Main Three rolls of the early Septuagint: Genesis and Deuteronomy : a photographic edition (Papyrologische..

Three rolls of the early Septuagint: Genesis and Deuteronomy : a photographic edition (Papyrologische Texte und Abhandlungen)

Categories: Art\\Graphic Arts
Year: 1980
Publisher: R. Habelt
Language: english
Pages: 142 / 158
ISBN 10: 3774914176
ISBN 13: 9783774914179
File: DJVU, 2.81 MB
Download (djvu, 2.81 MB)
 
You can write a book review and share your experiences. Other readers will always be interested in your opinion of the books you've read. Whether you've loved the book or not, if you give your honest and detailed thoughts then people will find new books that are right for them.
1

A Practical Guide to Designing with Data

Year: 2010
Language: english
File: PDF, 4.67 MB
2

Everyone Worth Knowing

Year: 2006
Language: english
File: LIT , 486 KB
THREE ROLLS OF THE EARLY SEPTUAGINT: GENESIS AND DEUTERONOMY

PAPYROLOGISCHE TEXTE UND ABHANDLUNGEN herausgegeben von Dieter Hagedorn, Rudolf Kassel Ludwig Koenen und Reinhold Merkelbach Band 27 RUDOLF HABELT VERLAG 1980 GMBH. BONN

THREE ROLLS OF THE EARLY SEPTUAGINT: GENESIS AND DEUTERONOMY a photographic edition prepared in collaboration with the International Photographic Archive of the Association Internationale de Papyrologues by ZAKI ALY with preface, introduction, and notes by LUDWIG KOENEN RUDOLF HABELT VERLAG 1980 GMBH � BONN

(;cdruckt mir Unte.rstiitzung der Deutschen Forschungsgemeinschaft B(,nn-Bad Gdesberg CIP-Kurztitelaufnahme der Deutschen Bibliothek [Testamentum vetus] Three rolls of the early septuagint genesis and deuteronomy: a photograph. ed./prepared in collab. with the Internat. Photograph. Archive of the Assoc. Internat. de Papyrologues by Zaki Aly with pref., introd., and notes by Ludwig Koenen... Bonn: Habelt, 1980. (Papyrologische Texte und Abhandlungen; Bd. 27) Teilausg. ISBN 3-7749-1417-6 NE: International Association of Papyrologists ? Inter- national Photographic Archive; Aly, Zaki [Hrsg.] HST ISBN 3-7749-1417-6 � Copyright 1980 by Rudolf Habelt Verlag GmbH Bonn Printed in Germany

v PREFACE This publication is the result of collaboration cit Egyptienne de Papyrologie, represented by its fessor Zaki Aly, and the International Photographic and Latin Papyri in Brussels (Fondation Reine Elisabeth), by the Association Internationale de Papyrologues in presented at Cairo by R.A.Coles and myself. In 1971, Dunand had Genesis and Miss ion of from the tion of cally all the fragments, old and new, in order to practice reading fragmentary Greek literary between director, Archive of published P.Fouad inv.no.266 containing Deuteronomy, G.D.Kilpatrick informed the the existence of a same rolls. With the Professor Zaki Aly, it was decided the So- Pro- Greek created 1971 and re- after Mlle Fr. fragments from Photographic large number of additional fragments enthusiastic approval and collabora- to publish photographi- to enable students manuscripts, exer- cise ing the on tification bility of textual criticism and evalution, and further their understand- of the textual history of these two books of the Septuagint. On basis of the photographs, scholars may form their own judgments the certainties and uncertainties of the of fragments. supervising the readings and the iden- Professor Zaki Aly accepted the whole task. responsi- Two missions of the Photographic Archiv in 1973 and 1976 to this project. and arranged in devot- In 1973, the bulk of the columns. M.Weber (Cologne) ed part of their efforts fragments were identified took the photographs under films (Agfa Agepan 35 mm; and the photographs had to be taken flashlights. The quality adverse conditions, for only documentary used with pink filter) were available, coordinated set of two was further impaired with a of the photographs by the fact that I did the fragments before the photographs cations were In addition, Genesis and Deuteronomy, and fragments, particularly of not succeed in identifying and placing all were taken. Further identifi- Cologne, partly by B.Kramer. made after my return to J.W.Wevers, the editor of the Gttingen editions of students of his seminar identified minute roll 3, though their recognition was

VI Preface sometimes fragments In 1974, arranged obscured by variant were mounted on the finally, a set of readings. These newly Mrs. clean photographs and photographs in their proper Shafia Bedier (Cairo; at that time, furnished identified position. Cologne) them with type- written numbers and labels of identification so that they were ready to be processed for printing the plates. It was nevertheless recognized that, in spite of all efforts, the plates would not meet the desired standard. When the printer was about to start his work in mission to Cairo gave us the opportunity tifications against the originals. A new taken by R.W.Daniels, films, a pink filter, The made tives delay. In in 1977, for cessitated The new 1976, another of checking the iden- set of photographs was results were accordingly better. to publish the new photographs, and arranging addition, which he rewriting this time using Kodak Panatomic-X 35 mm and two photographic lamps for illumination. Therefore the decision was though printing from the nega- the prints for making the plates caused much the publication of Wevers' edition of Deuteronomy 1973, ne- had used the photographs taken in the notes accompanying the plates. photographs may still not represent the quality which longer photo- could be obtained through more sophisticated equipment and a stay at Cairo. Though the superior quality of large-format graphy is obvious (see A.Blow-Jacobsen, ZPE 32, 1978, 225ff.), such a camera was not available for taking the photographs of P.Fouad inv.266. Also, we could not keep the water cool enough when developing the films, and this resulted in the sion. Glare was another problem, ical equipment had to be set up quently under changing conditions. Most mistakes of this sort were dis- covered in time, and new photographs could be taken. In a few cases, dislocation of small chips of emul- particularly since the photograph- anew for each working session, fre- Occa- of few in- meantime, With bits of handling, however, sionally the smaller fragments were small parts of the to be expected with each fragments-had broken off in the this extremely fragile papyrus. one risked the loss of minute particles and not available in 19 76, and this was necessary for other reasons as well. 1973. Some in a stances as was we had to resort to the photographs taken in

Preface VII ink. Thus the case hesive paper, photomontages became necessary again. Furthermore, in of a few fragments incorrectly stuck together with ad- the separation of the .papyri papyrus. joinings would have endangered the for correcting their Therefore the mistakes were few did remedied by cutting and pasting the photographs. Finally, unexpected identifications were made as late as 1979, when not have access to the originals. For casionally prints from Agepan had the to be combined. On a very (1973) and few plates, new photographs by parts cut from these photomontages, we even .photographs a we oc- Panatomic-X films (1976) had to supplement taken before the the � the the ad- notes. Photographic Mission began its work. In order to minimize verse effects, all cases of photomontages are recorded in In spite of all shortcomings, it is nevertheless hoped that photographs published here will fulfill their purpose. Many scholars and institutions worked together for this cation. To the names of the scholars already mentioned, I may Hanhart, whose initiative led to the mission's collaboration J.W.Wevers. Without the help of Wevers, the have taken the shape it finally did. He also draft of the introduction and notes. In addition, to discuss many of the problems with members of my seminar in Ann Arbor, especially with my daughter-in-law, M.E.Townsend. In a publication such as this it is hardly to single out individual and the right identification efforts. publi- add R. 1973 and gave a substantial contribution towards the Horace H. Rackham School of Gradua{e Studies which enabled me and R.W.Daniel 1976. The Deutscher Akademischer facilitated the participation of Sayed Omar who was provided a faculty research grant to participate in the mission in Austauschdienst graphic mission in the printing costs; 1973 the and 1976. activities of the photo- He saved the realization at Cairo per- The Deutsche of mitted us the use of its facilities in Forschungsgemeinschaft financed most of me M. Poliakoff read the final draft and the proofs. from numerous mistakes and improved the English. The help of several institutions was crucial for this project. The German Archaeological Institute merits contributions. Too frequently of a fragment was the result of combined with present work would not corrected the earlier I was privileged papyrological the former possible

