent of the Seder Rob Amrartt,. There can be no doubt that it was taken from the Seder 1 Rapoport, Biography of Kalir, note 20, gives so convincing a statement of the connexion between the Yozer and the Essenes that nothing remains to be added to his words. Dr. Hoffmann, in the Introduction to the cx:n cvro, goes so far as to conjecture that the Essenes were called c'D'in after mn " the sun," but this explanation of the expression C'C'in rfoo seems to me very forced. C'^nn would rather appear to be nothing more than a variation of D'rnn. Then c'D'in .-ran would be u " Collection of Proverbs." 2 Comp. G. S., pp. 48-9, where the vo formula is dealt with in detail. The statement made there that the Italian ritual, before being influenced by the Kabbalah, knew only vo, is corroborated by the words in bn'ic, 13 : 1.12 -raTJjmnjnrrcac. Comp. also Berliner Hoffmann, Magazin, Hebrew supplement 31:2 ISIN, 1886, p. 1 1 , where vo is given as the Kedushah, -TS 3n:c. 134 THE of Rabbi Saadia. Not only is it missing in the MSS, of the Seder Rob Amram, but we know from Bondi, Siddur des Rabbi Saadia, 13, that this short Yozer i& actually in the MS. of the Seder of Rabbi Saadia l . Whether the formula nil ranx for the second Shema* Benediction is really traceable to Rab Amram, is question- able, for as late as the time of Rabbi Sherira and Rabbi Hai it began with D^y nanx everywhere in Babylonia except in the synagogue of Kohen-Zedek, and there is no likeli- hood that Rabbi Amram would have given a decision deviating from the universal Babylonian custom. It seems that we have again met with a ! correction " made for the purpose of bringing the Seder into agreement with the views of the Franco-German authorities 2 . The addition of tan TTi for the summer is mentioned by Rabbi Abraham, in the Manhig, 16, as a Provenfal custom, not known to the Seder Rab Amram ; yet in our text of the Seder it is given 3 . Abudraham, 67, speaks with disapprobation of the " common people " who say N^y *t&J&1 in the Ninon N^nnp, the only correct form being IPDfa? D^yn, as the Seder Rab Amram has it. Again our text agrees with the supposed preference of the common people. The addition to the Geullah in the Evening Service in our text of the Seder, 1 9 a, is most suggestive. Rab Amram (6 b) is peremptory in opposing the insertion of the idea of the future redemption in the Geullah of the Morning Service. It is absolutely inconceivable that he would have 1 From V?r, I, 52, it may even be gathered that the short Yozer in the 5*iD read other than in our text. a It is true, so early an authority as the Gaon Rabbi Hanina, the disciple of Rabbi Jehudai, expressed himself in favour of nan runx ; comp. a"n, 125. But the statement . . . -|^i jan i:arm is contradicted by Rabbi Sherira. It may be that the Minhag was changed in the later time of which Rab Sherira speaks. 3 Accordingly, Rapoport (Kalir, note 33) is not right when he says that Kalir and the Sephardim agree in having Va for the summer, as the old Sephardic ritual did not have it. THE HALAKIC LITERATURE 135 been so inconsistent as to permit its insertion in the Evening Service. Moreover, from the Responsum by Sar Shalom given in G. S., p. 91, it appears that the insertion originally had its place in the Geullah for the morning. It is there- fore probable that it occupied this place in the copy upon which our text is based, as, indeed, the amplified Geullah was most generally identified with the Morning Service *. But the copyist of the Seder could not stultify himself to the extent of giving the expanded Geullah side by side with the Gaon's disapproval of it. Hence the insertion disappeared from the Morning Prayer, while, in the Evening Prayer, there being no remark of Rab Amram's to deter them, the copyists followed the custom with which they were familiar in the Geullah for the evening. Now, as neither the Sephardim nor the Ashkenazim in later times had an amplified evening Geullah, the inference is that the model for our text of the Seder must have been an old Spanish prayer-book containing these additions. As for their origin, the Genizah fragment enables us to say with certainty that they came from Palestine, whence they reached also the Morning Service in the old Orders of Prayer of the Ashkenazim and Sephardim, from which the opposition of the Geonim did not succeed in removing them entirely. Hence the fact that the insertion in the Geullah is missing in the Sulzberger MS. of the Seder proves nothing with regard to its high antiquity as com- pared with the printed text. It belongs to a time in which the amplified Geullah was no longer a general custom, and the copyists of the Seder therefore had no occasion to put it into their copies. For the endeavour to arrive at a valuation of our text, the noon hy B^p, i9b, is of great importance. In the Genizah fragment published by Professor Schechter in the J. Q. R., X, 655, there is a Shema* Benediction before 1313, running thus : PQJ31 D^B> 33b la^cnb l"3pK n*DN ^N3 nvsn. Recently, another Genizah fragment was reproduced 1 Conap. the Genizah fragment in R. E. J., LIU, 236. 136 THE GEONIM in the R.J&.J., LIII, 240-1, by Professor LeVi, and it con- tains a Benediction with almost absolutely the same wording. The accepted opinion is that this Benediction was unknown hitherto, until the publication of these two fragments. No explanation came readily to hand when and why this special Shema' Benediction was added to the other two of Tannaitic origin. Another striking point is that this Benediction is not directly before the Shema' in the two Genizah fragments, but before 13 "13. Does it seem reasonable to suppose that a Shema Benediction was recited before 1313? Light is thrown upon the bearing of this Benediction by a Kesponsum of Rabbenu Hai's, and by the remarks of a number of the old authorities about the Shema 1 Bene- diction before bedtime. Rabbi Hai, T\"w, 57, decides against the use of nanta i3^ni>i yov nnp by i"3pK n"DK '"3 before the noon by w"\>. Thus it appears that the Shema' Bene- diction of the two fragments contains nothing new. It is merely a variant of Rabbi Hai's form, a form to be found also in D^n 'ms, I, 430, Abudraham, 23, and 'Ittur, II, 34 c 1 . Its import is conveyed to us in an observation made by Rabbi Asher ben Yehiel, on the beginning of Berakot, which is repeated by his son Rabbi Jacob, in Tur, Orah Hayyim, 235. According to a well-known custom 2 the Evening Prayer was said at the synagogue immediately following upon the Afternoon Prayer, even if night had not yet set in. This necessitated the repetition of the Shema' after nightfall. As the Rabbinical injunction requires its recital at night, the authorities insisted upon its being said before going to bed, even if it had been prayed at the synagogue in the Evening Service. Some 1 Comp. also W;ir, 40, and Tosafot, on Berakot, 2 a, catchword end, and Hullin, 105 a, bottom. 2 This custom must have arisen in Palestine and spread thence to the European countries, but it gained no foothold in Babylonia, on account of the opposition of the Geonim. Comp. Rabbi Hai's Responsum in fi, 78: and n*r, 76 ; quoted also by many old authorities. THE HALAKIC LITERATURE 137 ordered, that with the Shema 1 the two Benedictions also were to be repeated 1 , for the reason that they, too, had been recited in the synagogue before nightfall. Rabbi Amram, however, says Rabbi Asher, was of opinion that it was not obligatory to say over again the Shema' Bene- dictions in their full wording. A brief Benediction. according to the usual formula of the rnaia, sufficed. There can hardly be a doubt but that Rabbi Asher found this view of Rabbi Amram's in his Seder under noon by t?"p. In our text it is missing, in consonance with the opinion of the later authorities 2 , who permitted neither this nor any other Benediction in connexion with the ntson by t/'p. There is only one MS. of the Seder in which the ab- breviated Benediction appears, the Oxford MS. Even there, however, it seems probable to me that the passage Dllpi '01 nnxnp was not derived from the Seder, but from some other source. My reason is that as it now reads in the Oxford MS., it contains a contradictio in adjecto. If stress is laid upon the recital of a Benediction before the Shema' at bedtime, and if stress is laid upon it for the reason that the Evening Service is held before nightfall, the appointed time for the Shema , then it would follow that the whole Shema should be repeated, not merely the first Parashah, as our text and the Oxford MS. provide 3 . It is also worthy of note that the passage in question is not in its proper place in the Oxford MS. It should have read bapb "p3Oi DK rr JD miptn rims Niipi rrobt? nw rnabo wby patsi yiDB>. The original Shema 1 Benediction before noon by K^p, which was nothing but an equivalent for the two long Benedictions which accompanied the Shema 1 when it was recited before nightfall, was looked upon later as a special Benediction 4 for noon by t/'p, without reference to the time of saying the Shema' in the 1 Comp. rev '-\ 'Yobn, Berakot, beg., and Caro, Bet Yos?f, n*s, 335. 3 Comp. Tosafot, Berakot, beg., and Albargeloni, quoted in *n*ac, 40. 3 Comp. Rashi and Tosafot, Berakot, beg. 4 Thence the opposition of Rabbi Hai to this Benediction ; he says. in men ITOD to c'ptn r:EO 138 THE GEONIM Evening Service, whether after or before nightfall. This is the conception that finds expression in the Oxford MS., as it does in later ritualists, and it is a conception that is not wholly in accord with Rab Amram's view. This analysis enables us to understand the Shema* Bene- diction in the Genizah fragments. A substitute for the prescribed Shema' Benedictions in the evening was a common expedient in congregations where the Evening Service took place in the synagogues before nightfall, as was the case outside of Babylonia 1 . But there were cir- cumstances requiring an alternative Benediction even in the Morning Service, either for an individual who had time only for the Shema', but not for the whole Morning Prayer, or for the whole congregation on fast days and holidays, on which the elaborate service was so long drawn out that the Shema' might fall beyond the proper time 2 . In such cases, and similar ones, Shema was recited in private devotion before the regular service, with the short Bene- diction in the Genizah fragments. For this reason it is not given as a Shema' Benediction after nn ronx or ronx D^iy, but as coming before wni, because only an individual, and he only if he does not recite jflDt? ni3"in, is to recite the short Benediction. It is, in fine, a special Benediction, which really should have no place in a regular Order of Prayers. It is highly probable that the introduction of Shema 1 with the three words fONJ "jta bx is only a remnant of this very Shema 1 Benediction. The opposition to it must have been strong enough to force out niatal DP, which was replaced by ita ta. Accordingly, the complete introduction must have run thus at some time after nia^oi DP was omitted : nsan s?BXM D^>P 33^3 ^taa few i?n ^N, and all that remained of it were the first three words. 1 Comp. above, p. 137, n. 2. 2 Comp. Yer. Berakot, 1, 3 c ; the Geonic Responsum quoted in Albarge- loni, D'nrn 'c, 255 ; 3>^D, 3 a, and n*N, I, 6 c. THE HALAKIC LITERATURE 139 An old addition, derived from the Sephardic prayer- book, is the congregational prayer 1K> rane", 27 b. So early an authority as Albargeloni had it in his copy of the Seder Rob Amram, as he tells us in DTi^n 'D, 250, while Tur, Orah Hayyim, 57, reports the reverse about his copy. That it was missing in the model for our text is evident from the fact that it does not appear in the Week-day Service, though there is no reason for reciting this prayer on the Sabbath exclusively 1 . The order of the verses '131 inp . . . pro inpnv is stamped as incorrect by Abudraham, and he recommends that they be recited as they succeed each other in the Seder Rab Amram. But our text has the repudiated arrangement, except in the New Year Service, where the order is that recommended by Abudraham. What Rabbi Abraham ben Nathan says in his Manhig, 33 a, makes it plain that in his copy of the Seder the Talmud sections are not set down to be recited at the end of the Afternoon Sabbath Service, and the passage KBIT *P"S is properly enough found to be missing in the Oxford and the Sulzberger MSS. The formula for pin nns at the end of the Sabbath, as given in our text, differs from that quoted in the Manhig, 33 b, from the Seder. As Maimonides agrees with the Manhig, it remains only to assume that our text was shortened in this passage. The prayer , . . ^nn X"IK, on page 31 b, is known to the Manhig only as a Spanish custom, and to justify it the author resorts, not to the Seder, but to a Yerushalmi passage, and we may be sure that it did not occur in his copy of the Seder. This throws doubt upon the authen- ticity of the whole section, from JVJHB>K until nyiB^, all the more as it is missing in the Oxford MS. That it is, in spite of this, an addition of respectable age may be inferred Albargeloni, it is true, deals with the Sabbath Service, but it is fair to assume that he had TOC name' of the Week-day Service also before him. The editor of the own 'D observes that it is not contained in our y"-\c ! 140 THE GEONIM from its being quoted from the Seder Rob Amram by Ibn Gajat, v"V, I, 15, as the Tur, Orah Hayyim, 299, does also. Nevertheless, it is recognised as an interpolation by the circumstance that it is a piece put in between the Habdalah and the draining of the Habdalah cup. It does not seem likely that between the Benediction over the wine and the drinking of the wine itself so long an interval would be interposed as is required for the recital of this piece, the rule being that a Benediction is to be followed at once by the enjoyment of the food and drink over which it is said. It is therefore much simpler to assume that it was taken from some other source, and as the copyist could not well attach it to the Halakic portion of the Seder, there was nothing for him to do but join it to the Habdalah. On 41 b, in the Order of Prayers for the second day of the Passover, the counting of the *0mer is missing. Yet it was present in the copy of the Seder used by B-abbi Aaron of Lunel, as appears from a remark of his in Dims Dn, I, 84 a. As an adjustment in conformity with the Sephardic rite, we may consider jJDI Ti 7X in the first Benediction of the 'Amidah for the New Year, which Abudraharn attributes to the ignorance of the people. He accuses them of having changed this Benediction as given in the Seder Rab Arnra/m,. Our text again agrees with the custom of the ignoramuses. If we call to mind how zealous the Geonim were in denouncing any change in the 'Amidah, there can be no doubt as to the correctness of Abudraham's version of the Seder in comparison with our text. Another change in the 'Amidah for the New Year is the insertion of trip tnpo 21B DV. Of the Seder Rab Amram it did not form a part, for which we have the clear testimony of the author of the Manhig, 52-3. It is a peculiarity of the Spanish liturgy, and our text was here subjected to an importation from it. The remark made by Ibn Gajat on the changes in the 'Amidah prescribed by Rab Amram for the Ten Peni- THE HALAKIC LITERATURE 14! tential Days (E^t?, I, 45) proves the sentences D"r DIM, 45 a, to be an addition from the Spanish Mahzor. The Benediction over the Shofar, in the copy of the Seder Rob Amram used by Ibn Gajat, read jnpn^ (v"&, I, 261), while our text offers the formula prescribed by Rabbenu Hai. On the other hand, the ivpn 7tfy0 in his copy of the Seder had the words KM rjR3 yiprb "pa OKI. One must despair of establishing the wording of this Benediction original to the Seder. The prayer n^rr.K is properly missing in the printed text, 27, while the MSS. Oxford and Sulzberger contain it as an addition from the Sephardic Mahzor. It is, doubtless, of Palestinian origin, as can be seen from the Mahzor Romania, where it has a place in the Daily Prayer. Besides, the closing Benediction *^pn Ttt> ''"Nil is known to us to have been used as such through a passage in the Yerushalmi Berakot, I l , and accordingly belongs to the rem- nants of the Palestinian liturgy, which have been preserved among the Sephardim, Ashkenazim, and Italiani. The words npnpn wi^no VW 2 , on 47 a, make it seem 1 Not in our text of the Yerushalmi, but in the text used by the old authors. Comp. Ratner, D'tem p'S run**, 33-4. Maimonides also has this formula, as well as Rabbi Saadia, in his 'Abodah given by Dr. Elbogen in his Studien, &c., 122. Curiously enough, Dr. Elbogen overlooked this, on p. 70, n. i. 2 Weiss (IV, 49) reproaches the Geonim for calling the Academy nmpn n:w. However, it is not the Geonirn who use the expression, but the scholars outside of Babylonia (R. Ibn Abitur and Mosea ben Enoch, in y*\r, 4 d, 29 ; 30 a, 9) or the correspondents of the Geonim (>":, 9). In the latter passage, the question contains the words : rrempn nya*2 an:orr TW, while the Gaon's reply was the simple i:3n2?3. Likewise in V^air, 172, mmpn unjTraa is a remark made by the compiler of the Responsa. In general, the Geonim either cite decisions by other Geonim or the custom of the Academy, but never a decision of the Academy, which, indeed, would have been odd coming from a Gaon, as all decisions were supposed to be issued by him and not by the Academy. In 01*123, 44, rronpn nruvron in Rabbi Sherira's reply is only a verbatim repetition of the expression employed by the questioner. It is interesting that in the Re- sponsum by Rabbi Sherira and Rabbi Hai jointly, found in the Responsa Collection of Rabbi Solomon Ibu Adret, V, 25, a-b, n. 121, the question contains the expression rroipn nrc'n, while the answer has instead of it sto rrrrr. 142 THE GEONIM very likely that mJ ^3 was missing in the original Seder, for these words were never used by the Geonim. If, besides, we take into consideration that m: ^D was un- known in Babylonia, as -we are told by the Geonim of Sura and of Pumbedita without a dissenting voice 1 , the probability of its not having formed a constituent part of Rab Amram's Seder rises almost to certainty. There would be no explanation to offer for Rab Amram's pro- cedure in first putting it into his Seder, and then character- ising it as a " foolish custom." We probably are troubled by two additions derived from different sources. The first addition, the ma ^3 itself, came, in all likelihood, from the Seder of Rabbi Saadia, and to this was joined, as a second addition, the disparaging criticism upon it made by Rabbi Natronai, introduced by the words xmTiDD nJB* nefipn. To the Spanish Mahzor, again, the prayers Kl^l nTjp and "p!>o, on 48 a, are attributable. As we learn from explicit statements in Ibn Gajat, w"v, I, 61, and Manhig, 60, it was Rab Amram's opinion that these prayers were not to be said on omaan DV. The author of the Manhig, and Abu- draham as well (133), add that none but the Spanish rite differs from Rab Amram. This point affords a striking illustration of the heedless way in which the copyist to whom we owe our text set aside the real Seder of Rab Amram. On 47 a, where a list of the initial words of the prayers for a"anv is given, he followed his model implicitly. There he included neither r6y nor "pta. But two pages further on he could not refrain from setting down what he was accustomed to connect with the services of the day. Our text contains no alphabetical NDn *?y, yet Abudraham, 153, cites one from the Seder Rab Amram. The prayer for a mother on the day when the child to 1 Comp. tD"ir, I, 60-1. Rabbi Saadia is the only one who knows Kol-Nidre, whence it follows that it was of Palestinian origin, as the Seder of Rabbi Saadia follows the Palestinian customs closely ; comp. below, pp. 166-7. Concerning the opposition of the Geonim to cm: mm, comp. above, p. 96, n. r. THE HALAKIC LITERATURE 143 which she has given birth is circumcised, 52 b, is a later addition, as was proved by the present writer in the Z. H. B. t IX, 1 06. The Geonic sources mention a prayer for the child, but none for the mother. The Benediction to be said at the circumcision of a proselyte, and of a slave, as set down in the Manhig, 98 b, from the Seder, is not in agreement with our text, which should probably be modified according to the Manhig. In view of all the passages instanced, it would be a wilful perversion of judgment to make an inference regard- ing the nature of the Geonic liturgy from the recensions of the Seder at present available. Our printed text cannot be looked upon as anything more or less than a Spanish Order of Prayer with some additions from the real Seder Rab Amram. The same characterisation applies also to the MSS. Sulzberger and Oxford 1 , though they deviate here and there from the printed text. Of the two MSS. the Oxford apparently is a more recent version, the in- sertions in which may have been taken from the Seder of Rabbi Saadia. This supposition is strengthened by the long passages, given by Marx, Unter&uchungen, &c., Hebrew part, 4, 6, 18, which are said expressly to have been derived from Rabbi Saadia, and p. n, relative to rim, which is quoted by various authors with the name of Rabbi Saadia attached to it 2 . The grace after meals in the MSS., having the same wording in the two, is also more recent than the printed text of the prayer, as is shown by *?2ib pron rp-O given at the end of the Seder. The prayer after pin pm in the Oxford MS. is doubtless a later addition 3 . Rab Amram would scarcely have sent the Spanish congregations more than the main prayers. Hence the difference between the forms of the nunn in 1 I have given the prayers in them only a cursory examination, but I am convinced they agree with the printed text in all essentials. 2 Comp. Miiller in (Ewares Complets de R. Soodt'a, IX, 156. s The sentence (28) ib inno .... ^xVo occurs almost literally in an epitaph at Brindisi, of the year 833, published by Ascoli, Inscriziotie, 66. Comp. also n*w, II, 635. 144 THE GEONIM the printed text and the MSS., as none of them were contained in the original Seder Rob Amram. There is, of course, no need to lose time in adducing proofs that the addition to Nishmat in MS. Oxford (24) is a late insertion, nor that the extracts from the Hekalot, to be found only in the printed text, most probably were not of the original constituent parts of the Seder. It is sig- nificant that while the Oxford MS. has no nw in the Week-day Service (p, 3), it has it in the Sabbath Service (13), exactly the reverse of what we find in the printed text. As has been demonstrated, Rab Amram did not have the PITS? in his Seder. THE HALAKIC PART OF THE SEDER RAB AMRAM. It now behoves us to explain how it happened that of all old works the Seder Rab Amram was subjected to such peculiar treatment. Like the others it suffered additions to its essential, original form. But that is not all the essential original form itself was not left intact, it was so modified, abridged, and extended, that we now have very little of what it was in the first place, when it left the hands of Rab Amram. Prayer-book making among Jews is a wholly modern trade. Rab Amram did not, by any manner of means, write a prayer-book. He merely sent the Spanish congregations the prayers prayed in Babylonia, well knowing that, to use a Talmudic phrase, " every stream has its own current." He had no intention of forcing Babylonian rites upon Spanish congregations. Incorrect readings, which had crept into some of the prayers in the course of the centuries, were rectified in the Halakic notes accompanying them, and at the same time the notes served to state the principles which had guided the Tannaim and Amoraim in settling the form of the prayers, and which still were to be applied as norms. These explanations of the Gaon subjoined to the prayers were the important part of Rab Amram's Responsum for THE HALAKIC LITERATURE 145 the Spanish Jews. There was no disposition on the part of the latter to abolish their local rites, but when the congregations had differing customs, or in doubtful cases, the directions of the Gaon were resorted to, consulted, and applied. The main task of the copyists, employed by those interested in spreading the Seder, consisted not in reproducing the prayers, but in recording the Halakic directions and the important variations from their prayers given by Rab Amram. In this way we have come into possession of Spanish prayer-books embellished with ad- ditions from Rab Amram's Seder as well as his Halakic instructions. Similarly, the Germans had their 3YT DniTHD D"iy l , prayer-books embodying their liturgy together with the Halakic portions of Rab Amram's Seder 2 . Of the same class is the Mahzor Vitry, which contains the major part of the Halakic element of the Seder, but in the prayers themselves it follows the French ritual. In view of the close relation subsisting between some of the prayers and the Halakot accompanying them, it may be assumed, without further evidence, that the Spanish congregations, and here and there others as well, yielded to the great authority of Rab Amram, and made changes in their liturgy in consonance with his directions, such as the excision of the reference to the Messianic redemption from the Geullah, which, as was demonstrated above, existed in the old Spanish forms of the prayer. Occasionally, compromises must have been made between the local custom and the version recommended by the Gaon. When we find the Sephardim using ina for the Musaf Kedushah, and je'npJ for the Kedushah of nnnsr, it is fair to conclude 1 fin, I, 26 b. 2 Naturally, many a Halakah was given a place in the Mahsorim that had the sanction neither of Rab Amram's name nor any other Gaon's. Hence, quotations from the y*SD in the works of the German authors that cannot be traced. For instance, a contemporary of Rashi's grandsons ('Vi s'n 'on, 3) cites the nbn nofci of Rab Amram, of which not a trace can be found in the r"So , and probably it never existed there. I L 146 THE GEONIM that we have an instance of an attempt at amalgamating different rituals *. The influence of the Babylonian ritual must, therefore, have been strongest in Spain, whither the Seder was first taken, which, however, did not hinder it from asserting itself among the Franco-German Jews. In pursuing this line, it must be borne in mind that frequently what is denominated the custom according to the Seder Rob Amram is nothing but the old Spanish rite, which agrees with the old Ashkenazic rite, both derived from Palestine 2 . While the liturgical part of the Seder was badly abused by the copyists, the Halakic part has reached us in com- paratively good condition. After what has been said, the reason is patent. The prayers the copyists knew by heart, and they paid little attention to their model. They wrote as their memory dictated. Besides, they knew that the value of their work was concentrated mainly in the copying of the Halakot. To these they therefore devoted conscientious care. It was inevitable, of course, that in spite of all attention, even this portion of his Seder should receive additions from other hands than Rab Amram's, and, also, several Responsa by him, which he seems to have addressed to Spanish scholars independently of the Seder, have been inserted at suitable places. For instance, the Halakot on pp. 26 a-b, bearing the name of Rab Amram, are abstracts 1 Comp. G. S., p. 49. a The great respect enjoyed by the y"-iD among the Franco-German Jews is apparent from the words of Rabbenu Tarn, in ixrrn 'r, ed. Rosenthal, 99, in which he maintains that the Seder was the chief source for the prayers. Rapoport, jn: 'l 'n , note 29, goes too far, however, when he says that the Germans were the only ones to accept the Seder Rab Amram, excluding the Spaniards as he does. Yet his instinct was correct in laying stress upon the influence exercised by the y*"iD upon the German liturgy. In his polemic against Rapoport, Weiss, Dor, IV, 121-2, is less close to the truth when he deduces the dependence of the Sephardic ritual upon the y'no, from the agreement between the former and our text of the Seder. We have seen that the relation is exactly the reverse. Furthermore, Weiss is mistaken in calling Maimonides' Seder Sephardic it is Egypto-Palestinian. THE HALAKIC LITERATURE 147 of Responsa of his addressed to the congregation of Barce- lona, b*3, 56-7. Ibn Gajat, e^t?, 1, 10, and Rabbi Abraham of Lunel, Manhig, 26 a, quote these passages, but it is doubtful whether they knew them from the Responsa as such, or from the Seder 1 . The sentence Drroa 1^ t 37 b, did not occur in the Seder used by the author of the Manhig (43 a) ; it is obviously a gloss calling attention to a Responsum of Rab Amram's, which gives support to this peculiar custom by means of the authority of the Yeshibot and the Geonim 2 authentication which was all the more pertinent, as not only did the European Jews know nothing of the recital of rniri>D on Purvm, but also the Geonim themselves were far from unanimity upon the point, as appears from TUT, Orah Hayyim, 693. It would seem that the custom prevailed only in Sura ; in Pumbedita no nin^D were prayed on Purim. So far as Sura is concerned, the testimony of Rab Amram is reinforced by the fact that Rabbi Saadia himself composed nin^D for Purim, published by Professor Schechter, Saadyana, 49-50. There is the possibility, of course, that these rnn^D may have been intended for ivayn TTIDK. The sentence nU'B" a"yx, on p. 32 a, is taken from a Responsum by Rab Amram, quoted in full by the author of the hi*??, IO2 3 . Additions from the Seder of Rabbi Saadia occur in three places in the printed text of the Seder Rab Amram, 4 b (bis), and 52 a. But, as was observed above, the MSS. do not contain the first two insertions, and as for the third, we know that it did not appear in the copy used by Ibn Gajat, as can be inferred from his words at the end of &"v, I. It seems to have been taken from the D M n mmx , 26 c, which cites the opinion of Rabbi Saadia in opposition to Rab Amram's. 1 In "jn'zc, 54, it was doubtless taken from a Responsum, and not from the Seder. * By a slip the author of bn'ac, 157, writes poioi p:n! For the meaning to be attached to pen in this sentence of Rab Amram's, see above, p. 24, n. i. On p. 29 of the 'jn'niD it has, properly, pen without pom**. 3 Comp. also Hazan, ;rn n , 45 a. L 2 148 THE GEONIM If the superscription (i4b) Nmaoia nX m is correct, then, naturally, we are dealing with an addition, as it is very improbable that Rabbi Zemah could have been quoted by Rab Amram. But one cannot help being assailed by doubts as to the correctness of the superscription. It is not impossible that the abbreviation 'x 'n, standing for pnx 31, was improperly interpreted as nox 'i, and then, to complete the verisimilitude, NJrnnoia N'n was added after ra^, as Rabbi Zemah was Gaon at Pumbedita. Originally, it must have read 3py pro mB* wn V""! 1 , without specifying the Academy. As was demonstrated at length above, only the heads of the Sura Academy bore the title Gaon. At first, and even later, when the heads of the Pumbedita Academy were already called Geonim, a distinction was still made between the 3py pJ ro'B* B'NI, the head of Sura, and the Gaon of Pumbedita, who were only r6tt be> mwn PNI. As early as Talmudic times (Rosh ha-Shanah, 23 b), r6i3 was synonymous with Pumbe- dita 2 . Later copyists, especially those in countries remote pb, ga, has the reading rras pns', plainly traceable to the abbreviation s"~\, for which the copyists had two explanations, prnr '-\ and nos 'i . That NTVIMIB NTT m'tra is a later addition is confirmed by TIDTZJN, I, 33, where it does not appear. The names pm" and pns are often confounded. Comp., for instance, Mekilta, Jethro, I, and Sifre, Deut., 38. In both places pis is to be read instead of pn^% as appears from Kiddushin, 32 a. The name of the Gaon Zadok is misread for pni" in z"n, 56, n"n, II, 414, SrVar, an, and in many other places. Comp. also Zunz, Gesammelte Schriften, IV, 274. MS. reads njnra p* rros n. a What Maimonides (Commentary on Bekorot, IV, 4) has to say on the use of these two titles at his own time is particularly interesting. He informs us that while ipy p: ramr trsi was used in Palestine, the Babylonians bore the title rrtu to nro' wi. The reason for the differing practices is obvious. In Palestine they tried to perpetuate the original title of the Gaon, while in Babylonia the title of the head of Pumbedita was continued, as this Academy survived that of Sura by two generations. The Hebrew text of Maimonides is corrupt. It reads nron pN instead of ?23. The Ai-abic text published by LOwenstein, Berlin, 1897, p. 22, has the correct reading psiy'jM, and the same is to be found in the MS. of the Arabic text of the Maimonides commentary in the Jewish Theological Seminary of America. THE HALAKIC LITERATURE 149 from Babylonia, did not distinguish the Geonim from each other by their exact titles. The mistake having been made of reading x'n as no* 'n , the expression 3py> pw r\yw BVI was retained, while the words NfmoiQ NTI were added : they bear plainly the earmarks of an explanatory gloss. The same error of interpreting an abbreviation incor- rectly may have changed ^antM 'n, 4b, into |lB>m '"i 1 . It is a vexed question, the identity of the Rabbi Nathan mentioned three times in the Seder, 35 b, and 37 a (bis). In the last two places he is called m^ e>N"i, both in the printed text and in the MSS., which gives no encourage- ment to his identification with the uncle of Rabbi Sherira. Rabbi Nathan ben Rabbi Judah. The latter was no e>Nl r\yw, only an pta, and if the copyists had desired to confer a more distinguished title upon him, they would have called him Gaon, the usual epithet bestowed later upon a very prominent scholar. But there was no Rabbi Nathan who was a ra^ B>N"i in Babylonia, and we have the choice of again resorting to a falsely interpreted abbreviation, and putting }DJ for 'NJnBJ 2 , or identifying him with the contemporary of Rab Amram, the Rabbi Nathan of Kair- wan, who was a n^B* t?N"i in Kairwan 3 . The difficulty of identifying this Rabbi Nathan is increased by the fact that Abudraham, even in his first edition (Lisbon, 1489), twice has foro 'i in citing the Seder. In the first passage, p. 79, jorti is probably a mistake for Amram, while in the second, p. no, corresponding to 37 a of our text of the Seder, the dictum ascribed in the latter to Rabbi Nathan, is quoted in the name of Rabbi Nahman. But fru would seem to go 1 Comp., R. E. J., LIV, 204, where this passage of the j?