VIII Preface studying at Cologne. He became, in fact, the Agathos Daimon of the mission in 1976. Above all, we are obliged to the United Nation's Educational, Scientific, and Cultural Organisation and to the Con- sell International de Philosophie et Sciences Humaines. The publisher Dr.R.Habelt deserves high praise. He waited pa- tiently when, in 1976, the plans for immediate publication were shelved and new photographs had to be taken in Cairo. I also wish to express special thanks to those members of the two missions who were not directly involved in this project: Revel and Marsali Coles, A.Fackelmann sen., A.GeiBen, M.Gronewald, E.Molnr, Sayed Omar, and R.Zachmann. The fact that some participants of the missions worked part time on the Septuagint papyri caused an increased load to those who were working on the other projects of the missions. Finally, I thank Professor Zaki Aly not only in the name of the missions, but also on behalf of the future user of this book: It will be the reader who profits from Professor Zaki Aly's scholarly supervision of the whole.project, from his letters in which he urged a quicker publication, and from his hospitality in enter- taining the members of the missions during their working sessions, thus giving them the strength to do a good job. L .Koenen for the Photographical Archive of Gre. ek and Latin Papyri

IX CONTENTS Preface Lists: I. List of Fragments II. Content of Rolls 942, 848, and 847 Introduction 1. Description of 2. Transcripts of (A) P1. (B) P1. (C) P1. P. Fouad On the 50 Col. 51 Col. 52 Col. II: IV: VII: 266 Addendum Arrangement the Rolls Three Columns Deut. Deut . Deut . of This of 847 11,10-11 31 , 26-28 33,1 4-20 Edition Signs (a) (b) and Abbreviations Plates Notes Notes and Plates Notes and Plates Notes and Plates Notes and Plate 848 ( Gen . and Notes Notes 1-3: Roll 942 (Gen.) 4-48: Roll 848 (Deut.) 50: Unidentified Fragments Deut. ) and Plates 50-55: and Plates 56-57: of Roll 847 (ffeut.) Unidentified Fragments Rolls 942 and o f Another V XI XII 1 3 lO lO 12 15 21 21 25 25 25 29 30 38 126 128 Roll 140

xI (1 ROLLS 848/942 I ! LIST OF FRAGMENTS rolls 848/942 cont. fr. G 1 G 2/3 1 2 3/4 5/6 7 8/9 ].2 13/4 15 16 17 18/9 20/1 22/3 24 25 26-28 29 30/1 32 33 34 35/6 37 38-40 41/2 43/4 45 46 47 48 49 50/ 52-4 55 56-8 59 60/1 62 63 64 65 66-68 69-7O pl. 2 col. D 3 col. E 4 col. 1 4 col. 1/2 5 col. 3 6 col. 4 6 col. 6 7 col. 7 7 col. 8 8 col. 9/10 9 col. 11 10 col. 12 10/1! col. 12/3 12 col. 14 I1/12 col. 13/4 13 col. 15 14 col. 16 14 col. 17 15 col. 19 15 col. 20 16 col. 22 16 col. 23 17 col. 24 7/8 col. 7 col. 25 19 col. 27 19 col. 28 20 col. 29 21 col. 31 25/6 21/2 col. 31/2 22 col. 32 22/3 col. 32/3 23 col. 33 22/3 col. 23/4 col. 32/3 33/4 35 36 37 38 39 4O 41 41 42 43 44 46 24 col. 25 col. 25 col. 26 col. 26 col. 27 col. 27 col. 27 col. 28 col. 28 col. 29 col. 30 col. fr. 71 72 73/4 75/6 77-79 8O 81 82 83 84 85 86/7 88-9O 91 92 93 94 95 96 97 98 99/1OO 101/2 103 104 105 106 107 108 109 110 112/3 114 115 116/7 118 119-123 124 125 126/7 128 129/30 131 132/3 134 135 missing (soee Dunand) pl. 30 col. 48 31 col. 51 31 col. 52 32 col. 53 33 col. 54 33/4 col. 54/5 33 col. 54 33/4 col. 54/5 33 col. 54 34/5 col. 55/6 36 col. 58 37 col. 60 38 col. 61 38/9 col. 61/62 39 col. 39 col. 40 col. 40 col. 41 col. 42 col. 42 col. 43 col. 44 col. 44/5 62 63 64 65 66 67 68 7O 71 col. 71/2 44 col. 71 45 col. 72 46 col. 73/4 46 col. 73 6 col. 74 47 col. 76 47 col. 78/9 48 col. 84 1 col. A 7 col. 8 49 16 col. 22 49 1 col. B I col. C 2 col. E 3 col. F 4 col. 1 4 col. 1/2 5 col. 3 6 col. 4 7 col. 7

XII rolls 848/942 cont. rolls 848/942 cont. fr. 136 pl. 137 138 139 140 141 142 143/4 145 146/7 148 149 150 151 152/3 154 155 156-58 159 16o 161 162 163 164 165 166 167 168/9 170 171/'2 173/4 175 176 8 col. 9 9 col. 11 10 col. 12 11/2 col. 13 12 col. 14 14 col. 16/7 15 col. 19 16 col. 22 17 col. 24 20 col. 29 21 col. 31 24 col. 34 25 col. 36 26 col. 38/9 26 col. 38 26 col. 38/9 27 col. 40 27 col. 41 28 col. 43 28/29 col. 43/4 29 col. 44 30 col. 45 30 col. 45/6 33 col. 54 35 col. 56 36 col. 58 36 col. 59 37 col. 60 40 col. 64 41 col. 66 42 col. 68 43 col. 70 45 col. 72 fr. 177 pl. 46 col. 178 47 col. 179 23 col. 180 26 col. 181 26 col. 182/3 49 (2) ROLL 847 fr. 73 78/9 33 38 39 1 pl. 50 col. I 2-4 50 col. II 5-6 50 col. III 7 54 8/9 51 col. IV lo 51 col. IV/V 11 51 col. IV 12 51 col. IV 13 51 col. V 14/5 52 col. VI 16-25 52 col. VII 26/7 53 col. VIII 28-30 53 col. IX 31/2 54 33 = 848 fr. 181 34 pl. 54 35 51 col. IV 36-42 54 43-49 55 (3) P.Fouad 266 add. 1-28 pl. 56 29-53 57 fr. I I, CONTENT OF ROLLS 942, 848, AND 847 Gen . 3,10-12 4,5-7 4,23 7,17-20 37,34-36.38,1 38, 10-12 Deut.4, 18 !O,3 10,8-9 10, 22; 11,1 11,10-11 942 pl.1 col.A* 942 pl.1 col.B* 942 pl.1 col.C* 942 pl.2 col.D 942 pl.2 col.E* 942 pl.3 col.F* cf. 847 pl.54 fr.7 847(?), cf.crit.n. on pl.53 col. IX, (9) cf. 847 pl.54 fr.7 847 pl.50 col.I* 847 pl.50 col. II* cf.847 pl.54 fr.7 Deut.11,13-14.16 17,1-2 17,14-16 17,18-19 18,3-8 18,15-16 19,3-5 I9,5-9 19,10-11.13-14 19,14-15.18-20 �20,3-5 20,5-8 20,8-9.12-14 20,16-19 847 pl.50 col.III* cf.848 pl.49 fr.117 848 pl.4 co1.1' 848 pl.4 col.2 848 pl.5 col. 3* 848 pl.6 col. 4* 848 pl.6 col.6 848 pl.7 col.7* 848 pl.7 col.8 848 pl.8 col.9' 848 pl.8 co1.10 848 pl.9 col.11* 848 p1.10 col. 12' 848 p1.11 co1.13'