*-c is quoted, but without the name of Rabbi Nahshon. There is no reason for doubting that it is taken from the Seder. * An interesting example of mistaking 3*S = 'JCITC: 'i and :"} = |ro 'T for each other is afforded in Tur, Orah Hayyim, 190. It occurs in the first Soncino edition, and in all following editions, while ed. Mantua, 1475, has 'terra: S as is proper, and as is confirmed by o'rt, 187 ; for indirect testimony by Rashi see above, p. 43, line 6 from below. s Comp. above, pp. 31-2. 150 THE GEONIM back to }1BTU, rather than to |nJ. The name of the Gaon is elsewhere found corrupted into pro l . Thus the reading jr>3 becomes very doubtful. Besides, the decision given on 37 a in the name of Rabbi Nathan offers a difficulty in the subject-matter. It contradicts a usage prevailing in the Yeslnl)ot y if we can put implicit confidence in the words of Rabbi Natronai, a^n, 187. The last point may be adduced in support of the assumption that the authority referred to is Rabbi Nathan of Kairwan, who recorded his opinion here at variance with that of the Babylonian Geonim. Apart from these additions, which can be attributed to definite authors, there probably are a number of anonymous passages in the Halakic part of the Seder that did not belong to it originally, but were inserted in the course of time. For instance, it is not at all likely that the references to the Spanish ritual, i a and 2 a, were made by the hand of Rab Amram himself 2 . The expression nxo "p rvniEW ni^NB>2 in the latter place is not a Gaon's way of speaking. 1 Comp. Rapoport's Introduction to p"j, gb, and also 531*03, 47, where, likewise, ptcn; is to be read instead of jnn:. The first edition of Abu- draham reads p: instead of jtsna in rv:rn 'n, 135, in agreement with y"-\~, 35 b, while all the subsequent editions have pn: \ Schorr, He-Halus, VII, 144-5, insists that there was a Gaon by the name of Jon:, though none is mentioned by Rabbi Sherira in his Letter. By way of corroboration, he adduces the fact that Rabbi Sherira has no reference to the Gaon Rabbi Menahem, of whose existence there can be no doubt. Schorr evidently was carried away by his opposition to Rapoport. In point of fact, the Gaon Rabbi Menahem is mentioned by Sherira. rr"on: is out of the ques- tion, the only Gaon by that name, the son of Kohen-Zedek, not having written any Responsa. In Abudraham, 139, the end of r*t~, 35 b, is also given in the name of p: S, but this can scarcely be correct, as in 31*03, 125, and c v n 'm, 90 a, the same passage is ascribed to Rabbi Jehudai, whom Rab Amram followed here as in many other places. i"c, 211, has fre n:3T which seems to corroborate our assumption that R. Nathan was not a Gaon, :n is never used in connexion with a Gaon. * Also lines 14-17, on p. 5 b, seem very suspicious to me. On the use of I::CN, Germany, comp. the Responsum of Rabbi Paltoi in rsi'ia, 149, where 2"i::t are mentioned. THE HALAKIC LITERATURE RELATION OF THE MANUSCRIPTS TO THE PRINTED TEXT. The fact that an old work has been subjected to additions does not preclude the possibility of its having suffered abridgment as well. However, it lies in the nature of these Halakic expositions to give suitable opportunities for additions, especially extracts from Geonic Responsa. It may, therefore, be formulated as a rule, that only the material common to the printed text and the MSS. can with certainty be considered as originally part of Rab Amram's work. Accord between the MSS. and the old authors is not in itself conclusive as to the genuineness of the passages found in them. At most, it proves that such additions, if additions they be, were made in a remote time. And in point of fact there are but few additions in the Sulzberger and the Oxford MSS. that cannot be followed up in one or another old author. Some of these parallel sources to the MSS. of the Seder Rab Amram follow : The resume of the ni3"n nxo in S and O, i , is met again literally in Mahzor Vitry, 3-5, and an abstract of it, in DW 'D, ed. Schiff, II, 235. Besides, the conclusion JNCI "p2Q*T is cited in the Manhig, 7 b, from the Seder. Never- theless, it does not seem at all probable that Rab Amram would give a summing-up of the 01313 nND sent by his predecessor to the Spanish congregations not very long before his own Responsum. The regulations regarding the benedictions over the Tejttlin, the Mahzor Vitry had in the copy of the Seder used for it, in agreement with O, 2, as appears from the remark of the author on p. 642, while the Manhig, 7 b, is in accord with our text *. Mahzor Vitry, 5-6, has the section rvpint^ Kncai? in MSS. S and O, and also TT N^J, found only in MS. S. 1 Rab Amram's view regarding the Tefillin Benediction has been trans- mitted variously in different Poskim. Hence the actual view of Rab Amram cannot be determined any more. 152 THE GEONIM MS. O, 5, is like Mahzor Vitry, 14 both contain the addition pD"M S^anem. Mahzor Vitry (28-32) also has the long piece on ftehn rmjJD, which is found in MSS. S and O (p. 7), and a part of it is described by Rabbi Meir of Rothenburg, in raiBTi '"'W, ed. Bloch, 299, as having been taken from the Seder. On the other hand, from the Manhig, 37 a, we should infer justly that it was known to its author as an independent Responsum, not as a part of the Seder, into which it may have been incorporated later. The reading of the Mahzor Vitry, ^n instead of , 50, that it is an abridgment. In fact, hitherto it has not been observed that a portion of the end of this Responsum is to be found, by way of supplement, on 43 a. The observation on the formula in Dlisn was originally a part of the Responsum given on p. 25 a. This we learn from the Manhig, and there can be no doubt that the author had the correct version. The copyist who omitted it by mistake and he must have lived in very early times, as is shown by Albargeloni, DTijn 'D, 173 atoned for his slip by putting it in under niyi3B> JH *no. How inappropriate a place he gave it is shown by the fact that it was over- looked there by all the scholars of our day. Professor Schechter published a Genizah fragment in the J. Q. R., X, 656, in which the formula of Dlisn has a wording different from the one we are accustomed to, as follows : Bt&rW HJUI $>tae* . . . Dlisn. This benediction runs in pretty much the same way in another Genizah frag- ment published by Professor Levi, R. . J., LIII, 235 : Q^rrv run) pnr DPUD bane* . . . omen. This supposedly new benediction is identical verbatim with that in a Responsum by Rabbi Natronai, quoted in the Seder, 43 a, and in the 1 Priority is not always in favour of the versions of the Seder used by the Poskim. For instance, there can be no doubt that what the 'irt'sc, 184, cites from the Seder is Italian Minhag, and equally i*V, 128, is not quoting an original piece of the y"~c. 154 THE GEONIM Manhig, 23. The Genizah fragments are doubtless of Palestinian origin, for not only, as Professor Levi remarks, was this formula in the Yerushalmi used by Rabbi Isaiah di Trani the Elder, Berakot, IV, 8 c, but the reading is also found in the Vatican MS. of the Yerushalmi. It should be noted, in addition, that the first verse of the Oeullah in the fragment published by Professor LeVi should read : 1J7 r\wy N7S DB> iniPJD }VV3, to which the verse . . . nt?y in n^nx by Jose ben Jose forms an almost verbal parallel further proof for the Palestinian origin of this Payyetan. .SPURIOUS WORKS ATTRIBUTED TO THE GEONIM NAHSHON AND HIS SON HAI. Many a work is ascribed to Rabbi Nahshon, the successor of Rab Amram in the Gaonate of Sura, but his authorship can be maintained with certainty only regarding one of them, the 'Iggul, a treatise on the Jewish calendar system, which Rabbi Eliezer ben Jacob Belin, a German author of the fifteenth century, incorporated in his book nunay, Basle, 1527. That the others have been ascribed to him rests upon a misunderstanding. Though Zunz in his work Zur Geschichte und Literatur, 221, properly said that the Rabbi Nahshon who was the author of the Halakic com- pendium noiNn, a compatriot as well as the namesake of the Gaon, was separated from him by an interval of five centuries, scholars like Miiller, in his Maftea/i, 131, and Weiss, in his Dor, IV, 123, continue to speak of the work noiNl ascribed to Rabbi Nahshon. In view of the fact that it is extremely rare, and that its form is very bizarre, I shall undertake to give a description of it, according to the copy once owned by Halberstam, now in the Sulzberger Collection of the library of the Jewish Theological Seminary of America. The work consists of sixty pages last one blank small quarto, and it was printed in the year 1565 (=DriDD) at Constantinople, according to a MS. in the library of Don THE HALAKIC LITERATURE 155 Joseph Nasi, accompanied by a double commentary from the hand of Rabbi Isaac Onkeneira. The title of the book, abbreviated, runs thus : Nswn f)DV fin , , . "PS NX iy:t? pN3 ^"T pt?ru -n"niD hnan nn nan pro 11 . . . D3nn Nin N^n . . . pnvi . . , n"jn '3 nx> ma enn e>N"i w ava Nr^NDnp . . . anrwa icnp anoo spppn. In the introduction the author names the 22nd Adar of the year 5560 A. M., or 1300 c. E., as the date on which he began his work, and mentions the fact that he was the head of an Academy, TIN 'nanna -iyje> '1N3, frequented by 400 pupils, for the use of whom he had written his little work l . Onkeneira tells us, in the intro- duction to his commentary, that Don Joseph received the MS. of the book from a distinguished old man, DJivro Dn^y 'n mrjn arryiN r~iN3, which probably means when Don Nasi still was in Portugal. At the request of Don Joseph, Onkeneira wrote his two commentaries on the book, the ratio of commentaries to book being ten to one. The last page contains the praise, in prose and verse, of the author, the commentator, and the publisher, Don Joseph, composed by Rabbi Joseph ben Samuel ha- Levi. Not until we reach this last page do we discover that the author bore the title Gaon, but Rabbi Joseph had no intention of identifying him with Rabbi Nahshon, the Gaon of Sura. Rabbi Joseph's own father is denominated Gaon. So far as I can recall, Rapoport, in his biography of Rabbi Nathan, note 30, was the first to be misled by the title of the booklet and to identify the author with the Gaon Rabbi Nahshon. It need not be said that if Rapoport had seen the book itself, he would never have entertained the idea of attributing it to the Gaon. Not only does the author, as was mentioned above, name the year 1300 as the date, but the book is based essentially upon Maimonides. What Onkeneira says, that Rabbi Nahshon 's title 1 What city is meant by -ji is hard to say, probably Bagdad. 156 THE GEONIM was composed of the two words no 1X1, "See Moses [Maimonides]," is probably nothing more than an ingenious conceit, but he is right in assuming the author's dependence upon Maimonides. In point of fact, the book is scarcely more than a brief abstract of the nanai n&ryff majl of Maimonides. The following illustrations show how closely Rabbi Nahshon followed the views expressed in tbe Tad : The first sentence, rv^l na* 1 rvpTCQ, can be explained only from Maimonides, Shehitah, I, 2. The other codes 1 , which follow the Talmud in their wording, speak of nmn, which is ignored by Maimonides and our author, who follows him. The view, p. 14, that the slaughtering knife must be examined after it has been used, is derived from Maimonides, She/iitah, I, 24. It is a view not shared by other authorities. The difference (pp. 3 1-3) between nom PSD and any other PISPIB pSD is inexplicable without the help of Maimonides, Skehitah, V, 3, who uses the case to exemplify his funda- mental view on the subject of the Sinaitic Halakah. The Halakic value of the little book is slight, as we have seen, but the form in which it is couched deserves some consideration. The author attempts to condense in thirty- eight brief and tersely expressed paragraphs the important regulations regarding ntaTiK' and nano. From the point of view of this object, it is not a despicable achievement. An interesting point is the author's desire to imitate the language and manner of the Mishnah, wherein he succeeded admirably. This is all the more noteworthy as the style he uses in the introduction may be called Kaliric, in strange contrast with the clear and pointed style of the book proper. But not even there could he wholly restrain himself from indulging his taste for the bizarre. To the end of each paragraph he adds a JOD, which in most cases is a conundrum, and one cannot but admire the ingenuity of Onkeneira, who succeeded in guessing all the riddles. 1 Comp. the commentators on this passage of Maimonides. THE HALAKIC LITERATURE 157 The explanations by Onkeneira which accompany the little book are of statements of facts and linguistic points. These are treated of in his commentary entitled mya ruav. In his other commentary, called *rn nprn, in allusion to Maimonides, npmn T, he deals with the relation existing between Rabbi Nahshon's statements and those of Mai- monides' Code. He does not attempt to enter into the views of other authorities. In a single passage (p. 23) he mentions Rabbi Joseph Caro, citing his Bet Yosef with these words : Tia nxp |DV nVno D^n Tonn pnaion ain D'O^n nytri D'nbx JV3 f)DV JV3 naoa. Furthermore, he men- tions his grandfather, Rabbi Judah Onkeneira, three times. On p. 12 he tells the following about him: nw npyo rvaa nnvra nx^pca ^vr ypr o^n oann Tonn miiT -n'rno Tonn ann nyis* jytDB D^D ^ ain vy Dp TN .... "sr 'Jpr Dn oann njrn . . . ipoa 'r6 nan vnan mim it^Ni ^y IP^JI ^ jenB> 7 j mirr. The name of his grandfather is not attached to this passage, but on p. 52 it is mentioned plainly, with the words Tiyoen Ti'mo D^n oann ^PT TDnn ^ao, in accordance with which *?"y min 11 we should read on p. 24. His uncle, Rabbi Moses Onkeneira, is referred to on p. 42, in the words ^ao Tiyon iD i?ape> i"n: m3p3iy nets i^nn ^ nn D^trn onnn nonn On p. 32 a saying from the Yerushalmi is quoted which is not found in our text. The Yet^ushalmi very probably refers to some Kabbalistic work 2 . The quotation occurring in a Yemen MS., published by L. Griinhut, in R. . J., XXXIX, 31 1-12, is probably taken from a mystical work attributed to Rabbi Nahshon 3 . 1 Rabbi Judah ben Isaac, Rabbi of Magnesia about 1500, author of a commentary on Ruth. 2 rrooini baro'b xc'i :T rrb n'i jun 'obci'a pnawia wo'2 NJTT : WIT ; the language is that of the Zohar ; so far as I know, however, the dictum does not occur in the Zohar. 3 The extract published by Griinhut was known before ; comp. R. .J., 158 THE GEONIM The Karaite Kirkisani, as we are informed by Dr. Harkavy 1 , who published portions of his works still in MS., speaks of " Hai, the head of the Academy, and his father, who translated the law-book of Anan from the Aramaic into Hebrew, and with the exception of two points, they found nothing that could not be traced back to the Rabbinic writings." As Kirkisani could not have been thinking of Rabbi Hai ben Sherira, because he wrote before the great Hai was born, he may have meant Rabbi Hai ben Nahshon, who studied the works of Anan with his father Nahshon. It is possible that the calendar in- vestigations undertaken by Rabbi Nahshon in connexion with his f lggul led him to take up Karaitic literature, and he naturally sought first of all to familiarise himself with the works of the founder of the Karaite sect. If we bear in mind that the Gaon of Sura, Natronai, barely one generation before Rabbi Nahshon, had to be told by a Spanish Jew of the existence of Anan's book of law 2 , it does not seem at all probable that an early successor of his would make it the subject of close study. And, in point of fact, Kirkisani's report bears the marks of falsification. Consider the monstrous exaggeration, that the Gaon Hai had found only two matters in the whole of Anan's book of law that could not be shown to be derived from Rabbinic sources, the truth being that there are barely two lines in his book that are in agreement with the Rabbinical authorities. It is equally out of the question that a Gaon should have busied himself with the translation of a Karaite book, and from Aramaic into Hebrew at that. The Babylonian Jews mastered XL, 128. Rabbi Nahshon is not the only Gaon whom the Kabbalists claim as one of their own. Even Rabbi Samuel ben Hofni could not escape them, in spite of his philosophic views ; comp. Steinschneider, Arabische Literatur, no, note 6. 1 In his additions to the Hebrew translation of Graetz's Geschichte, III, 493-5". * Seder Rab Amram, 38 a. THE HALAKIC LITERATURE 159 both languages, we may be sure, and it is not to be supposed for a moment that Rabbi Hai was desirous of making propaganda for Karaism among foreign Jews ignorant of Aramaic. WORKS ATTRIBUTED TO THE GEONIM ZEMAH, HAI BEN DAVID, AND HILAI. The contemporary of Rabbi Nahshon, Rabbi Zemah ben Paltoi, Gaon of Pumbedita, was the first of the scholars of Pumbedita to write a book, and this first Pumbeditan book was at the same time the first of the long line of Talmudic lexicons. The work "yny is known to have existed as late as the sixteenth century, in the possession of Rabbi Abraham Zacuto, the author of the Yohasln, who quotes from it here and there. It is, of course, astonishing that Zacuto should be the only one known to have made use of the work, still more astonishing that he was the only one to make mention of it. Kohut's opinion that Rabbi Nathan ben Jehiel resorted to the work of this predecessor of his is not based upon sufficient grounds 1 . In view of all this, I cannot refrain from expressing doubt as to the reliability of Zacuto's report. He may have come into possession of a Talmudic lexicon by some Zemah, otherwise not known, whom he or perhaps the copyists of the book, without taking the trouble to investigate the matter, identified with his namesake, the Gaon of Pumbe- dita. The restricted number of quotations from the lexicon hardly permits speculation as to the merits of the book. Zacuto tells us expressly that the arrangement followed the alphabet. An interesting feature is that it contained the names of persons and places in the Talmud 2 . 1 Comp. G. S., p. 294. 2 In the Introduction to his 'Aruk, 17-19, Kohut has put together all the quotations from Rabbi Zemah's lexicon, following the example set by Rapoport and Geiger. Rabbi Zemah's explanation of the oath Jiyon mrt, declared unintelligible by Zacuto, and by Rapoport and Kohut after him, is quite correct. Rabbi Zemah observes that ys~.n p msi 'i makes l6o THE GEONIM It must be mentioned that Rabbenu Hai is perhaps alluding to a lexicographical work by Rabbi Zemah ben Paltoi, when, in giving the explanation of a Talmudic word, in Harkavy, 200, he uses the expression ~ipno mimi . . . "IEX1 TOS m no "ipnB> " and in the investigation Rabbi Zemah pursued, in which he made the supposition." If he had been having a Responsum by Rabbi Zemah in mind, the expression used by Rabbi Hai would be very peculiar. That he did not mean Rabbi Zemah ben Hayyim, or Rabbi Zemah ben Kafna, is shown by a previous sentence, in which he gives the full name, Rabbi Zemah ben Paltoi 1 . On the other hand, the grandson of Rabbi Zemah, Rabbi Hezekiah ben Samuel, mentions nothing of a dictionary by his grandfather, in his letter published in the /. Q. R., XVIII, 401. As he was not a little proud of the numerous writings by his ancestors, it is not very likely that he would have forgotten the lexicon, if there had been one. A final possibility is that this lexicon of Rabbi Zemah is nothing but the explanation of Talmudic passages for which he was asked, and these are included in what his grandson says : " And also in the days of his [Rabbi Paltoi's] son, Zemah, the head of the Academy, my father's father, they [the Spanish scholars] sent to him asking him for explanations of the difficult passages in the whole Talmud, so many that several donkeys could not carry the load." These words would seem to point to a comprehensive work by Rabbi Zemah rather than his activity as a Responsa writer. The superscription reproduced in G. S., p. 28, from a Genizah fragment containing a collection of Responsa, "These Responsa were arranged [jpn] by Rabbi Zemah, use of the oath, because he lived during the time the Temple was standing, and being accustomed to swear "by this Temple," he did not change the formula even after its destruction. 1 Attention should be called to the fact that neither R. Sherira nor his son R. Hai refers to R. Zemah as his ancestor, though the former's grandmother was a daughter of R. Zemah, comp. above, p. 10. THE HALAKIC LITERATURE l6l the head of the Academy," might be interpreted to mean that Rabbi Zemah ben Paltoi (?) had made a collection of Responsa. This inference would receive support from the fact, that, as is shown in the G. S. t pp. 20 et seq., several of the Responsa attributed to Rabbi Zemah belong to his predecessors without a doubt. Thus he might be looked upon as a collector of Responsa issued by earlier Geonim. However, it is highly improbable that a Gaon should have engaged in the task of collecting Responsa 1 , especially in consideration of the fact that the Responsa Collections that have reached us were, in all likelihood, made toward the end of the Gaonate, and then outside of Babylonia. Accordingly, fpn should be translated by "composed," rather than " arranged 2 ." Among the doubtful Geonic works is the one on the Rabbanite calendar, ascribed by the Karaites (nVJlonp ^\b, II, 148-51) to " Hai, the head of the Academy." If this statement is not to be dismissed as a pure invention, at least so much may be asserted, that the author would have to be identified with Rabbi Hai, Gaon of Sura, whose father, Rabbi Nahshon, as was mentioned a little while ago, also wrote upon the calendar, rather than with Hai 1 Frankel, Entwurf einer Geschichte . . . der Responsen, 71-2, misunderstood the expressions nin'pN or rnaicn used by the old authors. It does not mean "Responsa Collections," but simply Responsa, the plural being employed because the correspondents in almost all cases addressed a number of questions to the Gaon. 2 Comp. Zunz, Gesammdte Schrtften, III, 51, on the use of fi?n, "to compile"; also Harkavy, 84: nos no ibN mtow?, "these [replies to] questions addressed to Rabbi Zemah." Luzzatto, Bet ha-Ozar, I, 83, maintains that Rabbi Zemah was the compiler of a collection of Geonic Responsa. He bases his view on Mordecai, Baba Batra, 471, where the JIM mas 211 m:w royron are spoken of. But there can be no doubt that the text of Mordecai is corrupt, and must be read as emended by Isserles, ad loc. The old name for Geonic Responsa was rnforo ruicn (D*rr, 45), which later was cut down to mto*D (VncN, III, 49), or chiefly rvunrn. The post-Geonic authors speak more frequently of nanrm m"J than of rnVwj rvuiicn, but there are well-known Responsa Collections by later authors that have appeared in print under tho latter title, for instance, the pn: p mm i:'rA m^Nir nuiujn. I M 1 62 THE GEONIM ben David, the successor to Rabbi Zemah ben Paltoi in the Gaonate of Pumbedita, as Harkavy does in his Additions to the Hebrew translation of Graetz, Geschichte, III, 506. Muller, in his Mafteah, 153, calls the Sura Gaon, Hilai ben Natronai, the probable author of niplDB rna/rt. But this rests upon a misunderstanding. The words of Rabbi Hilai, in s"n, 47, HDB3 pon paj6 utnw ITia, do not refer to a Halakic compendium but to his Responsum, D'n, 162, which he probably sent to the same addressee. THE IMPORTANCE OF RABBI SAADIA IN HALAKIC LITERATURE. As in many other fields, so in the Halakah, Rabbi Saadia was the most important author of the Geonic time. Not only did he enrich the various departments of Halakic literature with numerous contributions, but also what he wrote was so original that in many respects it served as a model for the succeeding Geonim and later scholars. His Halakic writings may be divided into four groups : (i) Introduction to the Talmud and the Halakah ; (2) Tal- mudic explanations ; (3) Codification of the Rabbinic laws ; and (4) Liturgy. Unfortunately, most of his Halakic works are lost to us, and the greater part of what we possess of them still awaits publication. In the first group belongs the nn j" B>na, published by Professor Schechter in the Bet Talmud, IV, 235-44, after an Oxford MS., and reprinted by Muller in (Euvres complets de Rabbi Saadia, IX. Originally it was written in Arabic, and it contains the fundamental hermeneutic principles applied to the Halakah by the Rabbis, its form being a commentary upon the "Thirteen Rules of Rabbi Ishmael." Each of the thirteen rules is illustrated by numerous examples, and at the same time all the variations falling under the rule are elucidated. For instance, the application of the first hermeneutic rule, the "iim bp, is exemplified by means of four Scriptural injunctions. The kw, says Saadia, tells THE HALAKIC LITERATURE 163 a man that in case he marries a second wife, he has three duties of a husband (Exod. xxi. 10) to fulfil toward his first wife. But the law contains nothing about the duties of a husband who has but one wife. These duties we derive by applying the "inirn i>p, and we infer that if the law puts certain duties upon a husband of two wives, although the fulfilment of them is twice as difficult as when he has only one wife, how much more is he obligated to fulfil them when he has but one wife. In this clear and intelligible manner, he continues to treat of all the hermeneutic rules in succession. The superiority of this work appears plainly on comparing it with the " Baraita of the Thirteen Rules," at the beginning of the Sifra. Neither in copiousness of examples, nor in lucidity of presentation, can the latter come up to Saadia's work even remotely. The relation of this Baraita to Rabbi Saadia's treatise, it should be said, has not yet been cleared up l . The Baraita, we know, contains many old elements, but it is not certain that, in the passages in which Rabbi Saadia's work and the Baraita are in agreement, it is always the latter that is to be considered the primary source. An Introduction to the Talmud by Rabbi Saadia was consulted by so late an authority as Rabbi Bezaleel Ashke- nazi (ab. 1609), and Azulai, in his }T1X Tjp, 36 b and 68 c, has published bits of it, after Rabbi Bezaleel's manuscript works. As Rabbi Bezaleel says expressly that it was originally written in Arabic, there is no room left for doubt as to the correctness of the supposition made by Professor Schechter, in his Saadyana, 128, that the 3Nna i>31O^K mentioned in a Genizah fragment is precisely this Introduction to the Talmud by Rabbi Saadia 2 . 1 Comp. on this point Miiller, in (Euvres complete de R. Saadia, IX, Introduction, 23-33, and Hoffmann, in Berliner-JubelschriJt t Hebrew division, 55 et seq. * Rabbi Saadia's 'i"n 'D contains matter of a nature introductory to the Talmud, as he himself mentions expressly ; see Harkavy, Saadia, 152, 160. The former passage is particularly interesting. Rabbi Saadia, M 2 164 THE GEONIM Rabbi Pethahiah, of Ratisbon, who travelled through Babylonia near the end of the twelfth century, reports that the Jews there used commentaries on the Bible and the nmo net? by Rabbi Saadia. Whether D"E> stands for the Mishnah, or, according to later usage, for the Talmud, cannot be determined with certainty. It is also open to doubt whether the *tma of Rabbi Saadia on certain Talmud passages which are mentioned in Geonic literature 1 are commentaries on the Talmud or part of the Talmud, or whether they are isolated explanations of definite passages in the Talmud, which Rabbi Saadia, like many of the Geonim, gave in his Responsa in reply to inquiries. From the list of works published by Professor Schechter in Saadyana, 79, it is plain that Rabbi Saadia compiled a " Vocabulary of the Mishnah." There is, accordingly, no reason for denying him the authorship of the Commentary on JTisnn, published at Jerusalem, 1907, by Wertheimer, from a Genizah fragment, under the title 2T B>TVa "iao pw iTiyo. The epithet t2>1Ta is somewhat inaccurate, because the book contains no explanations in the usual sense, but only very brief lexical notes. The sixty-three folios of the treatise Berakot are disposed of in two small leaves. However, it is not impossible that the JTD-O by K>na before us is only an extract from a much more detailed commen- tary by Rabbi Saadia, in which the philological notices alone are given, to the exclusion of all other sorts of matter. This hypothesis gains in probability from the fact with fine satire, takes the Talmudists of his time severely to task : " The reason for compiling this [chronology of the Talmud] is that I have met persons who call themselves Kabbis [Rabbanites ?], who have no understanding of it, and who do not walk in the way of our old teachers, whose names, however, are always upon their lips, and with whose food they nourish themselves." These words show not only that Rabbi Saadia was creating a new thing in this field, but also that he did it in opposition to the Zeitgeist so-called. 1 If 'cne may be taken literally, then Rabbi Saadia must have written commentaries at least upon Pesafnm, Sotah, Bdba Mezia, and Baba Batra ; comp. Saadyana, 59-61, and Albargeloni, nvracn 'c, 53. THE HALAKIC LITERATURE 165 that the first Mishnah l of the treatise is summed up in such wise that it may serve as an introduction to the discussions following in the Gemara. It cannot be assumed that Rabbi Saadia treated only the first Miahnah in this thoroughgoing manner, and not also the rest of the treatise. This odd contrast between the first Mishncth and the others would be fully explained by the supposition that we have only an extract before us. The epitomiser contented himself with reproducing verbatim the first paragraph of the book of which he was preparing an abstract; thereafter he took the shortest way possible. Rabbi Saadia's literary activity was most fruitful in the department of codification. Unfortunately, only scant remains have been preserved, but at least the titles of his works are cited by a number of old authors and in old lists of books. This enables us to assert definitely that at least the following ten parts of the Jewish law were codified by Rabbi Saadia 2 : iw; mjnap; nnas?; jnpB; nwno; nicrv; ruiro nuno ; manoi ntrnp ; mny ; mnm nsim Of these ten books, but one has been preserved, the first-named, "the book of the law of inheritance," which was published in the ninth volume of Saadia's collected works. Fragments have come down to us of two or three codes besides. The student need not be cautioned against judging Rabbi Saadia's achievements as a codifier by the insig- nificant remains enumerated, the more as it appears that the niETVn ">QD was his initial effort in the code depart- ment 3 . Despite its shortcomings, the book nevertheless 1 Also the three passages 1 1 b, 15 a, and 18 b, are more than mere verbal explanations. 8 Comp. Steinschneider, Arabische Literatur, 48-50, and Dr. Poznariski's " Schechter's Saadyana," and also his remarks in the Orienlalische Litteratur- Zeitung, VII, 306-7 ; to which is to be added Rabbi Saadia's treatise on rvyi, published later in J. Q. R., XIX, 1 19. Numerous citations from the 'moan 'c are to be found in Albargeloni's work of the same name. 3 This view, expressed by Miiller in the Introduction to his edition of this book, gains in probability from what is said, p. 166, below, on the relation of the book to Rabbi Saadia's other book, the jnpcn 'D. I 66 THE GEONIM gave scope for the display of Rabbi Saadia's originality. Not only is it the first Rabbinic book in Arabic, but also in plan and execution it reveals the influence of Greek- Arabic discipline 1 . Instead of ranging the decisions of the Miahnah and the Talmud next to each other, Saadia has presented the Biblical-Rabbinic laws of inheritance in an order quite independent of their sources. This book of his thus became in some respects the model of the Geonim Rabbi Samuel ben Hofni and Rabbenu Hai for their codifications, and it would not be going too far to assert that Saadia exercised some influence on Maimonides' code. It is interesting to note that the fragments of the jnpan ISD, published by Professor Schechter in Saadyana, 37, 40-41, show that in this code Rabbi Saadia pursued an entirely different system from that employed in the nttPHYi ISD. It is not impossible that Rabbi Saadia's method of not mentioning the Talmudic sources from which he drew gave offence, as similar action by Maimonides in his Tad aroused opposition. Saadia may have been led thereby to change his method. In the domain of liturgy, we cannot here give attention to the numerous prayers which Rabbi Saadia composed. We are interested in the prayer-book which he compiled at the request of the Egyptian congregations. Unfortu- nately, it still awaits publication, and we are, therefore, not yet in a position to pass final judgment upon it. So much is certain, however, that Rabbi Saadia did not, like his predecessor in the Gaonate of Sura, Rab Amram, execute his task according to the Babylonian ritual, but according to the ritual of his native country Egypt. Of course, it cannot be denied that his Seder was not without effect on the Babylonian liturgy. Rabbenu Hai (Harkavy, 97) states explicitly that certain changes in the liturgy of his country were due to the influence exercised by Rabbi Saadia's Seder. Although the Egyptian liturgy 1 Comp. Steinschneider, Arabische Literatur, 48, end, and Orient. Litt.