XIII Deut. 20,19-20; 21,1-3 21,4-8 21,9-12 21,14-18 21,23 22,1-3 22,6-14 22,21-23 22,24.26-27 22,26 23,3-5 23,7-11 23,14-17 23,21-22.24-25;24,1 23,25-24.1 24,4 24,7-11 24,19-21 25,1-5 25,6-10 25,14-18 25,19 26,1-3 26,3-8 26,11-12 26,13-16 26,18-19;27,1-3 27,3-9 27,9-10.13-6 27,23-24 27,25-26;28,1-4 28,4-11 28,13-14 28,15-17.20 CO1.14' C01.15 CO1.16' CO1.17' .49 fr. 848 pl.12 848 pl.13 848 pl.14 848 pl.14 cf.848 pl 117 848 pl.15 co1.19' 848 pl.15 col.20 848 pl.16 col.22' 848 pl.16 col.23 cf.847 pl.54 fr.7 848 pl.17 col.24' 848 pl.17 col.25 848 pl.18 col.26 848 pl.19 col.27 847 (?), cf.crit. note on pl.53 col. IX, (9) 848 pl.19 col.28 848 pl.20 col.29' 848 pl.21 col.31* 848 pl.22 col. 32 848 pl.23 col.33' 848 pl.24 col.34' cf.848pl.49 fr.117 848 pl.24 col.35 848 pl.25 col.36' 848 pl.25 col.37 848 pl.26 col.38' 848 pl.26 col.39' 848 pl.27 col.40' 848 pl.27 col.41' 848 pl.28 Col.42 848 pl.28 col.43' 848 pl.29 col.44' 848 pl.30 col.45' 848 pl.30 eol.46' Deut.28,31-33 28,49-50.54 28,54-55.57-58 28,58-63 28,63-67 28,68;29,1-4 29,9-10 29,17-20 29,21-22 29,26-29;30,1-2 30,3-4.6-7 30,9-11 30,16 30,19-20 31,2-3 31,5-7 31,10-11 31,14-17 31,17 31,21-26 31,26-28 31,27-30 31,29;32,1 32,1-7 32,2.4 32,7-12 32,14(?).17-20 32,25-26 32,39-43 32,44-49 33,14-20 33,21-22 33,21 33,25-27 848 pl.30 col.48 848 pl.31 col.51 848 pl.31 col.52 848 pl.32 col.53' 848 pl.33 col.54' 848 pl.34 col.55 848 pl.35 col.56' 848 pl.36 col.58' 848 pl.36 col.59' 848 pl.37 col.60' 848 pl.38 col.61 848 pl.39 col.62' 848 pl.39 col.63 848 pl.40 col.64' 848 pl.40 col.65 848 pl.41 col.66' 848 pl.42 col.67 848 pl.42 col.68' cf.847 pl.54 fr.7 848 pl.43 col.70' 847 pl.51 col. IV* 848 pl.44 col.71 847 pl.51 col. V* 848 pl.45 col.72' 847 pl.52 col.VI* 848 pl.46 col.73' 848 p1.48 col.74 848 pl.47 col.76 848 pl.47 col.78' 848 pl.47 col.79' 847 pl.52 col.VII* 847 pl.53 col.VIII* cf.847 pl.54 fr.39 848 pl.48 col.84 847 pl.53 col. IX*

INTRODUCTION The rolls of the Greek Genesis and Deuteronomy owned by the Soci- Egyptienne de Papyrologie (P.Fouad 266), a group of mostly fragments, are cent. famous for their B.C. ); among the age (1st Ryl. 458 acquired Greek text into able. Therefore, small use of the tetragrammaton and their witnesses of the Septuagint only P. is older (Rahlfs 957: 2nd cent. by P.Jouget in 1943. Mlle tion of the papyrus in 1961, but circumstances postponed printing and led to separate publications tion in 19662) and of the edition proper in 1971 3) later, the appearance of the new rolls was hailed the beginning of a new era P.Fouad 266, indeed, shows century B.C. the text of cally steady, though the results of continuous closer accord with the Hebrew agreements between the new papyri B.C. ) I ) p Fouad 266 was Fr. Dunand completed her edi- beyond her control of the Introduc- Seven years by R. Hanhart as 4) of studies in the text of the Septuagint. that already in the middle of the first the Greek Genesis and Deuteronomy was bas i- attempts to are clearly and the Masoretic bring the recogniz- text against the majority of the best manuscripts of the later tra- dition do not necessarily establish what may be regarded as the original text of the Septuagint, but may very well result from later assimilations. Textual criticism of t. he same type as is known 1) On P.Ryl. 458 see J.W. Wevers, Catholic Bibl. Quaterly 39, 1977, 24Off. 2) Fr. Dunand, Papyrus grecs bibliques, vol. de la Gen. et du Deut. (Intro- duction), Rech. d'arch., de philol. et d'hist. 27, Cairo 1966; cf. A. Henrichs, Bibl. Or. 25, 1968, 45f.; G.D. Kilpatrick, Et. Pap. 9, 1971, 221ff.; J.O'Calla- ghan, Stud. Pap. 12, 1973, 96ff. (O'Callaghan is concerned with the general problem of identifying minute fragments). For the use of the tetragrammaton see Dunand, Introd. 39ff. 3) Et. Pap. 9, 1071, 81-150; for photographs see pl. Greek Manuscripts of the Ancient World, Oxford 1971, 56 Waddel, JThS 44-45, 1943/4, pl. facing p. 160; here pl. 106); O. Montevecchi, La papirologia, Turin 1973, pl. 25 I-XV and E.G. Turner, (= Dunand, pl. XIV; W.G. 44 and 45, fr. 104 and (reproduction of Dunand's pll. II (Deut. 19, 10-11) and IX (Deut. 25, 15-17); here pl. 7 fr. 10 and pl. 24 fr. 49. I did not see the New World Translation of the Christian Greek Scriptures, Brooklyn and New York 1950, to which Dunand refers for photo- graphs (Introd. 2 n. 2). 4) OLZ 73, 1978, 39-46, particularly 40.

2 Introduction from the Christian era and is particularly connected with the name had already begun in the first This should be of no surprise. translation existed, attempts and to eliminate discrepancies century B.C., if not even As soon as an authorita- must have started to im- between the Greek and the of Origen 5) earlier. tive Greek prove it 6) Hebrew. of 116 ments. ready outdated. the Socit fragments 7) Most Unfortunately, by the time Dunand's 119 additional fragments Egyptienne de Papyrologie, published by Dunand bring of the new fragments are could be directly joined with Dunand's fragments were published by Zaki Aly in the 8) tained the edition of Dunand's fragments; ing fragments could be identified, although disputable identification (see preface edition appeared, were found in the which when added the total to small, but sometimes fragments. Four of the same volume that and most of the a few resisted p. vf.). it was al- archives to the 235 frag- they new con- remain- an un- The new hand. They fragments contain 49 fragments written in come from a second roll of Deuteronomy. This a different brings the 5) Hanhart, loc. cit. (see n. 4); J.W. Wevers, Text Hist. (see p.26) 64ff.; cf. also his treatment of P.Ryl. 458 (see n. 1) and the roll of the twelve prophets found at Murabbat and published by D. Barthlemy (Les devanciers d' Aquila, Leiden 1963). The earlier optimism that the new variants would confirm the theory, formed on the analogy of the targum, that the text of the Septuagint was not an authorative translation made once and for all, but rather the standard revision which completed the process of gradual coalescence of a large number of translations (P. Kahle, The Cairo Geniza, London-Oxford 2 1952, 209ff. and, on the new fragments, 218ff.; in German: Berlin 1962, 232f.; Dunand, Introd. [see n. 2] 30ff.; Henrichs, loc. cit. [see n. 2]) was premature; see G. D. Kilpatrick, Et. Pap. 9, 1971, 221ff. and Hanhart's sober judgement. 6) Wevers also stresses the unconscious influence of the Hebrew on bilingual scribes during the period when the Bibl. Quaterly 39, 1977, 244; Text some communities, but the original grammaton and calculated the space Septuagint was actually used by Jews (Cath. Hist. 69). Such scribes may have existed in scribe of 848 was unable to write the tetra- so that it would fit x6puoq (see p. 5f.). Even with the Hebrew letters came out of use in the the second scribe who then added % was not familiar (see n. 27). The Septuagint was needed because Hebrew Jewish communities in Egypt. 7) The number of new fragments was originally even larger; which were joined with each other before the first photographs not counted separately. In her Introduction (see n. 2), Dunand ments; her edition contains 116 fragments. 8) Et. Pap. 9, 1971, 227f. with photographs. but those fragments were taken'were counted 115 frag-

1. Description of the Rolls 3 9) total of the new rolls to three. 1. DESCRIPTI ON OF THE ROLLS The fragments were according to Rahlfs' attributed 10) list: to the following rolls numbered 11) 1. 942: Genesis, middle of the 1st cent. B.C., relatively broad columns of c. 1 5 cm. excluding the margins; (+36 letters per line) , riginating shown here on plates 1-3. Only 9 fragments o- 12) from 6 columns (chapters 3, 4, 7, 37 and 38) have been identified. The papyrus is of good quality. No intercolumnation is extant. Also, the numbers of lines per column and the height of the roll are not known. Only parts of the upper and lower margins are preserved (see pl. 2 col. D and pl. 1 col. C). The use of the tetragrammaton is not attested for this roll, but may be inferred from the fact that 942 has probably been written by the same hand as 848 or, at least, by a scribe belonging to the same school 1 cola. little and scribal tradition. Little blanks indicate new 3) There is also a tendency to mark Hebrew names by 14) blanks before and after the names. 9) None of these 49 fragments was published by Dunand. She assumed correctly that her fragments came from two different rolls, one containing Genesis, the other Deuteronomy (see Introd. In. 2] 3). Only 3 of. her 116 fragments came from Genesis; 6 new fragments were added. 10) The number 942 was originally given to the whole lot (Dunand, Introd. [see n. 2] 2). When, however, it became clear that P. Fouad 266 contains frag- ments from two different copies of Deuteronomy, a separate Rahlfs number was attributed to each of the three rolls. 942 was kept for the roll of Genesis so that Wevers' edition of this book of the Septuagint (see the next note) was not affected. The numbers 847 and 848 were given to the two rolls of Deuteronomy by R. Hanhart. 11) For the date see Dunand, Introd. (see n. 2) 9ff.; Turner, Greek Man. (see n. 3) 96 ("perhaps c. 75-25 B.C."); J.W. Wevers, Genesis. Septuaginta, Vetus Testamentum Graecum auctoritate Academiae Scientiarum Gottingensis editum I, Gttingen 1974, 25 ("urn 50 v. Chr."; see also Wevers' description of 848 in his edition of Deuteronomy [see n. 28] 14). Cf. 12) For details see list II (p. XII). 13) P1. 3 col. F, (2). The practice was nn. 24 and 32. obviously the same as in 848 and 847. 14) P1. 1 col. C, 21(?); pl. 2 col. E, (9). Cf. n. 33.

4 Introduction Although too few fragments are extant for this roll to be textually significant, 15) a few passages deserve attention. 16) 2. 848: Deuteronomy, middle of the 1st cent. B.C. (see note 11 ) , shown here on plates 4-49 (unidentified fragments on pl. 49). The quality of the papyrus is the same as in the case of 942, and both rolls have probably been written by the same scribe. The columns of 848 are smaller (11.5 cm. ex- cluding the margins; +37 letters per line); the height of the writing area varies between 15.5-16.5 cm. (21-23 lines) 17) The upper margin was originally at least 3 5 cm., 1 8) the lower margin 4 cm. 19) This indicates close to 24 cm. for the height of the entire roll. Kolleseis are oc- casionally visible, but since too much of the roll is miss- ing, the length of the kollemata cannot be determined. 20) The intercolumnation varies around an average of 1 .5 cm., 15) Cf. J.W. ta-Unternehmens Wevers, Text History of the Greek Genesis, Mitt. des XI, Abh. Akd. Gttingen 1974, 186. Septuagin- 16) See notes on pl. 1 col. A, (1)ff. and B, (2); pl. 2 col. E, (8)f.; pl. 3 col. F, (8) f. 17) Cf. pl. 15 col. 20: 15.5 cm. (22 lines); pl. 27 col. 40: 16.2 cm. (22 lines; the distances between frr. 60, 155, and 61 were probably O.1 and 0.2 cm. larger than they are arranged on the photograph); pl. 29 col. 44: 16.1 cm. (21 lines; frr. 161/68 should have been moved up 0.3 cm.). For the variation in height of the writing area see, e.g. pl. 8 col. 9f. and pl. 10 col. 12f. (the lines of col. 12 raise in the second half). - Most columns had 22 lines. For 21 lines see pl. 9 col. 11 (?; cf. note); pl. 24 col. 35 (?); pll. 28-29 coll. 43- 44; pl. 32 col. 53 (?; cf. note); pll. 39-40 coll. 63-64 (?; cf. notes). P1. 44 col. 71, however, had 22 lines (pace Dunand, Introd. [see n. 2] 4, cf. the note on 71,22; for pl. 45 col. 72 cf. the note on 72,21f.). For 23 lines see pll. 14-15 coll. 17-19 (cf. note on col. 19); pl. 18 col. 26 (cf. note); pll. 19-20 coll. 27 (?) and 29 (?), also col. 30 (cf. note on pl. 21 col. 31); pll. 22-23 coll. 32-33; [cf. note on pl. 24 col. 34]; pl. 31 coll. 51-52 and pl. 41 col. 66 (see note).- All measurements were made on the photographs. 18) See pl. 15 col. 20; pl. 42 col. 67; and pl. 27 col. 41 (3.3 cm.); Dunand, Introd. (see n. 2) 4. 19) The lower margins vary, of course, depending on the height of the area of Qriting. Col. 44 (pl. 29) has an extant margin of 4.1 cm. (21 lines), but the margin of pl. 43 col. 70 (22 lines) actually measures 4.4 cm.; cf. also pl. 15 col. 20 (22 lines): 3.8 cm.; pl. 26 col. 39 (22 lines): 3.7 cm. 20) No search for kolleseis was made on the originals. Those which were ob- served on the photographs are indicated in the notes accompanying the plates. For the terminology see E.G. Turner, The Terms Recto and Verso, Papyrol. Bruxellen- sia 16, Brussels 1978, 15f.

1. Description of the Rolls 5 but occasionally it bottoms the lines have more to the left, thus (Mass ' s Law) 22) paragraphoi narrows down to the tendency producing slightly longer marking the beginning O. 2 cm. 21 ) Towards the to begin progressively lines of verses are used throughout, though not regularly. An additional long oblique stroke marks the beginning of chapter 21 23) Frequently small blanks indicate new verses, sentences, or names are not The ends of to start the 26) are rare. col. 4. surrounded by blanks, lines are occasionally next verse on a new A gloss appears on the cola, 24) while Hebrew as is the case in 942. left blank in order line. 25) Corrections lower margin on pl. 6 It is this which is characterized by the use of � Where it was to occur blank equal to 5-6 letters written in full) and A second scribe roll tetragrammaton ( ) scribe left a size of at its beginning� ters. They cover only the middle marked it by filled in the the the original about the a high dot Hebrew let- of the blank, usually the 21) P1. 47 col. 78/9, 18/9; 0.4 cm.: pl. 31 col. 51,3. 22) See particularly pll. 11-12 coll. 13-14; pl. 14 col. 17; pl. 18 col. 26; pll. 22-24 coll. 32-34; pl. 34 col. 55; pl. 43 col. 70; and pl. 45 col. 72. Turner, Greek Man. (see n. 3) 6. 23) P1. 12 col. 14. This is %he only instance; a reason for this special treatment is not apparent. Chapters 27 and 29, beginning in the middle of a line (pl. 26 col. 39 [no extant paragraphos]; pl. 34 col. 55), are not marked by an oblique stroke; chapter 28 begins on a new line, but is marked only by a para- graphos (pl. 28 col. 43) . 24) New verses: e.g. pl. 31, 5(?); pl. 30 col. 48, (2) [pl. 22 col. 32, 17(?)].-Cola, and 22; pl. 16 col. 22, (3); pl. and 22; pl. 25 col. 36, 19 etc. the blank. In addition, 22 col. 32, 23; pl. 26 15 col. 20,5 (see also n. 26.12.13; pl. 21 col etc. These blanks may be combined with paragraphoi sentences, groups of words: pl. 13 col. 15, 7 22 col. 32, 20(?) and 21 (?); pl. 23 col. 3, 20 The size varies according to the function of there occur blanks for which I see no reason: e.g. pl. col. 38, 17(?; see note ad loc.); cf. n. 13 and 32. 25) e.g. pl. 5 col. 3,10; notes on pll. 24 col. 34, 18(?)f. and 44 col. 71,22. 26) P1. 8 col. 9,18(?); pl. 15 col. 19,6(?); pl. 15 col. 20,5 (see loc.); and pl. 23 col. 33,15, where the scribe had originally omitted a had left a blank for it; later he filled in the missing word (see note All the correcsions may have been made by the first scribe. note ad word and ad loc.).