- Zeilung, VII, 206-8. THE HALAKIC LITERATURE 167 is not free from Babylonian influences, yet, on the whole, it is an offshoot of the Palestinian ritualistic system. Whether the kinship that exists between the Seder, of Rabbi Saadia and the Order of Prayers by Maimonides, which I have pointed out elsewhere *, is attributable to the sole circumstance that both authorities were concerned with the needs of the Egyptian Jews, is more than questionable. It is very probable that Maimonides was intimately acquainted with the Seder of Rabbi Saadia, and permitted himself to follow it in many respects. Rabbi Saadia's place in the development of Halakic literature can be summed up in this way : The many- sided scholar endeavoured to free Halakic literature from its exclusiveness. His Introductions and his methodo- logical works tended towards a historic-critical understand- ing of the Talmud, while as a codifier his aim was to arrange the Rabbinic law in a unified logical system. THE THREE GREAT SUCCESSORS OF RABBI SAADIA. The last three Geonim, Rabbi Sherira, Rabbi Samuel ben Hofni, and the son of the former, Rabbi Hai, all stand 1 Z.H.B., IX, 104-7. After an examination of the MS., which I gave it later, even though it was cursory, I do not entertain the slightest doubt that Rabbi Saadia's ITC embodies the Egyptian ritual. The Genizah fragments comprise only a few insignificant tattered pieces of the y'to and very large pieces of Rabbi Saadia's Seder, further evidence of the assumption that it was destined for and went to Egypt. To the liturgical decisions by Rabbi Saadia given by Muller, in (Euvres compkts de R. Saadia, IX, 150 et seq., most of which are probably derived from the Seder, a quotation is to be added occurring in Ibn Gabai, spy njAin, the section on TOWO ifjcn. Ibn Gabai, it must be confessed, does not seem to have taken it direct from Rabbi Saadia. The anonymous commentator of the German Prayer Book, printed at Trino, in 1525, was acquainted with Rabbi Saadia's Seder. He quotes it in his commentary on the Haggadah on the verse wuyi. The passage quoted by him is not found in the Oxford MS. of the Seder, but it occurs in the Haggadah according to the Yemen ritual, in the 1*0, 293, in a MS. of the Haggadah according to the German ritual, of the year 1329, in the possession of the Jewish Theological Seminary, and was known to the author of the Vnjrt 'o, comp. , 52, ed. Hoffmann. I 68 THE GEONIM under the influence of Rabbi Saadia, manifesting itself peculiarly in the case of each. While Rabbi Samuel followed the example of Rabbi Saadia in the field of philo- sophy and Bible exegesis, as well as in his other interests, Rabbi Sherira and his son Rabbi Hai remained true to the old traditions of the Geonim. Of philosophy the latter would none, and the study of the Bible was a subordinate pursuit. To their core they were Talmudists, and Talmudists only. But in their capacity and work as Talmudists they could deny the influence of Rabbi Saadia as little as Rabbi Samuel ben Hofni. A work entitled D^DD nbo is ascribed to Rabbi Sherira, but the statement is rather doubtful. In his Introduction to his Menorat ka-Maor, Rabbi Isaac Aboab quotes a state- ment of Rabbi Sherira's from DnnD fbjD 1 . What Aboab meant was probably that he had taken the words of the Gaon from the book Dnno fbso by Rabbenu Nissim. Like his Maftea/t, this book by Rabbenu Nissim is also made up in large part of Geonic Responsa 2 , and of these Aboab made use in other places, too. It is equally doubtful whether the *pliM by Rabbi Sherira, cited several times by Rabbi Isaac of Vienna in his book jmr "UK 3 , is an independent work, somewhat of the cha- racter of a commentary on several treatises of the Talmud, or explanations of Talmudic passages in the form of Responsa. 1 The correct reading is DTTIC, not we. a This is confirmed by the Responsum of Rabbi Hai, in the appendix to Rabbi Sherira's Letter, ed. Mayence, 64-5, which likewise was in- corporated verbatim in Rabbenu Nissim's cnrc 'm . Comp. also Harkavy, in reran, V, 53 : Briill, Jahrbucher, IX, 121 ; and G. S., p. 273. 3 II, 168 a ; Baba Kama, 72 ; Baba Batra, 40. The Geonic sources used by Rabbi Isaac, the author of the i"w, which are of great importance for the valuation of Geonic literature, have not yet been exploited sufficiently. Wellesz, in Monatsschrift, XLVIII, 369-71, is neither exhaustive nor com- plete. For instance, the Sheeltot quotations from I, 159 b, II, 50 and 163, are missing; also Rabbi Hanina Gaon. I, 209; Rabbi Nathan ben Hananiah, I, i76b, and several others. Vow, II, 76, seems to indicate that Rabbi Sherira wrote a commentary on Baba Baira, comp. also Steinschneider, Arab. Lit., 98. THE HALAKIC LITERATURE 169 Rabbi Sherira's reputation as one of the most prominent authors of the Geonic period rests upon a much surer basis than is afforded by these doubtful productions upon his celebrated Letter to the scholars of Kairwan. The Letter is a reply to a question addressed to Rabbi Sherira as to the origin of the Mishnah and the other Halakic collections by Tannaim, and as to the heads of the Academies during the time of the Saboraim and Geonim, together with a number of other points connected with these two cardinal matters. The lasting value of his epistle for us lies in the information Rabbi Sherira gives about the post-Talmudic scholars. On this period he is practically the only source we have, and his report is all the more important as it is partly based upon documents in the archives of the Geonim. But we should be doing Rabbi Sherira injustice if we thought of him merely as a chrono- logist. The theories which he unfolds, in lapidary style, regarding the origin of the Mishnah, its relation to the Tosefta and the Baraitot, on the beginnings and develop- ment of the Talmud, and many other points important in the history of the Talmud and its problems, stamp Rabbi Sherira as one of the most distinguished historians, in fact, it is not an exaggeration to say, the most distinguished historian, of literature among the Jews, not only of an- tiquity, but also in the middle ages, and during a large part of modern times. But the fine historical percep- tions which he displays in literary criticism, and his searching investigation of the problems he encounters are almost unthinkable in the Geonic period without the preliminary work, or rather the personal influence, of Rabbi Saadia 1 . By far more direct and tangible was the influence of Rabbi Saadia upon the work of Rabbi Samuel ben Hofni, who was a serious competitor of Rabbi Saadia in point of versatility and productiveness. He cannot, however, 1 Comp. the observation by Rabbi Saadia given above, p. 163, n. a. 170 THE GEONIM vie with Rabbi Saadia in originality. The Halakic works of Rabbi Samuel, some of them, perhaps, nothing but works of Rabbi Saadia recast 1 , were written in Arabic like those of his predecessor, and they share the fate of the latter, too, in that they are completely lost save a few fragments. The Genizah fragments have made us acquainted with a large number of titles of books, as many as forty, all to be added to the Halakic writings of Rabbi Samuel 2 . It is fair to assume that these are not independent works 3 , but rather parts of a great code. WTB^H "Commands," by Rabbi Samuel, may have been the general title, which was accompanied by a number of sub-titles for the various divisions of the code. The gigantic compass of the book may readily be judged from the rwu njJP, " The Portal of Benedictions," which was published by Weiss in the Bet Talmud, II, 377-86. This division, doubtless an insig- nificant portion of the code, exceeds in size the correspond- ing parts in Maimonides' Yad and Caro's ShuUan 'Aruk together, and it must be remembered that it has not been preserved in complete form. Probably this prolixity is a partial reason why both the Arabic original and the Hebrew translation, which were in the hands of the German authors as late as the fourteenth century 4 , have dropped into total oblivion. Of the other Talmudic writings of Rabbi Samuel, we should mention a commentary on Yebamot, listed in a 1 Comp. Schechter, Saadyana, 43. 3 Comp. Steinschneider, Arab. Lit., 108-10, and Poznanski, Orientalische Litter atur-Zeitung, VII, 313-15. In the recently published nunn nn (Bernard Drachman, New York, 1908), 53, the nvnyn 'c (on witnesses?) by Rabbi Samuel is mentioned. 3 A supposition made by Rapoport, Biography of Rabbi Hai, note 8. 4 The author of fcno 'DO D'Eip 1 ?, published in Coronel's 'yip 'n, quotes Rabbi Samuel's onyizj, and also the author of mcicn 'c, living at the same time. Some of the decisions by Rabbi Samuel, reproduced in Miiller, Mafteah, were not Responsa originally, they are taken from his code. THE HALAKIC LITERATURE 171 catalogue, /. Q. R., XVI, 411, and an Introduction to the Talmud, of which a considerable piece is to be found in the Taylor-Schechter Collection. The influence of Kabbi Saadia is patent in the nwu ''"W, especially in the grouping of the material and in the style of presentation. It is altogether likely that Rabbi Samuel used the work of his predecessor as a foundation for his Introduction to the Talmud as well as for his Code. Rabbi Hai, the last of the Geonim, who as a Talmudist may perhaps be called the first of them, and who in respect of Talmudic scholarship, profundity of conception, and incisive judgment, is excelled by none, not even by Rabbi Saadia, is known chiefly for his numerous Responsa. How- ever, he is the author of independent works on subjects in every department of the Talmud, too. Of his commentaries on the Talmud nothing has been preserved, though it is certain that he expounded several treatises. Quotations from his commentary on Berakot are to be found in Ibn Gajat, SJ>"B>, 1, 14 ; Albargeloni, DTiyn ISD, 288 ; in the MS. of the nuBTi 1 of the RaBeD ; and in rr'atn, 24. Rabbi Solomon Ibn Adret makes copious use thereof in his commentary on Berakot. We may also be sure of his having composed a commentary on Shabbat 2 , to which reference is made in h"l , 59, and that the expression ""C^iTM wm in this passage does not mean an explanation made by Rabbi Hai in one of his Responsa is evident from the word nmeTi that follows soon after. It is obvious that in this Responsum a difference is made between fc^Ta and maiBTi. It is questionable whether Rabbi Hai wrote a commentary on the treatise Hagigah. Albargeloni, in his commentary on the book Yezirah, cites explanations of passages in this 1 I am indebted to Dr. Alexander Marx for calling my attention to these jrutEn against Rabbi Zerahiah Gerondi ; they are in the Sulzberger Collection of the Jewish Theological Seminary of America. 3 Comp. G. S., p. 56, and 'Aruk, s.v. Ntco p, which quotes Rabbi Hai's explanation of this expression from Shabbat and not from Kelim I 172 THE GEONIM treatise 1 five times, once (p. 26) as wan 'a3, and again as penn px 'aa rcrpaa. What is certain is that the view of Weiss, Dor, IV, 187, cannot be correct, when he holds that whenever the author of the 'Aruk quotes the words of Rabbi Hai with the intro- ductory formula en^a he had a commentary of the Gaon before him. It is curious that Weiss should have dropped into the incorrect statement that Rabbi Nathan, s.v. HitD^N, was quoting Rabbi Hai's commentary on Kiddushin. The words pBTipl tnna Yaai show plainly that Rabbi Hai's explanation could not have had a place in a commentary on Kiddushin. In such a case he would have had to say ppnaai. Indeed, some of the explanations of Rabbi Hai introduced in the 'Aruk with SPTa are found in Responsa. For instance, that s.v. TTin riTini is literally in Harkavy, pp. 128-9. Likewise, Rabbi Hai's authorship of the brief commentary on the Order Teharot of the Mishnah seems to me very dubious. My reasons against the prevailing assumption that this commentary ascribed to him is actually his, are the following : Rabbenu Hai, like many other Geo- nim, did not consider it beneath his dignity to give short linguistic explanations of Talmudic passages, when he was asked for them. We have, indeed, a large number of such by Rabbi Hai in various places in the Responsa Collection edited by Harkavy. On the other hand, it is highly improbable that a Gaon, especially a scholar like Rabbenu Hai, who was mainly concerned about a proper understanding of the Halakah, should have composed a commentary on a most difficult part of the Miahnah, without making the slightest contribution to our actual knowledge of it. The explanation offered by Weiss for this peculiar fact can hardly be taken seriously. He maintains that as this Order of the Mishnah was studied only by great scholars, it required nothing but linguistic elucidations ; 1 Probably it refers to a comprehensive Responsum on the difficult Haggadic parts of the second section of this treatise. Comp. G. S., P- 273- THE HALAKIC LITERATURE 173 the matter itself contained therein needed none. In other words, Rabbi Hai might presuppose in his readers an intelligent appreciation of the most difficult parts of the Halakah, but not acquaintance with such words as ^DBD, "ina, V3\o, and many similar terms. They occur frequently in the Talmud, yet Rabbenu Hai must define them for his great scholars 1 There are other circumstances that militate against Rabbi Hai's authorship. In this commentary on Teharot, Greek equivalents for certain words are not infrequently cited, and we are certain that Rabbi Hai understood no Greek \ The numerous quotations from the Yerushalmi also testify against Rabbi Hai's author- ship. Though he does now and again make references to the Yerushalmi elsewhere, the frequency with which it is done in this commentary arouses suspicion. Moreover, not only is the Yerushalmi drawn upon freely, but also contemporary Palestinian custom is cited (Kelim, XXV, 3), which hardly fits in with our notion of Rabbi Hai. Though Rabbi Saadia and Rabbi Nahshon are named in the com- mentary (Kelim, XXVIII, 3), Rabbi Sherira never is, which would be rather curious in a work by Rabbi Hai. Also Rabbi Hai never speaks of the Responsa of the Geonim as JT6w; he calls them nniETi, while in the commentary m^NB> is the term constantly employed. And what ex- planation can be given of the fact that the author of the 'Aruk quotes it seventy times without once mentioning the name of Rabbi Hai 2 . In view of all this, Rabbi Hai's 1 The explanation of the word sophist is quoted by Rabbi Hai, as we learn in Harkavy's Introduction, 25, note, from a work by .Alfarabi ! His ignorance of Greek is evinced also in his remark on caiiTN, Harkavy, 196-7. In another Responsum, 1. c., 23, he says with regard to the names of certain fish in the Talmud : jm ]Y3O i: f m p :v jvri j*o pin jrw tai ! This would seem sufficient to refute Weiss' statement that Rabbi Hai understood Greek. 