6 Introduction 3. 847: space of 2 1/2 The bulk of roll half 27) - 3 letters. the fragments (177 (chapters 17-33). Because of of this roll has disappeared jecture that the whole of rolls the first of which pieces) belongs to this the fact that the first entirely, we might con- Deuteronomy was left no traces. tions for this type of preservation are contained But other possible. in two explana- Deuteronomy, second 28) 1 st cent. A.D., fragments on similar to pll. 942 half of shown 54 and and 848. 1st cent. B.C. or, perhaps, early here on plates 50-55 (unidentified 55). The papyrus is of a quality Only 49 very small fragments of a 27) % and are frequently not clearly distinguished in writing. For this and other pecularities see Dunand, Introd. (see n. 2) 13 (at pl. 5 col. 3,6 the � is broken off, not "absent"). See also n. 2 28) In many details, this hand is similar to 848, but larger, thinner, more rounded and irregular, and less bilinear, though the extensions of letters above the "upper line" and below the "lower line" are only small (see particularlye). The occasionally [Schubart' � letters are . decorated by quite heavy strokes s decora- tive style; see particularly � at pl. 51 col. IV, (5)]. Also the middl horizon- tal strokes are stressed (see, e.g., p at pl. 52 col. VI, (6). In partmcular m, (, see Turner, Greek Man. [see n. 3] pl. 21 [R. Seider, Palaeographie der griechischen Papyri II, Stuttgart 1970, IX 16; P.Oxy. 4, 659]: 1st. cent. B.C.; Turner pl. 37 [P.Oxy. 31, 2545, a much smaller hand]: end of 1st. cent B.C. or early 1st cent. A.D.), and ( , though this shape continued well into Roman times) look definitely Ptolemaic. () fits best into the 1st cent. B.C. or A.D. The large s () with the heavy upper curve and the shape of some of the (; not much different from 848) point towards the end of the Ptolemaic and the beginning of the Roman times. P.Fayum 7 (C.H. Roberts, Greek Literary Hands, Oxf. 1955, 9b; Od. book 6; dated: end of the 1. st cent. B.C.) shows the same de- liberately oldfashioned style (cf. particularly m, , and and, on the other hand, s and ; cf. also and p).In general, cf. W. Schubart, Griechische Palaeogra- M6nchen 1925, 116 (Turner, Greek Man. pl. 20; W. Schubart, P. Gr. Berol., Bonn 1911, l lb 9775): 1st cent. B.C.; .Roberts, Lit. phie, Hb. d. Altertumsw. I 4,1, P.Oxy. 15, 1790): 2nd. cent. B.C.; (Seider, Palaeographie II, VI 12; P.Ber. Hands 1Oc (M. Norsa, La scrittura lett. Greca, Florence 246): 66. A.D.; P. Gr. Ber. 18 (Roberts, Lit. Hands l/a; II, XIV 27; P.Berol. 6926B): 1st cent. A.D.; Aur. Leone, scrittura nei pap. Gr. del Vecchio Testamento, Barcelona 1975, 1080): lst/2nd cent. A.D. The last three examples are, however, later than the new fragments, and one may assume that 847 was the same time as 848: On first look, however, the hand older. This could be a deceptive impression. Although I this roll to the 1st cent. B.C., the trema see n. 34) 1939, pl. 7; P.Oxy. 2, Seider, Palaeographie L' evoluzione della pl. VI (P.Oxy. 8, distinctly written around of 847 looks slightly had originally assigned later made me prefer the 1st cent. A.D. Correspondingly, J.W. Wevers assigned this roll to about 50 A.D. (Deuteronomium. Septuaginta, Vetus Testamenturn Graecum auctoritate Academiae Scientiarum Gottingensis editum III 2, Gttingen 1977, p. 14). I am now more in- clined towards a date at the end of the 1st cent. B.C.

1. Description of the Rolls 7 few ments come from chapters margin is extant in col. II (pl. 50). 3O) have had �21 lines. interlinear space are 16.4 cm. Comparing this size of the might assume the same (24 ters, which correspond roughly to 17 cm., letters each are extant. Fortunately, some of the frag- fit together. Those that were definitely identified 10-11 and 31-33.29) Part of the upper column with that the overall height of the roll cm.). The width of the lines averages followed The columns seem to Since the height of the lines and the each about 0.4 cm., this amounts to 848, one 847 was 24 let- by an cm. But the number of letters per 31 ) occur on pl. 52 extant on pl. 51 32) sentences, or cola large blank precedes Moses' the end of line (47) of blessing may begin at intercolumnation lines varies of �1 col. VII col. V. and mark blessing VIII (pl. considerably. Kollesis may (see note ad is Small blanks loc. ) . Paragraphos separate verses, 33) names. A presumedly at order that the Hebrew of Dan, 53) in new line (see n. ad loc. ). No 60Lo full; thus Christian use after tant. ,9c6 is written in scribe did not use the names, which came into (indicating the beginning of col.IV, (11 ) ] . This confirms or tetragrammaton is we may as sume abbreviations of 7O a word) occurs once the date assigned to that holy A.D. "Inorganic" [pl. this col. a ex- the t'r e ma 51 hand, 29) For details see list II (p. XIIf.) . The general distribution of these frag- ments renders it unlikely that fr. 7 (pl. 54) is Deut. 4,!8 or 22,26 (see note ad loc.). 30) This calculation results from pl. 51 coll. IV/V, (4). 31) The length of lines varies between 18 and 24 letters in col. III (pl. 50), between 22 and 23 in col. II, and between 24 and 27 letters in col. I and IX (pl. 53). The lines of col, IV (pl. 51) range from 25 to 29 letters and those of col. VII (pl. 52) from 20 to 30 letters, if this count is not distorted by scribal errors in the lost part of the column. The top of col. VII was consider- ably smaller than its bottom. 32) Before the beginning of verses: pl. 51 col. IV, (11); pl. 52 col. VII, (2); pl. 53 col. IX, (1); cf. also pl. 54 fr. 39, (3).-Before the beginning of subverses, sentences, cola etc.: pl. 51 col. IV, (6) and (10); pl. 52 coll. VI, (!). (7) and VII (10). (15) . (18) . Small blanks also occur elsewhere: pl. 52 col. VI, (2). Cf. n. 13 and 24. 33) P1. 52 col. VII, (14). (15). and (20), all in front ample survives for a blank after those names. Cf. n. 14. of the name. No ex-

8 Introduction though it 34) above. favors the Corrections later part of the 35) are frequent. range indicated identification of The 942 and 848 were probably these fragments of Genesz$ width of nevertheless three rolls is separate written by the and Deuteronomy roll. Such an are no extant the may more ) rolls of rolls 942 different same scribe hardly certain. (see assumption would not account for fragments from chapters Gen. 3 9 - the columns is larger in 942 than have been part of the same ensemble of 36) of the torah. and 847 which, hands. Similar arguments in addition, were come from a the fact that Deut. 16. in 848. preclude clearly Though above ) , single there Moreover, Both rolls 5 (or even the identity written by As for 847 and 848, the same conclusion is obvious. These two rolls were written by clearly different scribes, and Deut. 31 ,26- 32,4 is extant in both, though fragmentary. 37) In fact, 847 and represent different recensions. 38) Nevertheless, both rolls were 848 ap- 34) "Inorganic" is the use of the trema whzch marks the nztal or emphasizes %he fnal vowel of a word, whle ts use for separating vowels in a cluster s called "organic." The "organic" use s attested as early as the 2nd cent. B.C., though t does not become common before the 2nd cent. A.D. For "inorganic" use I am not aware of any example before the end of the 1st cent. B.C.: P.Oxy. 24, 2387 (Turner, Greek Man. [see n. 3] pl. 15, dated to the end of 1st cent. B.C. or the early 1st cent. A.D.); cf. Turner p. 12 and Koenen, ZPE 16, 19u5, 222. 35) The correctzons at col. I, (4) [pl. 50] and col. VII, (16) [pl. 52] may have been written by the original or a second scribe; col. VII, (19) [pl. 52] s surely by a second scribe (cf. and see n. 28), perhaps also col. II, % (pl. 50. 36) If 942 and 848 were part of an ensemble of rolls of the torah (cf. Dunand, Introd. [see n. 2] 3), they could represent the same type of text. But thzs zs by no means certazn. As soon as the torah was dlvzded znto dzfferent rolls, each of them could have zts own textual hzstory. The fact that only remnants of two rolls survzved, whereas at least three rolls entzrely dzsappeared illustrates the zndzvzdual fate of the rolls. For thzs reason zt seemed advzsable to regzster these two rolls under separate numbers. The small quantzty of text of 942 does not provzde conclusive znternal evzdence. 37) ]y): Deut. 31,26 k. rip6 " ' 3 2,2 xcq;c13frrco &q; The following words are extant n both rolls (sometimes only fragmentarl- uktov; 31, 5. 31,28 5uepTb. 31,29 Xsv. 32,1 xe (xouT). ACe% 6poq; vueT6q n. 32,4 T& (pye); Kpcuq. 33,26 - 38) Cf. the verszons of Deut. 31,27f. :n both rolls: 848 (pl. 44 col. 71,4ff.): ms- 847 [pl. 51 col. IV, (8)ff.]: [ms-

Introduction 9 parently used by the same Jewish community occasionally the original form of the text 39) of and witness the early, 4O) Deuteronomy. On the other hand, both rolls show the tendency of harmonizinq the text of the Septuagint with the Hebrew parent text. 41 ) Both are extremely im- portant for our knowledge of the history of the text of Deuteronomy. The value of 847, mentary character however, is substantially reduced by its frag- 42) which allows only tentative conclusions. I a] XC'rov 'rou quk6, pxcq [ xc . I 8'epooq q:b]v e[bv I 5]v (?), o6x, co,'l:'ov ] %'05 [ o,I (11 )v6,%'o]u'p,ou; [ xxXriou.q:', ( ? ) pog IJ,, ] 5.1;t,[Av I ( 12 ) ,,l)UX6, PXOUg (?) xo,, q:'o5g q:',l (13)poug x.%'.X. The addition of 5 ]h and om. 71; uv x6 Touq om. 125) only in 30'� On the other hand, position of ?[v in 11 are pecularities of � oeoTg occurs also in V and 314. These are text of Deut. 847 has the correct o]Teu in ]crv of a minority of mss ; kyt0v ' ' � o. noko uo %'ouq the omission of 7 upmv after uk&pxq (Spry are peculiar to 848� The form ukpX q occurs the addition of (10) ]v (?) and the advanced 848. xxXOT6 (?) instead of all deviations from the original 32,25, while 848 seems to offer the may have been omitted in 847 [pl.53 coi:i9,, (97) see n. ad loc.], whereas 848 does not give an indication for the same mistake.- Other peculiar mistakes of 847: omissions of To% noov [pl.50 col. II, 5: Deut. 11,10; the missing words were added interlinearly] and xe Lpl. 50 col. III, (2): Deut. 11,13]; xngnp]guoeu instead of xngnpgucygg [pl.50 col. II, 3: Deut. 11,10] assimilates this form to the preceding g[oopgOq. For the details see the discussion of coll. II, IV, and VII (below) and for other columns consult the notes attached to the plates. For the idiosyncrasies of 848 see We- vers, Text Hist. (see p. 26) 64ff. 39) The fragments of 847 have been registered under the same number as those of 942 and 848 (P.Fouad 266). Together with the fragments published hereas an ad- dendum (pll. 56 and 57), the fragments of 847 might have been in the box which Fr. Dunand mentions as containing "une masse de dbris minuscules; quelques lettres, parfois A peine lisibles, apparaissent sur les plus grands d'entre eux" (Introd. 2, n. 6). All these fragments were bought together from Phokion Tano, a highly experienced dealer who is not likely to have mixed up different lots of fragments. He probably bought them from the same fellah or �ellahin. If so, all the fragments were probably found at the same spot. 40) Deut. 11,10 &f, crrtop&6 and xr[6n6p6uo't:36 (pl. 50 col� II, lff.; the papyrus has xngnop]guoeu and confirms the perfect form though the second pers. sg. is wrong (see n. 38); Deut. 11,11 go]op[Ou (?) [ibidem, line 6]; Deut. 33,15 [no x6[ at the beginning of this verse: pl. 52 col. VII, (2)]; and Deut. 33,17 n' [ xpou [ibidem, line (11)]. See the discussion below. For 848 consult Wevers, Text Hist. (see p. 26) 64ff. 41) For assimilations of the Greek text of 847 to the Hebrew see the dis- cussion of pl. 52 col. VII, (16)f. and (19) [below p. 18ff.]. For 848 see Wevers and Hanhart, locc. citt. (see n. 5) . 42) It is for this reason that J.W. Wevers made only limited use of 847 in his edition of Deuteronomy (see n. 28).

10 Introduction 2. TRAN. SCRIPTS OF THREE COLUMNS OF 847 An exhaustive within the scope cilitate the use stantiate to offer parts evaluation of this new manuscript does not of this photographic edition. But in order of this extremely fragmentary material and the general evaluation given aboe, 43) it a transcript of the three columns of which can be reconstructed and to add a few textual fall to fa- to sub- is appropriate substantial 44) notes. (A) PL. 50 COL. II: DEUT. 11, 10-11 The top margin is visible. The actual beginnings of lines are certain, and the present arrangement is arbitrary. The preceding column ended (Deut. 11 ,10): [so, by yp q yq sL qv SbOOpSUqb] un- 11,II 1 [ ss u kqpov ] oqc u [ rqv oh- ] frr. 4 [X 03OTT,$p q yq] Atyuou e[atv] [oOsv sxsop] suoat [ I n 6e Yn mm m m m m m m mm m m mmm mm m m m m m mm mm m m m mm m mm frr. frr. 2/3 2/4 2/4 End of the preceding column; guoopgu (-gu 19 75'- 767) B Sa, Rahlfs : Go guoop&u 29 121 = Q Sam TargNeofiti Aeth : uoop&u&o9 & 68' -83 : up&uq variants), TargOnkelos u&uq A 392 (with Prok. 904, Compl, = Mas Lat. 1OO Arm : &uoop&u&o F M V and all others 847 seems offers in line 3 the 2nd pers. sg. (SXTSTop]eoocLL. This to be an assimilation to saaopsunu (see below). Therefore it may be assumed that 847 in fact read eaoopeuqa, as B b n (and others) and Mas (+TargOnk.). If its sLoopsuqu is a correction 43) See above nn. 38, 40, and 41. 44) The collation is based on the edition of Wevers (see n. 28) and A.E. Brooke - N. McLean (The Old Testament in Greek, I 3 Numbers and Deuteronomy, Cambridge 1911). P. 963 was collated for the passages that are extant. For the Hebrew, the Biblica Hebraica Stuttgartensia (fasc. 3 Numeri by W. Rudolph'and Deuteronomium by J. Hempel, Stuttgart 1972 and 1977) was used. The general prin- ciples for the following discussion, the abbreviations, and sigla are the same as will be used in the notes accompanying the plates;see p. 21ff.