3 Kohut, in his Introduction, 14, maintains that Rabbi Nathan, s. y. nD, ascribes the commentary on Teharot to Rabbenu Hai, and calls it -co p*u. But if this passage proves anything, it is that Rabbi Nathan did not consider Rabbi Hai the author, inasmuch as he never calls him anything but p*ert. 174 THE authorship of the commentary is, to say the least, very doubtful. The codifications by Rabbi Hai encountered a more favourable fate than his commentaries. Following the example of Rabbi Saadia probably, he wrote them in Arabic, but only the Hebrew translations have been pre- served, and they only in part. Rabbi Isaac ben Reuben translated l Rabbi Hai's book On Sales as early as the year 1078, giving it the title "DKO npon IQD. It has been printed and published a number of times. To this book with its sixty gates are added three comparatively short treatises on the law of pledges, patron 13D ; the law of conditions, DBB>O ; and the law of loan and sale, niwSn DBPD . A second work of importance by Rabbi Hai in the same field is his work on oaths, of the Hebrew translation of which, niyDE> "Hyt^, we also have a printed edition. Of these two works there is a metrical version, which, however, does not own Rabbi Hai as its author, the statement of the printed editions to the contrary notwithstanding. These two works by Rabbi Hai are to be classed among the most excellent achievements in the department of Rabbinic code literature. As Rabbi Hai treats only certain portions of the Rabbinic law, he naturally goes into detail, without, however, dropping into the longwindedness of which his father-in-law, Rabbi Samuel ben Hofni, is guilty. The logical development of the subjects treated is presented in a clear and comprehensive way, and the systematic grouping is masterly. The 13ED1 npnn IDD is to this day the best exposition of the Rabbinic law of sales with all its essen- tial branches. Equally, his nijnatr nyt? shows the cunning of the great systematiser and the acumen of the great jurist. In the first-mentioned book, XLI, 77 a, he refers to his work, 2 Sinn "nan, which seems to be lost. Perhaps the treatise niNl^n 'DBIPO is nothing but a chapter of this book. 1 On the translations of Rabbenu Hai's works, comp. Steinschneider, Arabische Literatur, 99 et seq. 2 Comp. Briill, Jahrbiicher, IX, 120. THE IIALAKIC LITERATURE 175 Among the lost works in codification by Rabbi Hai there is one on IJTm niDN, arranged, like the other, in "gates," which is cited by some old authorities l , and also a treatise on the prerogatives 2 of the owners of adjoining possessions, nmm Mention is made, besides, of Rabbi Hai's p^an niabn. This may have been an extract from his Seder 3 , which probably, like the Orders of Prayer of his predecessors, contained the prayers and the Halakot bearing upon them. The Seder seems to be lost irretrievably, and nothing can be conjectured about it, except perhaps this one thing, that it may have been put together either for the congregations of the Crimea or for those of Byzantium. At all events, the Jews of those regions had a tradition about having received a prayer-book from the Geonim 4 , and as neither Rab Am ram's nor Rabbi Saadia's could have been meant, Rabbi Hai's naturally suggests itself. One other circum- stance should be mentioned in connexion with the Seder of Rabbi Hai. He himself reports (Harkavy, 105, bottom) that young men from Constantinople studied the Talmud under him, and it may have been at their instance that he arranged a Seder. A Halakic work by Rabbi Hai, his Book of Documents, was found recently among the Genizah fragments. It con- tains twenty-eight forms for drawing up documents, together with brief directions. Dr. Harkavy, who publishes four of these documents in the Hebrew Journal ruoan, III, 46-50 5 , 1 Rapoport in his biography of Rabbi Hai, note ai, refers to a quotation from a work of this sort. However, traces of it can be shown to exist in several authors. Comp. DTIE, 17 b and 17 c (?), and the index to authors in Vn*air, ed. Buber. 2 Not boundary disputes, as Steinschneider, Arabitche Literatur, roo, says. 8 In VrVaip, 267, end of paragraph, .vn '-\ nco means his Seder ; the author applies the same word to Rab Amram's Seder : ncca aro cio? ail. Buber's emendation, 137, moa for ricoa is superfluous. Other references to Rabbenu Hai's Seder in ^n'asj are 264 and 294. Comp. also Stein- schneider, Arabische Literatur, 102. 4 Comp. the Hebrew monthly, Vocwn, I, 147. 6 The concluding sentence of the tree B3, 48, which Harkavy could 176 THE GEONIM ascribes the book to the Gaon Rabbi Hai ben David. The reasons for such ascription were inadequate to begin with l , and they have now been nullified by another Genizah fragment, come to hand in the meantime, wherein Rabbi Hai ben Sherira is explicitly called the author 2 . Rabbi Hai, like his father Rabbi Sherira, and his father- in-law Rabbi Samuel, is unmistakably under the influence of Rabbi Saadia. This influence is betrayed plainly by the arrangement of his works in codification. The interests of Rabbi Hai centred largely in the civil law. His independent works belong almost exclusively to this domain. Well aware that his acute analysis of certain legal discussions might be applied in dishonest ways, he tries to guard against abuse in the following words at the end of his rnyus? l nj?B>: "And if an interested party should derive arguments from this presentation to twist the words and win his cause, he will bring evil down upon himself. I am innocent before my Creator, for I have composed this work only for those who walk in the straight path, to understand how to give just decisions. . . . The Holy One, blessed be he, will be my avenger, that the readers of my book use it in fear of God and in truth, and also the Lord, before whom all hidden things are manifest, will espouse the cause of my innocence, as it is written : ' As for such as turn aside unto their crooked ways, the Lord shall lead them forth with the workers of iniquity, but peace shall be upon Israel.' " not explain, must be read as follows : pcb (frf) = ) jb J'lrro (j:i = ) jw 1 Dr. Harkavy's argument, FyoNn, V, 152-6, that this nrrairn 'D must be older than Rabbi Saadia's, for the reason that it is less comprehensive, cannot be taken seriously. The same logic would make Rabbi Samuel, the author of rrvya rftro, older than Albargeloni, the latter treating seventy-three documents in his work, the former only fifty, and yet Rabbi Samuel lived six hundred years after Albargeloni. 2 Comp. Wertheimer, D^IDW 'uj, III, Introduction, 1-3. THE HALAKIC LITERATURE 177 ANONYMOUS CODES OF THE GEONIC TIME. The transition from the works of individual Geonim to the collective Responsa compendiums is formed by a number of writings, most of them originating near the end of the Geonic period, which are composites made up of Eesponsa and one or another of the kinds of works mentioned above. At the head of them is the D^wn "HD DWiONl, written probably in the year 885, which has come down to us in several recensions. Its purpose is methodo- logical as well as chronological. It, therefore, contains a chain of traditions from Moses until Rabbi Judah, the compiler of the Mishnah, an array of data about the Amoraim and Saboraim, and also a number of methodo- logical rules for the use of the Talmud, especially its application to the decision of practical cases. The recensions at present available are such a medley that it would be unfair to charge any writer with having perpetrated it 1 . Obviously, the text was badly used by glossators and copyists. In G. S., p. 322, proof is adduced showing that a piece of the D H N11N1 D'wn 'D had been taken verbatim from a Responsum by Rab Amram. This suggests the conjecture that the rest of the little volume is made up partly of Geonic Responsa, partly of the niyioc? current in the Academies. These " Traditions " are mentioned by Rabbi Saadia in two passages in his commentary on Berakot 2 . His references to them give us no specific notion of their character, but the word 'fm shows that they were in writing and probably consisted of old 1 The Taiinai in and Amoraim are mixed together confusedly. 2 6 a (perhaps a gloss) and 12 a. What Rabbi Saadia tells us of these rnyoo in the latter passage, called an enigma by the editor, seems to me an intelligible remark, only it has happened in the wrong place. It refers to Berakot, 37 a, and puts the question, how Rabbi Akiba came to use the words -\i -raw rtn to his teacher Rabban Gamaliel, unbecoming words according to Baba Batra, 158 b; he should have said "p TaiN irrm?. Accordingly, we should read TDN im w^i, instead of the meaningless ION iny: '2i iNVi. I N 178 THE GEONIM explanations of difficult passages in the Talmud *. Eabbenu Hai, quoted in >V3, ed. Luncz, XII, 320, speaks likewise of DnMPin bw nyiB>, apparently referring to post-Talmudic traditions. An extensive collection of Geonic Responsa and extracts from the codifications of the Geonim was called 1DD niyivpcn, which was compiled at Kairwan, perhaps during the lifetime of Rabbi Hai, certainly not long after the extinction of the Gaonate. This book was one of the chief sources from which the German authors of the twelfth and the thirteenth century drew their knowledge of Geonic literature. The opinion of some scholars, that Rabbi Hananel was the author of this work, cannot be defended. Indeed, if anything can be asserted positively, it is that Rabbi Hananel was not the author 2 . The pan IBD was a collection similar to the one just mentioned, and it probably belongs to approximately the same time and place. Whether Rabbi Hefez ben Yazliah, the correspondent of Rabbi Hai, actually was the author, seems to me not quite certain 3 . An argument against his 1 "The books of the Academy," of which, according to the statement of his pupils in their commentary (p. 36) on Chronicles, Rabbi Saadia made use, do not mean Geonic writings, as Harkavy holds, in Samuel ben Hofni, 28 ; they were books in the library of the Academy, and have nothing to do with either rwioizj or niyno . 8 Rapoport in his Biography of Eabbenu Hananel, note 36, called attention to many differences between the mrispan 'c and Rabbenu Hananel. His conjecture that the 'pon 'D was begun by Rabbenu Hananel and then elaborated and worked over by another hand is a theory faute de mieux. The passage in fw, 1, 167 a, to which Berliner in bMin 'nsD, 20, refers, is to be emended to read ':n m instead of ':n 'm, for, as appears plainly from the quotations taken by Berliner from the "w, the author did not ascribe the 'port 'D to n""i. This also disposes of Berliner's statement that the mjrapDrt 'D was in part arranged according to the treatises of the Talmud ; moan J"D in this passage of the i"i refers not to the 'port 'D but to bN^n , contains, in addition to the Geonic Responsa, decisions by Rabbi Moses of Cordova, a contemporary of Rabbi Sherira, by his son Rabbi Enoch, and his disciple, Rabbi Joseph ben Abitur, and by Rabbi Meshullam, the last three contemporaries of Rabbi ,Hai ; and also decisions by Alfasi, who was twenty-five years old at the death of Rabbi Hai. Likewise in the Collection p"3 no authors younger than Rabbi Hai are named. We now have two sets of facts before us. On the one hand, we have seen that the impulse to make Responsa Collections cannot be proved to have manifested itself earlier than the time of Rabbi Hai. On the other hand, we have seen that in the three Responsa Collections instanced, certainly among the oldest of their kind, no younger authority than Hai is mentioned, if we except Alfasi, while the non-Geonic authorities mentioned are contemporaries of Rabbi Hai outside of Babylonia. This would seem to make it im- possible to declare the Responsa by non-Babylonian authors in the Collections as later additions. Or, we should owe ourselves an explanation of the fact that they include no Responsa by scholars living after Rabbi Hai. In scrutinising the arrangement of the Responsa, two points can be fixed upon which seem to have been of significance to the collectors. As these two points are incongruous in character, the result is that there is not one of the Responsa Collections executed according to a consistent plan. The two points are authorship and related subject-matter. The questions submitted to the Geonim were either dubious cases of practical bearing, hence unconnected one with another, or dubious cases coming up in theoretic study which were more likely to have some relation to one another, especially if their common point of departure was a given section of the Talmud. An example of the latter class is afforded us in the fragment published in 1 86 THE GEONIM G. S., pp. 328-36, containing a number of Responsa by Rab Amram on rwv nia^n. These have not been arranged in the order given by a later hand. The order is original with their author, who obviously was requested to explain and codify the laws on nTX given in the fourth section of the Talmudic treatise Menahot. This example shows that it is not always safe to attribute a logical arrangement of Responsa according to subject to the collector. It may be the work of the Gaon in the same sense in which he is the author of the Responsa themselves. However, it cannot be denied that the Collectors were particularly concerned with arranging the matter at their disposal in the most logical manner possible. Isolated portions of the printed Responsa Collections, and some of the fragments published in " Genizah Studies," have been spared the systematising hand of the collector, but no complete collection known has been similarly fortunate. This lends peculiar interest to the Responsa lists published in G. S., pp. 56-71. Their authenticity can hardly be doubted, guaranteed as it is by the name of the Gaon, the name of the addressee, and their checkered contents. A comparison of one of these lists with the printed Responsa Collections reveals how imperfectly even such among the latter as are supposed to have reached us in their original form have preserved the initial order in which they were arranged. Of the thirty- two questions on pp. 67-8, below, addressed by Rabbi Jacob ben Nissim to Rabbi Sherira and his son Rabbi Hai, there are but two and these two in widely separated places that occur in the Collection published by Dr. Harkavy, which he describes as having been planned on the basis of the duplicates kept by the Geonim in Babylonia. But this pitfall of not being able to re-establish the original order of the Responsa is not the only one. Care must be exercised not to mistake decisions by European and North African scholars for decisions by the Geonim. This applies particularly to the large number of anonymous THE HALAKIC LITERATURE 187 Responsa in the Collections of Geonim, not by the Geonim. The quotations from the Responsa of the Geonim in the older Halakic literature are an excellent guide. To the authors of this literature Geonic Collections were accessible, more accurate and reliable than ours. But these same authors have a far higher function to perform in the study of Geonic literature. Their main value is that they knew a multitude of Geonic Responsa that have come down to us through no other channel besides. In his Mafteak Miiller has made the first attempt to bring them together, and as a first attempt it is most satisfactory. But he has not dug out even the half of the hidden treasure to be found in numerous works, beginning with Rabbi Hananel and extending down to Caro 500 years later l . As an exemplification of the importance of Halakic literature for the study of the Geonim, there follow three lists of Geonic quotations in the works respectively of a Spanish, an Italian, and a French scholar, parallels in our printed Responsa Collections and in other sources being marked. The first list contains quotations from three works by Albargeloni as the representative of the Spanish school : SPANISH SCHOOL. ALBARGELONI. own n*o rn'oa NTT im 17 a"n n*? im 17 ''jp E*n ;n"nn*ir naitcnai 17 n*: ,'a Vwn "jpcrtjQTnw im 1 7 1*3 ,'a Vcn i P D n ornNcc 18 *D *)"} jwa 3 WVTC 10 2 'nn 13 nvw: 15 I'D 01*03 niwrai 17 1 David Kaufmann, in the Bet Talmud, III, 64, published two Responsa by Rabbi Sherira and Rabbi Hai from a MS., not noticing that the same are to be found in the Responsa of Rabbi Solomon ben Adret, V, 25 a-b, no. 121. a In his commentary ? i88 THE GEONIM b'p 'n a'n n op 1*3 O 1*3 O 25 a 3/So 25 a y-iD 25 b S"-ID i": I, 7 t3"ttJ I, 9 c"o I, 10 \e"izj 12 d a"n n": b"a n"o 'n ; 12 b a"n 13 a a"n 'n ; 12 125 132 i35 mpiDDai 135 Drni^Dd 135 pa 136 Npccai 136 jwa 139 i43 H3 i43 i43 ma bn 143 ana bi 144 NpDsai 144 nbstrai 148 nai^nai 149 Cnbi^^rTiDT 149 maittnai 150 i53 i54 arm 157 1 60 160 172 173 !74 VL: myo 176 178 D: 178 182 'sn 182 rrnno 183 n 183 ':niD3 183 nos 185 DTOO 189 190 190 I 9 I mbnaai mpicDai 193 J 93 195 J 95 196 199 a"o o ioa 'n ,-p-W a"p ,a*n 13 a a"n 13 a a"n ;n*y b'a *'P nnso "op n"np 18 snnan 19 jwab 26 anbwoc 26 26 26 27 27 34 37 38 39 39 mbnaai 39 rr 43 jiwb 46 "n 4 8 49 49 52 53 56 57 64 jia 64 nawiai 65 Npcoai 66 naiirna 74 74 76 naiirrai 91 jwan 94 nipiDEai 104 pDDai 109 j:no 109 pab no im DribNuwi 114 im 114 *^wn ii^ "Nn 116 124 mi 'aiTyi 124 nbNTCH? 124 Tmr 124 In his commentary. 2 Comp. ffi'o, I, 18-19. THE HALAKIC LITERATURE n 260 niwnai 199 I, 69 Vow rrnrra 264 I, IO c"lZ7 cite n 199 29 a y*~\r> '10102 266 3*0 b*3 NpD'Bl! 2OO >'Nn 267 nrrajnai 200 rrnno 267 3"3 V3 "xn 200 'nn 268 rnyo 20 1 rvnyo 208 l"3 b's C1O3? 2O2 NY-It? 268 i*icp n'c "nrr 203 'nn 268 1*7 n't? D1O3? 203 rrnno 269 '3 'n ; 13 d 3*n mpicci 204 mpicD2i nm 270 nos 204 nipiDEai 270 0*7 n*o "n 204 28 b y"^c '*3TD: 270 I, 5 "c '3T1B3 204 28 a 3TI3O 'ncbo 270 'a 'n ; 13 d 3*n nib 11321 mpicsai 206 "Nn 271 D^Qyc 'n\D 29 a y*iD 'N21Y23 272 I, 3 ffi c Mnnn 207 n 275 3*3 'n ; 12 c 3*n mb nsai mpiCDai 211 ?lYnDl mj?D 275 no^ 211 'n 275 rpnrra 211 ?vvnD3. mro 276 1*0 ,'n ,iac 3*n niVnan 211 "Nn 276 n": V"3 w n 212 ^l^Q/D 270 'HH 215 OYoy 277 26 b r*-o 'N311E3 2l8 NH 277 B": b*3 ^Mn 222 3*2 b*3 "n 278 "j"p D*n nnbxnjd 235 ii a J?*TD aiVo 281 B*3T n*CJ Drt^cci 235 3*3p 0*103 >*prDa 281 23 a ; i y*tt? on'rNii^m 235 i"?p D*n pN3 287 23 a ; 2 y*c cnbucci 235 n 288 23 b 53 y'tu nari 236 "xr? 288 23 b ; 6 y*c orfowDi 236 'l '*3 ,'l y"lD n*3YTC3 289 25 b ; 18 y*ic DnbxiCd 236 J1N3 1 ) 289 26 a ; 20 y*c in 337 n^'w 3*3 ITDD 301 26 a ; 21 y*o cn^Ncti? 237 "NH 304 26 b ; 27 y*\D mi 238 |1N37 306 2 7 a ; 35 y*c rtncfi 238 JW3 310 27 b ;37 ?* Drr?^TC^T -| ( ' 3*n rvofci 310 J1N3 240 n*sp 3*3 Drtwunoi 316 'n 248 n*sp 3*3 Vicii 316 MT1C 248 NpDDTI 317 "NTT 2T 249 'vn 337 vanes: 249 'n 339 ? TO1TS2 "Nn 252 p3 341 rrnno 253 pH3 343 "xn 253 p ,-piy "Nn 347 156 n"3 nn 253 103 n*3 "xn 257 1 Are the following five quotations taken from R. Hai's commentary on Shabbatf 190 THE GEONIM I'D ,n*3 ir'p nnro ,n*p 3*n 82 8 4 86 86 ?2r\vh 87 sn p 105 124 125 126 126 n psmp |8< 24 17 18 24 53 71 76 78 "n 114, 128 "" ^T, 138 6 Nn 149 8 nn 154 1 66 1 66 262 II, 44 Vim II, 4 6 ->"trn '3 ,'n ,3*n n"3 D^PI DTO 25 26 28 74 85,86 HTT 103 The second list illustrates the Italian school by references to the Cp^n 'fyytf by Zedekiah dei Mansi. In this list and the third, special devices have been adopted for two purposes. A cross ( + ) indicates that the Responsa cited are not by Geonim, but by old French and Italian authorities, called Geonim by courtesy. Again, when there are doubts as to the origin of the Responsa with actual Geonim or Geonim by courtesy, a query is put against the citation. 1 In his commentary on Baba Batra. 8 In his treatise on Witnesses ? 3 Compare Halberstam's remarks. * Comp. Tosafot on Baba Batra, 10 b, catchword D R. Hananel's commentary. 5 Comp. above, pp. 171-2. * In this collection, as in Harkavy 199, the responsum is ascribed to K. Hai. \vho quoted it from THE HALAKIC LITERATURE ITALIAN SCHOOL. ZEDEKIAH DEI MANSI. NT1TD 13 i np a a 'Mano: i n*ap :*n TOS 13 ? pNabi i + c<:iNan 14 D':wan 3 7 c'awab 14 22 b DTID 'Kn 14 + c'awab 15 ? 8 D'3wab 16 mro 16 Vrp a*a 'WIITD: 3 + a pai 6 .en ,TTI? rtxco 7 1,31 ;59 XD'C D'siNab 17 1*0 a*n D':wa^ 18 p*oi 19 II, n, 12 Vim 'n 19 'MrT 8 9 wirr 8 58 a CTID "MH 20 'D rt*a c':iwan 21 ' T nnco "Mn 22 + * D':ian 9 47 b maia ,C"NT 'n 9 9 mj?c nat 22 58 a DTID n 22 I'D Va a^awan n ? pa^ 22 2*S D*n 1>NC3 II pwaVi 22 + n^isa? 12 + pwab 23 pnab 23 4 b-s a y'nc Dioy 13 H"DT 'n ,a"n ptcna 13 58 a CTID B ':no: 13 44 a cmo pis ;HD 25 4by*x5 rnro 13 1 Not a Babylonian authority, the prayer Dtob is of Palestinian origin, comp. Ratner, T*in, Berakot, 199-200, and Tur, Orah Hayyim, 46. 2 Comp. G. , p. 273, n., where cbprpi is used by R. Hai, and R. E. J., LIV, 195. s Is hardly a Babylonian Gaon, the explanation shows the influence of mysticism ; comp. Parties, 57 d~s8 a. * German authorities, as indicated by the name rrnrr -\"i pns' 'i. 5 The text of bn'ac is to be amended in accordance with Parties and I'tn, I, 52. * Italian authorities ; R. Daniel is the brother of R. Nathan b. Jehiel, comp. above, p. 179, n. i. 7 Comp. Ratner, T'ITW, Berakot, 51-2. ' Comp. Tur, Orah Hayyim, 66, and the authorities given in Bet Yosef, ad loc. 9 This passage is undoubtedly of Geonic origin ; perhaps a literal quotation from R. Hai's Seder. 19 Comp. Tur ; Bet Yosef; Orah Hayyim, 594, and c"c, I, 30. 192 THE GEONIM a*D Q*1T33 ^*^SJ^T A C i_ ratsi 45 'a a'n wrrca 25 c'^inan 46 35 b y'-c l pa 28 I'D 'n ,a'n nVma 46 60 d DTIS "n 28 s'p n'c c'aisan 46 II, 42 Vcn w n 28 II, 9 b i* D'aisan 46 19 a f"^3 cibtD ~TT 28 [pis] pa 47 jiwai 28 i*op Q'TOJ NTTC3 48 37 b y'-ic Din? 29 s TI n c D'aiani 49 ai'in D"n mix iv^ pN3i 29 Vp E*n ixarvQa 49 a pa 30 25 a r'-c 9 DlVffi 50 n*: ,a"n 'anca 31 25 a y'-c **wrfl 50 3 Dior 32 26 a y"TC HDIOJ 51 29 a y"~c mo? 33 25 b y'lD m mro '-i 51 + Jisabi 33 'K311B3 104 TD'D 'n, a'n nobn 67 's nrtiD 'rnSna rnrforoa 106 + * D^IWI 70 38 b DT1D nn 1 08 ? D'3inan 71 win 1 08 a* 1 ) o*ioa craixan 72 41 C DT1D wi 109 + pwi 77 ? D'3iNan no + pai 77 D'jwa'j na n*D nnco xmn 1 78 I, 3 w*c ''ITD^D 114 ? D'aiMb 79 I, 43 bocN "Mn 115 4- 8 pwai 79 D':iwan 115 + pxai 80 II, 99 C'UJ n 115 TL"?P D*n 'Miin 4 83 10 a a'n mabnai 115 pai 83 1*3 Va o~my 116 c'Vp D*n 'hnin 1 83 9 a a'n 'nin 116 ? D':in 83 I'o Va D':wab 116 tD /p D n vtiirp 84 1*3 ninbHrc 'KHN Il6 pai 84 I'D 'n 59 d a'n no'vn 117 B'SW, a'a '*nin' 86 D':ian 117 I'D Va *n 88 n'sp n'i TITO D'3iai 118 T*n n'c "n 89 8d a'n 'MTinr- 120 pHJI 90 rVo Va 10 'ITD'JD 120 4- D'3iwn ga i'a Va n^iNan 123 *OT n'w D'3inan 94 n'o Va '*TCT> 126 pMabi 95 n'yuj "niman nyra '3Dn p 126 B*p 1*0 pa 97 II, 56 i*cn D'jiMan 126 30 a y*TD Dib 98 I sn l DttJ < i 127 V'p 1*0 D':i:n 99 n'cw rvoian nyw <:cn p 127 30 a y*TO cibc 99 B y\c manan *"V?TD ':cn p 127 31 a S'ID 'Hinioj 101 7 c a'n msbm 127 31 a y'iD moy 101 1*3 'n 7 a a'n nisVm 128 Rino ioi "> 53 a T 1 ^^ * '01 f]DV 103 1 Comp. Twr, Graft Hayyim, 291. 2 Comp. ibid., 382. * Comp. Bet Yostf, Graft Hayyim, 301 end. 4 The view ascribed to R. Hai in Va is opposed to that ascribed to him in bn'atj. 6 Comp. Miiller, Mafteah, 80. ' Comp. above, p. 147. 7 Read sax 10 na psa r\cv '. 8 Read mVna mibnai mnVwca and comp. a'n, 9 b ; ed. Hildesheimer, 67. 9 Ibn Gajat quotes it on the authority of R. Hai, but R. Hai uses the words 10 Comp. also Coronel, 57. 11 Published in Bet Talmud, vol. III. I 194 THE GEONIM isn ,-piy D1/Q7 ^^? loo 33 r'lD oioy ^37 n 166 N' 1 ?? ,a*n C'jiNan 140 + D:wan 167 II, 12 bl3TT jwai 140 *op 'n ',a8 a'n rro'mi. 171 33 b y'lD moy 141 o'n n*i TIB [pis] pns jns 171 1*3 a'n 1 D^wan 142 ? 7 c'2wan 171 ib*i n'c NrQTTOO ^44 I3*vi 'n ; 15 c a'n 'unn 1 1 72 ab'i n'tD Mm^QD 1 44 -,.- ^'.. 1 p U 1UJ *c':iwn 172 1*3 mnbtj 'Mn 144 + D':wan 173 D':wan 145 i^p 'n ,a*n '131 na'm 174 + D'awan 145 a^wn 175 36 a 3?*iD DTO3J 147 rrnwn 175 M'm ,'n ,a'n 'nn 147 4- o^ixan 176 n'n b'a 'MTin 147 a*7 ,1D*TO3 [pis] pm jns 176 142 = 'N21TO3 148 I'DI n*c pis p3 178 *3>'lD moy 1 50 I'D? c'n n Nni'D'Ju 170 I'cp D'n ^n 152 n 179 pab 153 + o^isan 179 CnbXTTd 153 'D mnVtD n 181 *pw 155 4- D'JINJ'J 182 t*Ep D*n 4 TWsn 156 DT01 S 184 II, 109 Sj'tiJ 5 nos 156 n*Dp a'a D'^Nan 197 II, IO9 M' C *vn 156 o'p 'n ,a'n n^iNan 201 37 b y'lo 8 moy 157 II, 53 c Tray 1*31plS p3 201 n'p n'o D>:ian 158 40 b JJ'TD DTiwan 202 Soadyana, 59 n'TTD 161 i*op 'n ; 30 a a'n rnsbnn 202 I'D H'TD "Nn 161 Denary 202 + m3in3 161 II, 146 nn'n "Nn 203 l*Vp 'n ; 22 b a'n noyrn 161 + D^iMan 203 a's n'rc M n 162 ? D^iNab 209 D'jiNa 1 ) 162 II, 26 c'o ni3itrn3 211 E'I n'c C'31NJ^ 162 t'n n*c 10 'i3i pna 211 onrc Mn 162 II, 24 j*TB pwb 212 n*:i I'D '131 n 163 pnai 212 n'ci n' n'jwan 164 + maitDni 213 a*Di n's? D'3iab 165 1 In bn'axr, 148, ascribed to R. Natronai ; comp. also Tur, Yoreh Leah, 401. * Comp. Marx, Untersuchungen, 20. 3 R. Samuel b. Hofni? Comp. n* 79 and Mafteah, 171. * Comp. above, p. 43, note. 5 Comp. above, p. 43, n. 8 Comp. above, p. 147, n. 2. 7 I doubt whether this Responsum is Geonic ; notice especially the use of the word D33i. 8 Comp. it'tt, I, 6. Comp. above, p. 153, n. i. 10 Read pm instead of prtjr, and comp. above, p. 143, n. i. THE HALAKIC LITERATURE 195 II, 120 d I'M 8 %.*-._< **.* 'Kll~ 207 >*op 'n J3b J*" mo/n 753 217 N11J? jr,D rr-rrc 267 y'-in 'n ; 137 c a'n 'HTn^ 217 ?YYnD n 267 1 0*3 wan 217 i*c n*S ,v*tr\ xn 267 + D'awan 220 220 = D'J^Nan 267 i*Dp n't? D'jinan 220 pxaS 268 >*Dp 'n 536 a a'n niD^na 227 6c a'n nobn 268 x'cp 'n 536 a a'n NT'p 228 44 a DTID Hn 268 pnai 229 44 a DTIC '131 DTay 268 i*yp 'n ;s6b a'n nia^n 237 267 = nn 269 n*3wan 242 45 a 3?*iD D'31Na 269 Dn^MurTDi 243 maiTDna 270 + tyainan 246 10 Dioy 270 ? pto^ 247 + D':iwan 271 * MT'p 248 '.. P i ^ jwab 273 i'o D*n pis jno 248 n*o ,^3 voim 276 + D':wan 249 I, 25 TT'TD vn 278 N ^D npn * rnyD 252 InjRM 4.2 u, U? Q n 1 iiu 20 1 D'awab 254 46 b J?*TD moy 281 II, 2 Vou? *pnan 254 I, 42 c'c n 281 + D'3ian 254 i*op rrvs' 'D 'jVya ma'w 282 a'sp o'n 5 a':ian 254 n't iyo toin 'T M D'3iNan 283 D':isa 257 281 = mor rnrD 284 n*:p O'TD A .... O H3TU? 250 n't I'D ,fn ia nrno en 285 III, 9 c DTI anan >*n 258 I, 44 w*w 19 D>:wan 286 + craiMan 258 I, 43 m'rc 15 "Nn 287 n'o a'a mwan 259 o n^ijTD 207 I, 22 C'TD pMab 259 50 a y'TD Mwan 288 I, 21 tt?*1T pxan 259 50 a y'-o 'tOltll 288 i D mrv?No KTO 260 Q _*_ ^^'^w*. T a D u ioa pen: 288 n'jyn n 't pnab 261 1'cp mn'iwD 'Hnx ago o^awan 261 + D':ian 295 D'3iKab 263 niTDi n 295 i* n'c 7 p-n? pa 263 + I4 o':ian 296 I, 50 VOTI D'swan 266 1 Comp. j*c, II, 108-9, and 0. S., p. 185. 2 Not found in our two versions of a*n. 8 Comp. Parties, 48 a mro n ^ ^D^, accordingly not the Gaon R. Saadia. 4 Comp. also Alfasi, Ta'anit, . . . and Vru'c, 261. s Comp. 0. S., p. 263. Comp. above, p. 104, n. i. 7 In n'c ascribed to R. Hai. 8 Comp. Muller, Handschriftliche Jehudai, &c., u, and ff. S., p. 263. ' Comp. D'ny, 252 and 288. 11 Comp. Jerusalem, VII, 167. = n 'i, comp. c*c, I, 42. 14 Comp. Parties, 44 c. 2 19 Comp. above, p. 141. 13 Comp. (?. S., a6i. 196 THE GEONIM I, 42 TD'TO I'T 'n ;42b a'n 22 C DTID ,f"-\ 'n 541 d j'n 20 a ; 3 X>*ID 'n ; n*T 'n 542 a a'n 22 b DTID 22 b ; 13 y'c 22 a 55 y'TD 22 a 53 y'\u 22 b ; 12 y'o 22 b ; ii y'ttj I'p 'n ,a*n i'p 'n ,a'n i*p 'n ; 24 a a'n n*n ">'a 1 ,:'a n*DT ,n's ,T>B 344 344 345 346 346 347 347 347 348 35 D'JiNan 350 35 2 353 354 355 355 357 357 360 362 362 364 365 366 12 'Nn 370 pa 370 pwa 371 wirr 373 W (TO 373 nos 374 Dl^tD 374 Tirr 374 'XTirr 374 pwab 374 NYV 376 48 b y'no 44 d DT1B I'D n'? D*p 'n ;32b a'n c'p 'n ;32b a'n a'cp 'n ;32d a'n I, 89 ID'ID I, 100 TD'\O n'Dp 'n ;ssd a'n II, 107 TD'TD n* n'sj 50 b S'TD II, 42 d T1EW QTJJ ^1TQ3? 50 b y*~[D 34 c a'n I, n6d n' II, 42 n'xcn n'jaa FJID f]'n a'c mail f]'n n"? 296 297 299 299 303 3<>7 rrobn 308 308 309 3" 3M 3H rabc 315 3i5 317 322 323 325 327 328 328 rnyo 328 >n 328 'TO^D 328 328 338 330 330 33 331 333 333 B pa 333 "n 340 *wnt$ 34^ 1 Not in the Halakot Gedolot. 2 C!omp. Hildesheimer, ad loc. 3 Our versions of the a'rr read differently. * Comp. G. S., 310, and n', I. 113 e. s The Seder is also the source for 'I/tur, . . . I n'c has only an extract of this Eesponsum. 8 R. Amram in his Seder, 51 a, differs from this view. Comp. above, p. 194, n. i. " The author is E. Hai. II This is the Eesponsum to which reference is made in bn'ac, 257. u Comp. Vow, II, 123. " Read with y'c : * Comp. "jrr'iur, 216, 270. THE HALAKIC LITERATURE 197 I, 114 b i*w n3*pn 'n ; 128 a j'rt 1'3 ,3*3 TD'3 .*._ .'_ 3 3p J n n'?pn 'n ; 130 a a'n -En p 393 mwn 395 m3"?n 395 + r;iNjn 398 3'D D'n fwa 398 II, 26 d ticy msbna 399 II, 86 "JOCN nos 399 3'D C'n ':wan 399 3*D C*H )'31M3H 399 3'3 ,3'n '3iwn 399 ni^Sn 400 3':p n'c '31H3P! 400 I, 46 ^'J.-l ]1NJ 400 I, 47 I'tcn 6 "xn 408 ('Mb'rr) 383 = ':i3n 408 258 = K-PP 376 n 381 38r 38i 382 <:wan 383 '**n 3 8 3 cite 383 383 383 ITOS 384 F|cr 384 ntro 384 rrunr 385 n 386 The third list illustrates the French school by means of the no'l Time ; quotations from y*"iD not described as such aro disregarded. FRENCH SCHOOL. 25 a y'TD 25 a y*TD 25 b y'iD 25 b y*iD i': a'n II, 40 n'cn 28 a y'iD 29 a y'TD 1 I'D rfoo ,r|*'T 29 a y*TD ii a y'TD 81 81 81 mno 83 83 87 9 1 9i 9i 93 94 94 98 99 '3 'o ,y'io 14 b y'-o 4 b y*-c 1 1 a y*io 1 1 a y'no 24 b y*io ten ,-pny 14 b y*TD 's b'j 'i 'o ,y*-c 'n 'o ,y*TD i i a y*TD 5 cwa 8 IttTTO 8 :TB3 23 VII 23 :rre 23 rroo 23 mas 25 26 20 S^ 50 50 50 1 In n'c, 153, ascribed to R. Hai, but i*cn, I, 45, agrees with brt'ac. 2 In the 'Irtwr, ascribed to R. Hai. 3 The words rrmrrci pi are in the wrong place, they belong after mirro rroru. 4 Comp. above, p. 151, n. i, and VOCN, II, 86. * Comp. o'w, I, 5. ' = Marx, Untersuchungen, &c. ' DC ? 8 Comp. Hurwitz, ad loc.; there can be no doubt that this mystical passage is not of Geonic origin. Comp. also O'TO, 260. 198 THE GEONIM I ,iT3yn ,Ffn 35 b J?*TD 35 a r'TO 35 b y'-o 'n ;44a-43c a'n 20 b 59 y'rc n'o ,y a I, 14-15 Van 3*y o'loa N*op 'n a'n T 3 mnVNttJ I, 99 tt'ro n*3 'n ; 7 a a'n II, 103 TD'TO op 'n ; 29 d a"n n'op 'n ,a"n o'p 'n ,a'n o'p 'n ,j*n 'op 'n ; 29 d a'n 26 b y"-\o "n 231 D'3wan 104 JH3 232 ^bn ,IITS n^'xan 104 17'OTJI 233 Vp a'n D'3wan 105 niTTC' 233 lb*p D*n on'jSffiiDi 107 nine* 233 i i a y'TD anoy 108 D'Oisan 234 30 a y"~nn 272 36 b S'ID unas 208 D':ian 276 37 a r*TD 'K31TE: 208 rrobn 278 I'D 2 mn'w 211 1 D'3iNan 278 37 b y'lD jn: 211 D'3-wan 279 TO*p c"n 3 D'3iNan 212 'Tn> 279 37 a r'no mop 213 12 nos 280 * WVP 213 '>nra 281 II, 109 TO*tD 6 n 213 nyi 281 n'' 'o r'-io ntco 214 TW 281 43 b r'lD 'N31V.C3 228 D"\D2? 284 43 b y'-o 6 onoy 229 1 Not Geonic, conip. above, p. 193, 1. 27. a Not a verbal quotation. 3 The author is R. Natronai, comp. above, p. 43, note. * Read wrros in agreement with c*\c 1. c. and other authorities. 4 Hurwitz is mistaken in maintaining that R. Nathan in his 'Aruk, s.v. "?ao, ascribes this view to R. Zemah. 6 MS. S of the Seder has likewise oro and not orn as printed text. 7 In o*ioa : 'Wnsa ; comp. above, p. 149. 8 Comp. bn'aw, 162 ; if not for this statement of brVnw I would be inclined to ascribe this Responsum to R. Hai b. David, the contemporary of R. Hilai, and not to R. Hai b. Sherira, who according to Mordecai, Pesahim, 583, holds an opposite view. ' In n'tD ascribed to R. Matthetias. 10 Read pis jro. ll niNDTan ? 12 Read KS'H ; in xc'u?, V, 100 and n'c, 102: nos without the name of his father and accordingly one of the Geonim. THE HALAKIC LITERATURE 199 JITS' 1 D W11E3 1 1 a a'n n'o ,y a n'o ,"5*a Sab *TD 22 b ;n y'ffi 69 d a*n U> 37 "J12BK 1'opn'n ;i26d j'n III, 48 532 mro 556 nrrtna 587 n 591 608 612 614 623 am 623 624 624 1 c':wn 637 12 N-riiu 640 IS 'nn 640 wirr 642 13 DTO3? 642 rrobnai 644 D';i*on 652 740 754 755 "'NTirr 787 ?36a y*iD 537 0-38 b a'n a*:p-*i*op 50 a y"-\o n'o 'n ,a*n 46 b J?*ID a'ain 'n ,a*n n*cp-n*:p I, 24 T*ujn !, 24-35 n'cn a'p 'n ; 34 c a'n 50 b y"-o 42 b 2?*1D 3'in a'n 29 a 3?*1D TVS' 'D wnBa 'anbMtrtn 353 356 D^:wan 365 WYC3 374 382 387 ''Itto 387 rriyo 388 409 414 416 *ni3 1 jna 423 433 435 435 'TO^D 437 pns 440 'onor 445 Niirr 458 DID? 463 519 1 The passage of the Seder is quoted literally on p. 202, and it seems there- fore that the source for the Responsum given on p. 353 is another one than the Seder. * In n*ic ascribed to R. Hai, but comp. above, p. 195, n. 8. 3 In i"n anonymously. 1 Neither in j"n I, nor in j'n II. 5 Comp. G. S., pp. 309-10. Not in the printed text of the Seder nor in the MSS. 7 Comp. Hildesheimer, ad loc. ; our text reads differently. ' Comp. Albargeloni, rrvs' 'D 'c, 177 and 341. 9 Comp. n"c, 6, where this Responsum is made use of. 10 In his code, comp. above, p. 165. 11 Comp. G. S., 250. 12 Comp. Auerbach, in his commentary on Vorc, II, 82. 1S Comp. above, p. 151. u Neither in a'n I, nor in a'n II. 200 THE GEONIM THE IMPORTANCE OF THE GEONIC RESPONSA. Defective and incomplete as is the state of the Responsa transmitted to us, so must be our judgment of their value. From Rabbi Shashna, about 680, until the death of Rabbi Hai in 1038, about eighty Geonim officiated as such, but barely more than a third are repre- sented in our Responsa literature 1 , and yet it is hardly open to a doubt that, if not all, at least a large majority of them must have given written expression of one kind or another to their views upon religious questions. But even of the Geonim from whom Responsa have come down to us, we know only one side of their activity, and of that side not enough to furnish grounds for an impartial and adequate judgment of their place in Jewish develop- ment. In the Responsa Collections available at the present day the Geonim appear as Halakists exclusively 2 . Even the few Responsa that deal with Haggadic material touch upon it merely in the course of explanations of Talmudic passages. Thus what we know of the Geonim in relation to the Haggadah is not their independent view, but only their activity as commentators. And yet it was precisely in the domain of the Haggadah, in other words, in theology, religious philosophy, and related subjects, that the Geonim made no attempt to harmonise their views with those of the Talmud; their purpose was simply to explain the Talmud regardless of their own predilections. "Know that we are not, like some others, in the habit of explaining any matter apologetically, in contradiction to the real 1 Almost all are on record in Miiller ; the only ones to be added are the two Geonim by the name of Kimoi, whose Responsa are found in an anonymous Halakic treatise published in J. Q. R., IX, 681-761 (comp. above, p. 104, n. i), and Rabbi Hezekiahben Samuel, who, to be sure, was not actually a Gaon ; comp. above, p. 7, n. i. 2 p*a, 15, is surely not a Responsum, and its Geonic origin is very doubtful. THE HALAKIC LITERATURE 2OI meaning of him from whom it proceeds. We will there- fore expound to thee the opinion of the Tanna, his real meaning and his true purpose, without pledging ourselves for the correctness of the assertion made by him." These words of Rabbi Hai l , who, in opposition to Rabbi Saadia and the philosophising school that followed him as its head, insisted upon an unbiassed explanation of the views of earlier teachers, characterise not only his own intel- lectual attitude, but also the spirit prevailing in the Academies so long as they remained untouched by alien influences. At the same time, his words make apparent how difficult it is to reach a knowledge of what the actual views of the Geonim themselves were. And yet, if any doubt had been entertained as to the theological trend of the discussions in some of the Responsa of the Geonim, it would have been dispelled by the list of Responsa printed in riD7K> J"6np, 69-70, containing twenty-eight items, almost all of a theological nature 2 . In that batch there were Responsa on the translation of Elijah and of Enoch, on Shabuot as the Feast of Revelation, on the suffering Messiah of the fifty-third chapter of Isaiah, on the death of the Messiah referred to in Zechariah xii. 16, and on many other interesting points, not one of which has been preserved in the Responsa literature now known to us. A com- parison of Responsa lists in nobc> r6np with our available Responsa Collections, leaves no room for doubt as to the guiding principle adopted for the latter. It was plainly intended that they should consist of Halakic and Talmudic material exclusively. This is the only possible explanation 1 "j*:, 99. The expression men 1 ) is probably an imitation of the Talmudic x-nnx ra Sy men 1 ? in Git tin, 17 a ; comp. 