2. Transcripts of Three Columns of 847 11 on the basis of the Hebrew, one would follow the Hebrew. Since this is not be the original Deut., though, in the the case, following the original form probably was see Wevers, Text Hist. 125 (without reference on line 6. expect that line 3 would also s Loopsu L may well part of the sentence, For the opposite view to 847) . Cf. also n. 2 0o&p : 0 (OS 54) 54-75' 7 : 7nq B : 550(orig.) d(exc.106): Y 458 3 oS&v - &x&uS&v om. d(exc. 106) [ 1 &x&, - 3 &x[&o- p&uou om. 527; 1 xkpovoou - 3 &x&ue&v om. 381'] oe&v A B F M and most mss. : o8&u Ku 46�-52'646 (cf. the erasure of 3 letters after o8&� in 53) &K&op]&uou 847 alone: existi Lat.lOO Aeth (exc.G) Sa : &x&op&uoO& (-ou G) B V G-426, Rahlfs : 246 z, cf. D% Mas : &xop&u&oO& A F M and all others (&xop&uoO& 318 : &,O710p&u&080,, 767) v The 2nd pers. sg. is best understood as an assimilation to the preceding s uoosu u (see above), which may have independently taken place in some of the translations of the Greek (Lat. Aeth. Sa.); the Hebrew offers the 2nd pers. pl. On the other hand, the perfect (ssoQ]suou) confirms the perfect ssoQsuos of B V G-426. ( Text Hist. 141 ) the Greek translation would have used the the verb of the 6s clause was to express time of speaking. Indeed, this is Wevers because z), if to the peua9e means. argues for the present tense (ops6sos) , aorist (cf. 246 an action prior not what the perfect s6- It denotes rather the fact that the Hebrews suitable. A F M) is in the process of their migration, and this is situation the agreement of 847 and B V (against In spite perfect 40[OpO]� as A B F M V and almost all mss. : opou oou in G Syh) O(exc. 82) Syh [xeu] as A B F M V and almost all of 847 being wrong with regard to the 2nd pers. form helps to recognize Deut.'s E6pEUOE. are still In such a decisive. its sg., asterized mss. : ore. Lat.91 92 94-96 Aeth Pal Palaeographically ] is preferable to ](u (cf. notes on line 5). 5 'TOuq] OOuV as A F M V and most mss. : T. . oou (oou asterized in G Syh) G-426 Syh; = Sam ] %, cf. Mas 3 : T. . eur0� B Prok.904 Aeth Bo : om. 75 [0q] as F oII (exc.707) 59 407 :

1 4 Introduction dominure Pal, = % Mas 9f. Julpray : om. by all mss. and Mas final v with u whether (or of lines about 6 stood might [ u] v is though of so; otherwise the a unique (Ocvc-o I u) the text xGpL, ov) (8)/(9) letters suggestion implies in 848 and 847 are secondary obvious next 45) of which, of 847 was not line reading as ] cannot be the ( 11 ) would have to begin impossible. One wonders to. the full formula p course, is expanded sbv 0v. In this case, the remaining text would have shorter. If it at the end of line (10) have been omitted as in long lines related to manipulations of the been either c. 12 letters longer or was longer, ro] [ so could have [ instead of 9 ] . Otherwise o6spo 120, or r as in 72 71. The latter in (8) and (9) . Whether the epansions each other or not, they are clearly next of Deut. &xxkqoea&Ts (- 319) gxxkoueoeTg A B F M 963 and almost 509 : gxxkoueo& uq 669 (orig.) all mss. : In the context the imperative of deed had the present tense (cf. n. 48), the this with V and 31 4. Since, however, the mistake lography), it would not constitute a link if the reading wmre certain. Cf. p. 12 on aorist is needed. was a mistake which is of a between pl. 50 If 847 in- it shared common type (hap- these mss. even col. II, 51. (11) cont. Sa Syh alone : most al 1 44 col. pouq mss. : upmv A (T. om. and almost (13) f. 508) upv poq (11)-(12) ukePXOUq preceded by our O(exc. U[0v (see p. 7) TOUq Up0v (p0v 46corr.'; mss. (with variants) : Touq 71,7) 848 (apparently) 125 71 om. B 72 75' Compl. (12) om. 125 (12)-(13) B F M V and almost all 44-125) xpL,a,q upmv (u. all mss. : om. 376' TOUq (T. om. (U. om. 72 509 82orig. )-58 ouq ukePXOUq u]c9. v 963) A B ukpxouq (-Xq 848; (12)-(13) x. om. 71) Touq as p a u I po uq mss. (13) om. 44-125 71 : 550' 130 319 Syh, 125) : F0v 54) A B F M V and almost all 128-630-6691 Arm (or -Xeq) 847 F M V and al- see n. on p l. ouq pau- � almost all 71 44-125 : pgauTpouq (x. om. 44-125 71) F0v 246 669 I) A B F M V Compl., cf. Mas (ypa, H, HaT ,q xa,. mss. : om. 550' 45) For the 12. fact that line end has to occur between the two fragments see

Transcripts of Three Columns of 847 15 As the text of lines ( 1 2) - ( 14 ) is lost, I refrain from recording sub- sequent variants. The Hebrew is much shorter: "Assemble unto me ,,46) all the elders of your tribes, and your bailiffs, so that ... Some of the Greek mss. also omit some of the officers. The article and u are frequently left out. It is for these reasons that the position of fr. 12 remains hypothetical. The If frr. least u is assumed have case � ou . reconstruction of lines ( 11 ) - ( 1 3) is particularly puzzling. 11 and 35 both originate from this column (see above), at must be displaced. If uv precedes ou uXQXou, as in the reconstruction given above, 47) 2QsoBursou may not been accompanied by a second uv (Om. in 71 44-1 25) . In this the next uv could easily switch back into the position after All the idiosyncrasies of 847 in lines (9)- (1 3) are of for Deut. no avail (15) eurv : u0v b Pal Aeth M Pal [ apparently] , : om. 64(txt) evTe (obelized Syh; = Mas) Tou : Tou evTe d(exc. 125) t om. 58 No particular (15)-(17). problem surfaces in the few extant letters of lines (C) PL. 52 COL. VII: DEUT. 33, 14-20 The left margin is Q and a blank in the the blank in line 20 ments were joined in Nevertheless, uncertainty Gad (see n. 33). Normally far as we can tell on the extant. Fr. following line. The already begins at the Cairo, partly on the basis of prevails. The blank 24 contains only the single suits 19 soQ [ , end of fr. 23. The fibres. letter and frag- this type basis of the precedes the name of of blank is much smaller, so the limited evidence. The first extant line continues the phrase: 46) Onkelos , Haphtaroth and Rashi ' s Commentary:Deuteronomy, New York 1960. Translated by M. Rosenbaum and A.M. Silbermann (PentaLeuch with Targum 47) Alternatively one might consider a simple scribal omission (or tion) of &xkc& pO & and reconstruct (11) [<&xxko& OO ukepXou] [0�. The precise amount of letters lost in the lacuna is disloca- ,&> oo not known (up to 16 letters), because the beginning of v. 28 may have been marked by a longer or shorter blank.