'Aruk, s. v. vnn ra and F]n. 2 It will not do, of course, to assign all these Responsa to the end of the Gaonate and ascribe them to Rabbi Hai. In fact, the list is headed D'yihO 1 ?. It is noteworthy that the first list, a'a-'i, deals with difficult chronological problems in the Holy Scriptures, some of them being the data used by Hiwi Albalki as weapons against the authenticity of the Scriptures. Dr. Poznanski in his essay on Hiwi, Jnn, VII, 112-37, makes no mention thereof. 202 THE GEONIM for the phenomenon that most of the Responsa of Halakic bearing recorded in the lists just referred to have been preserved in our Collections 1 , while those of Haggadic content have disappeared wholly and entirely. Limited thus to pure Halakah, the Responsa nevertheless are of very considerable value. In the first place, they called forth a new species of literature, which in a measure shares with the Talmud the distinction of being the only department that can be described as peculiarly Jewish. Correspondence between scholars existed before Geonic times, nor was it an activity confined to Jews. But Re- sponsa are something more, at all events something other than correspondence between scholars. The Geonim were not requested to give their views upon vexed religious questions merely on account of their scholarship and attainments, but because they were at the same time, in virtue of their high office, the representatives of legal authority. It is true that in an overwhelming number of cases the Geonim appeal to the authority of the Talmud. The Tannaim and Amoraim had a similar relation to the Bible as the only source of law. Yet it would be ridiculous to say that the teachers of the Talmud did no more than explain the Biblical law ; their activity was equally fruitful in elaborating the fundamental law. Halevy holds that, barring two ordinances, there is nothing in the whole of Geonic literature not taken from the Talmud. The same logical process would properly lead to the conclusion that with the exception of the so- called "seven commands of the scholars," pl*n JYIVD jae>, the Talmudic time produced nothing but what is prescribed in the Pentateuch. The Tannaim and Amoraim felt justi- fied in considering their " ordinances and fences " as devised in the spirit of the Scriptures, and the Geonim were 1 Of the fourteen Responsa in the list, p. 72, the following can be traced : 'N in b"a, 55 ; 'a in n'cn, II, 46 ; 'T in n*:, 197 ; 'n in IITQT, I, 25 b ; 'i in nancn nynj of Rabbi Meir of Rothenburg, ed. Bloch, 177 ; 'n in y'tr, 43 b, i ; N*> in E*n, 187; i*' in b*a, 61. THE HALAKIC LITERATURE 203 persuaded of their implicit adhesion to the Talmud in all their decisions. This view taken by the Talmudists and the Geonim of their own activity may be conceded to be correct theoretically, but we are not thereby hindered from recognising it as a fact that Biblical law is not identical with Talmudic law, nor the latter with Geonic law. Every age has its problems, and though the law remained unchanged for all times among the Jews, the laws underwent modification along with the times. Let us consider only the varied development of Divine worship in the Geonic time. Built up on principles laid down in the Talmud, it yet is totally different in form from the service customary during the Talmudic time. Or, to take another illustration, in ?"&> 6? b, 60, we have the Geonic decision that a husband may marry a second wife only with the consent of the first. The aim of the Talmudists, to entrench and increase the rights of women, is evident in a large number of their enactments, and the Gaon who gave the above decision felt himself in accord with the spirit of the Tannaim and Amoraim, though in this given concrete instance he was striking out into his own new path *. And as the rights of women were developed during the Geonic period, so also were the rights of slaves. Thus we have a number of Geonic Responsa that grant liberty to a slave whose master has had intercourse with her. The reasons adduced against the validity of this Geonic decision on the basis of the Talmud cannot be set aside lightly 2 . No doubt, the Geonim were aware of their opposition to the statements of the Talmud taken literally. They felt secure in the other consciousness that they were acting in its spirit. Rab Amram's decision 3 , that it is not per- mitted to take usury from a non-Jew, cannot be authenti- cated by resort to a Talmudic expression. If, nevertheless, Rab Amram forbade it strictly, in any circumstances, he 1 Yebamot, 64 a, bottom, is another case ; comp. toc'n on the passage. z Comp. the Responsa in Saadyana, -76-8, and i*u, I, 164-5. 3 y'tr, 40 a, 20. 204 THE thereby proved the potentialities for development latent in the Rabbinic law. These examples, which might readily be multiplied twentyfold in every department of the Rabbinic law, will probably suffice to give an indication of the real value of the Geonic Responsa. Viewed thus, the Responsa are much more important than the codifications by the Geonim. In the latter, it is the Talmud that is given the opportunity to speak ; in the Responsa it is the spirit of the Geonic times. For this reason, the Responsum became an example and a model for later generations. Their leaders and teachers used it as a means for making the Rabbinic law effective according to the changing circumstances of the times. The Responsa literature, created by the Geonim, developed, as to quantity and quality, into one of the most important branches of Rabbinic activity. The chief distinction of the Geonic Responsa, in com- parison with later Responsa, is that they became of fundamental importance for other departments of Rab- binical literature. The older commentaries on the Talmud, those of the North African school, for instance, are scarcely conceivable without the Responsa of the Geonim l . It may be said confidently that Rabbi Hananel's commentary is an outcome of the Responsa by Rabbi Hai and Rabbi Sherira. They not only served him as a formal model for the explanation of the Talmud, but they contain such wealth of material for this very purpose that to this day they may be resorted to with great profit to the student. And as for Rabbi Nathan ben Yehiel, the great lexicographer, for him and his investigations, especially those into Aramaic word- structure, the Responsa were a veritable treasure -trove. His 'Aruk is in large part a collection of Geonic glosses on the Talmud. Let the interested student compare the frag- 1 There is no telling to what extent Rashi made use of the Geonic writings. The different readings he offers often go back to differences of opinion among the Geonim ; comp., for instance, Bosh ha-Shanah, 28 a with TO'TD, I, 36. THE HALAKIC LITERATURE 205 ment published in 6.S., pp. 318-25, containing linguistic explanations bearing on the treatise Shabbat, with the corresponding headings in the 'Aruk, and he cannot but be convinced of Rabbi Nathan's dependence upon the Geonim. Rabbi Abraham ben David, of Posquieres, showed keen insight in judging of the value of Geonic contributions to Rabbinic literature. He said, " At the present time we may not explain a Talmud passage other than the Geonim, unless we have irrefutable evidence against their conception of it which is never the case." LIST OF ABBREVIATIONS OF TITLES OF BOOKS I'IN or j*N=sm TIM, by R. Isaac of Vienna. n" or n*iN = D"n nimx, by R. Aaron of Lunel. Vi3ttJN, by R. Abraham b. Isaac of Narbonne. a*a = rmano D'owan nuiirn; the second volume of this book. n*a = Responsen der Geonim, by A. Harkavy. b*a=-o*->n pb ,a'aian rrowi. oi'oa rroVn Vo ,-bV>n ;aiym mw awa niawn. a*n=m l >'na mabn ed. Venice; ed. Hildesheimer is quoted as 'n a*r. mpico rvo^n. m ,mi3a mon. nnsn. , by R. Isaac b. Abbamari, ed. Lemberg, 1860. c, by Rashi, ed. Constantinople, 1802. rrchv nbnp, Geonic collection, ed.Wertheimer, Jerusalem, 1900. aw, ed. Buber. U3*\i} = nrras? 'nyc, by R. Isaac Ibn Gajat. \cn a:*xc n nan . . . oiNn mm. an ma jrrnn Coronel = Nb'in ]3^n bonip *y C':iNa 6. S. = GenizaJi Studies ; the second volume of this book. Graetz = Geschichte der Juden, vol. V, third edition. Halevy = Dorot ha-Rishonim, III. Harkavy. See n'a. 'Ittur. See 110?. J. Q. R. = Jewish Quarterly Review. Mafteah Einleitung in die Responsen der Geonen, by Dr. J. Miiller. Pardes. See DinD. R. E. J. = Rei-ue des Etudes Juives. Sherira = Letter by R. Sherira, ed. Neubauer. Z. H. B. = Zeitschriftfur hebrdische Bibliographie. ADDITIONS P. 4, n. i end. The Geonic Responsum in i>n*2B>, 38 and E. Sherira, 33, 22 ; 34, 6 refer to the same persecution during the reign of "1131P, and there can be no doubt that either fcrtf in ^rfyff is corrupted from ^Dim or niD'QB' from nB>3. Friedmann, in the introduction to his edition of 'l Ifl'^N X D, 1012, has drawn unwarranted conclusions from this corrupted passage. P. 8, n. r, 1. 8. Attention should be called to the fact, that " the Ten of the first row" have their parallel in the -rrparroi Sena of the old Palestinian councils. Comp. Schurer, Geschichte d. jud. Volkes, II, 253, fourth edition. P. 10, 1. 8 read Kimoi. P. 1 2, 1. 12 from below. Nahmanides, NiVn, 28d, quotes a Geonic Responsum where the triad (?) pK31 PJ^X D3n occurs. P. 12, 1. u from below. Comp. Midrash STiemud, XX, 106, ed. Berber: N^N m"On m$>n pKB> 3*3. Does it refer to the triad of the presidency of the Sanhedrin ? Yerushdlmi, Sanhedrin, II, 20 c reads nC^tJt3 PIDS. P. 13, 1. 13 (note). Attention should be called to the fact that DSN '"1, the successor to R. Judah ha-Nassi, was his secretary, comp. Genesis E. LXXV. P. 25, 1. 14. Comp. -|t5>Tl 'D, section pp, 773 ed. Venice, where f*"iN=~vy. P. 25, n. i end. Comp. Midrash ha-Gadol, 190, niTO i?y ''JT'Dyn, and the same in Gaster, nWJ?D, 4 ; the Aramaic equivalent is: 7JJ Dp; comp. Hullin, 97 b. P. 29, 1. 12. R. Sherira speaks ofE.Elhanan as one who was: nniBVi Eni'B'D nhnj nnt^3; "the three rows " are referred to in Mishnah, Sanhedrin, IV, 4, and the Midrash ha-Gadol, 741 : p3CW D'COPl n-'O^n *?V nmt^ vhw l^K ny i>33 nn^B^. P. 32, n. 3. But more likely B>N")n is to be read, the title of the head of the Kairwan academy. P. 40, n. 2. The distance between Bagdad and Sura as given by Funk in the map attached io his Juden in Babylonien, II, is by far too great. P. 51, 1. 9 (note) read Bi*O2, 32, 86. P. 53, 1. 6 from below. In the Egyptian academies the title JH JV3, shortened from JV3 3N p*l, was used (Saadyana, 81), and there can be no doubt that |1N3 as title of the head of the Suran academy is shortened from }1N3 nTB Bn. The description of the academy as 3pN' pN3 208 ADDITIONS reminds one of the Aramaic expression Nfl I| *V1K*T Nip" 1 used by R. Ashi, Berakot, I7b, in speaking of the grandeur of the Suran academy, and there is no need to look for Latin or Persian models for the }1NJ as Kohut (Aruch Completum, s.v.) and Sachs (Beitrage, II, 83) do. Comp. also Abul K. Nathan, 25, ed. Schechter: pNJ JJB^KI lit^N HT p-pn. P. 53, 1. 12 from below. Comp. W*W, I, 63: KTID1 KjmfU, where Njmm = KnnaDIB and J.Q.R., VI, 222. P. 58, 1. 8 (note). Comp. M. Coen ^K^DT, 297, and Jacob Schorr D^oan Wy "VND , 2 7 b 2 8 b, concerning the use of the Talmudic expression .TOp 3TV. P. 71, 1. 20. Comp. HBt3 D^Ha, where Enoch introduces himself to Moses as "pUN *3X. P. 71, n. 2. Comp. Targum, Isaiah xi. i, and Midrash Tehillim, XVIII, 157, where p p = descendant. P. 77, n. 2 end. The scholars of Kairwan (?) probably had in their mind the passage of Yerushalmi, Maaserot, IV, 51 b: m:? 3"iyb = nn 'DHIon, which statement implies that "IB>J?K& njop TOW pDD, else the Talmud would have said TOBTI pSD3 instead of rat? 3"iy. Halevy I.e. and Eatner DvtJTVI JVX rnnK, Pesahim, 124, are of the opinion that the scholars of Kairwan refer to a passage not found in our text of the Yerushalmi. P. 87, 1. 8. Comp. however NTTIN in G. S., 390. P. 88, n. 5. Comp. I^TI 'D, ed. Eosenthal, 80 : ""NHS X 11 imp D^wab {TNT nTi:? ; for -mB> is to be read PTQ5?. P. 93, n. i. Lerner, Jahrbuch d.jud. lit. Gesellschaft, I, 210 et seq., tries in vain to prove the dependence of the Yelamdenu on the Sheeltot. P. 93, n. 2. There can be no doubt that the author of the n, a part of which is read on fastdays, and not to ^Hp* 1 ! as the editions have it; ^n"3B>, 260, quotes this Sheelta ADDITIONS 209 properly as ne> ^nTl Nn^NS?. P. 96, n. i, 1. 8. Comp. 'D 1B*n, 98 and 210. P. 1 08, n. i. The author of the hrfyff quotes a number of passages from the 3'n which are not found in our versions, comp. the list of quotations given below, pp. 1917. P. 112, 1. 2 (note). As late as the time of Maimonides the Rab- banites had to fight this Karaitic heresy, comp. his Responsa, n. 149; comp. also 'n liT^K 'D, XVI, 75, ed. Friedmann. P. 122, 1. ii (note). It is even doubtful whether R. Natronai while speaking of the Haggadic D*BVB thought of Kalir ; the pre- Kaliric Payyetanim, for instance, Yose benYose made use of the Haggadah for liturgical purposes. P. 133, n. i. Comp. Wisdom, xvi. 28 : " That it might be known that we must rise before the sun to give Thee thanks, and must plead with Thee (="]^N ^203) at the dawning of the light." P. 137, 1. 5 from below. Del. the three Hebrew words. P. 142, n. i. The objection of the Babylonians to Kol-Nidre and DH13 man in general is partly due to the fact that there were no DTIO1D in Babylonia, while the Palestinians continued to confer the ordination. P. 145, 1. 16. Comp. -IBTI 'D, 82 : an 'TS ia wv nnetr 'T anan urns? -nrnca moy 'n TIDO D'-nn. P. 145, n. 2. In IPVI 'D, 82, top, the words JWVN y""~\ "V1D31 belong to the preceding sentence, and are to be translated : " and the Seder of R. Amram contains it," namely the benediction over the kindling of lights. A quotation from the Seder not found in our texts is given in 1B"n 'D, 97. P. 149, n. i. Comp. ^Tfyp, 42 with 3*n, 48 and To, 251 with eVca, 72. The differences in the names go back to a different reading of the abbreviation 3""). P. 152, 1. 21. A reference to this part of the Seder is found in -|B*n 'D, 98. P. 167, n. The Seder of R. Saadia is referred to in -|B>V1 'D, 82. P. 179, 1. 20 (note). The j*Bn 'D is quoted in KH n3J?Q, section niDB' towards the end. P. 181, n. 2. Comp. Sachs, pan 7 D, 9-14. P. 182, n. 3. Comp. however the words of R. Hai in f&, 6 ; 94 d. P. 191 (22). Muller, Mafteah, 2 1 o refers to Pardes as the source for this Responsum of R. Hai, but it is not found there. P. 193 (67). Comp. Hildesheimer, ad loc. P. 193 (89). Comp. no^t? n^np, introduction, 15 et seq. P. 193 (101). In the Seder ascribed to Sar Shalom. P. 193 (115). Our texts of 3*1 read differently. P. 195 (258 : NH). The view ascribed to the Gaon ( = Hai) in D'JH D'Dn is just the opposite of that ascribed to R. Hai by the author of ^Y'3B'. P. 197 (399 : I P 210 ADDITIONS Comp. Mordecai, Huttin, 420, j'DD, Commandement, 63 and fix, I, 1 1 4 b, who had the same text of the J*n as ^n*3K> ; Hildesheimer's remark to j'n, 527, n. 59, is to be corrected ac- cordingly. P. 205, 1. 5. This remark of RABeD is found in his MS. niJtJTl against R. Zerechiah Gerondi in the Sulzberger Collection of the Jewish Theological Seminary of America. University of California SOUTHERN REGIONAL LIBRARY FACILITY 305 De Neve Drive - Parking Lot 17 Box 951388 LOS ANGELES, CALIFORNIA 90095-1388 Return this material to the library from which it was borrowed. ARY below. if ^/ea- rn isti v.l