16 Introduction qkou (33,14) 33, 15 33,16 33, I 7 33,18 33,19 33, 20 (1) (4) (8) (12) (16) (20) [ 'r pOT ] coy [ I [ 'x, opuq)q c2 copa.'u yqg O [ eTO OUVO6COV 1J.n-] .opuq)qc200E:C0� Of.p- ] rXqpoaeo3 xc ] TOCO'TO"30XO[C 'reupou 'To 'x.e,'l,'loc eu'rou] xepcxrcx 1J.o[�oxepc0roc sc xepcxrc eu-] � rou ev [eu'roL, c eOvn xepee el.J.e] ew em [a. xpou YqC uT z [ cloukcov sv ] a. Xa. [ p ev 'To L, C [Eupevq] OOg K I[oo uT I'[ o'r, T[,,lOU'rO ] C e,'%.e[ cane 'to opoc xc ] [ r(x6 os] fr.16 frr. 16/17 frr.21/22 e L e ] v euk [ oynuevoc ] frr. 23/24/25 (2) eo B 707 ms. 566) : xeu ePXlS xe euo C" b s 28 407'Chr. I 80 Arm te Bo Sa(Pierp. Morgan Copt. eO A F M V and all other mss., = Mas xopu om. d (exc. 125) WIt 120 59 646 (2)-(3) op00v o [line (2)] follows a blank. This fact renders it virtually certain that verse 33,14 begins with co (cu om. as in B and oth- ers). The omission of is the lectio difficilior; its addition, on the other hand, corresponds to the Hebrew; it pattern of xcel d6 opucpq in 32,15b (cf. xcel 16d). Since 847, B, and a number of other mss. out initial u, this may well be wha.t Deut. lacking in line (9). may also follow the TT.L 'x. opuq) in 33, offer the text with- once had. Cf. the (4) kposq A B F M V and most mss. : kpcsq eu%nqq (e. under obel. in-

2. Transcripts of Three Columns of 847 17 stead of aster. Syh Mas : X, eurou 75 29 344 (marg.) 407 A B F M V and most B V 707 WI -127-767 O(exc.82)-58 d n(exc.75) : k. euroov 799 (7) (8) To om. 458 mss. : xeu x&peTe d 318 407' Tert. passim 59 646 Lat.1OO Ambr. passim Aeth Arm Bo : 7nq B M V and most mss., = Mas : T %03 A of mss., = Sam Targ Pesch. eure u A B V oII (exc.29) F t (exc. 799) almost all euou om. 414 t Aeth Arm, = Mas : om. 376' 54-75' e(O) A F M and F 72-381' 414 b M and most mss. 59 Syh mss. : (9) m (11) & 18-68' -120 all others and a number 746 confirms the s5 [ line (11 ) ] 48) tion. of B V (16) euoov n(exc.127) 344(marg.) 120 59 Arm A B F M and most mss., Rahlfs Wevers : oou V and others. See Wevers' : eurou (eLroo 346[corr.]) = Ma s :om b Lat.lO0 Aeth, . edi- 4O7 The variants of the development of pattern of ly used in line (16) the text. these blessings, reference to the parallelism zBoukv urou the pronoun of represent three different stages of (1) 6o5 instead of do5 suits the in which do5 and uro5 are respective , aou / the 3rd person in tribe. � IooXig, v woEc the second colon promi s cuous- Nevertheless, in the is harsh and, compared with the 2nd person of V b Lat. 100 Aeth and Mas, the lectio difficilior. We may conclude that Owo5 is the reading of Deut. �2) o5, consequently, represents a to the Hebrew. (3) 6wv is not paralleled its attestation is blessing certainly wrong; allelism between ZBoukv and that this Issachar. � I aoXp. is aimed blessings. secondary assimilation in the other It is weak, and it destroys the par- It may result from the fact particular at the tribes of Zebulun and But it also may be just a simple scribal error (urou > . This type of mistake indeed occurs frequently. For in Deut. 33,11 a in 33, 11c, 630(orig.) ur B 58-426 and a u ins tead the mss. 107' have aoXuv ur instead of offers sXpv urv instead of s. number of mss., = Rahlfs); and in of u. urou is the reading of 664 33,18c, this common type of mistake n [cf. p. 11f. on pl. of Deut. eur > eur) example, uo6 In the case constitute II,5] . UTOU (E. 33,25a, (orig.). does not 50 col. a link between 847 and 48) For (5) Tu eT00u (Too eroo B and others) Walters prefers T T (The Text of the Spt. , see Wevers, Text Hist. 137f. P. Cambridge 1973, 323.

18 Introduction (16)'sev.[] s'.[x]so[vT (or '&Sv.[]' .[ek]&oo[vTeu; the serif at the peak of the vertical stroke of x is all of this letter that is extant) &AS To opo 847 alone; = 1p% 0%D Mas (populus ad montem vocabunt Vulg.) : &Sv &ok&Sp6uoououv (&.with minor variants) xeu &uKek6o&o& (&. [or spelled -o9eu] B V 58-376'-707 344[marg.] z 319 509 646, Compl : &uxek&oovTeu d n t 59 Tht., Dr, = Mas [above]; &Kek&o o& 72 : &uxek&oo9& [or-oSeu] all other mss.) Deut. (16)-(17) [xe I sxs] (= Sam Pesch.)or Isx6 (Mas) : &x& xe Deut. (17) 8uoouou(v) d n t 59 Tht., Dt, = Mas : 8uo&T6 A B F M V and most mss. : 8uoT& (-TEA 319) 29 C'' b 53'-129 85'-344[marg.]- 346-730 and others 8uouev almost all mss. : ictimas Lat.lOO, = Mas (%) : &x&u 8uouev A 246 z 646 At first sight, it is tempting to assume that the scribe in line (16) simply forgot c [o,kcpuaoua t," a.t, 6u- and merely wrote: Llines (1) and (2) 66 L . of this Although c o 1 umn: Qccasionally short lines occur 21 and 20 letters plus blank respectively; respectively pl. 50 col. III, 1 (?) and (identification doubtful) ] 2(?): 17 and 18 letters , the remainder of line (16) is unconvincingly short (19 letters); the preceding line has 29 letters. Moreover, the supplement which results for the beginning of line ( 1 7) is also too short Alternatively one might try to add L at the end of line (16) and read (17) Luoouou] bLxLooulvn], producing two lines of 22 let- ters. On the whole, however, it is much more likely that the orig- inal text closely followed the Hebrew and ran x[X]soo[vrL ro oo. This may seem hypothetical. The correction, however, brings us to firm ground: either the original or a second scribe (see n. 35) restored what corresponds precisely to the. Hebrew. The textual facts are curious and significant. The text of the Hebrew and of 847: "they shall call peoples (i.e. "tribes" or "na- tions") unto the (respectively "a mountain, (and) there they shall offer sacrifices sacrifice," 848 [ ? ] ) of righteousness," 49) became 49) For the addition of "and" as a variant in the Hebrew tradition of the text see the critical apparatus. For the slight possibility that "unto the moun- tain" was not part of the text of 847 see the preceding paragraph.

Transcripts of Three Columns of 847 19 in the you a sacrifice ly into the the Septuagint proper: "they shall utterly destroy nations; (respectively "they") shall call thither, and (there) of righteousness." The Hebrew/847 version fits context, though it escapes us which historical offer benediction announces. mentators whether "people" There was discussion refers to the tribes 50) Christians would easily have understood of Pauline theology: they shall call take part in the sacrifice. If Origen of this version in the mss. of Deut., it the textual tradition of Deut. But it did the Septuagint 51) had long before cleared ty of the word "people" (DD; vq) and expressed and pagans. in the light gentiles to any trace vived in lators of perfect- events hopes: the tribes of Zebulun and Issachar "shall among medieval com- of Israel or the this sentence the pagans and had discovered would have sur- not. The trans- up the ambigui- their national e. the pagan enemies) ness. Obviously, Christians. Thus shall destroy the tradition shall call peoples to the Hebrew; and tradition to Deut. nations ( i. righteous- the ( "they within destroy and (then) call for a sacrifice of this was the only version of Deut. known we might conclude: (1) The version of nations") expresses Jewish sentiment and is, of Deut., original. (2) The version of 847 ("they [to the mountain]") is a secondary assimilation (3) it never made it into the mainstream of the of the Greek Deuteronomy. (19) t_top ucv reading) snopue (sopue eno V 46'-529 (orig.) corrected either sSvc0� 121 68' -83) A B F M and most 53 767 59 Tht., Dt. ep]eku0� into ep]ekue� or ep]ekue by m 2 kuev 58-376'-707 C" (exc.528) Dt. : Repekue (-ku 767) 528 M V and most mss. : eDekueq A B F M V and almost all mss. WI-127 s(exc. 85marg.) 28 59 610 767 : 85(marg.) 407; om. : 'rl:o, pouxouv%'(.,0v 77 mss. : (unique : (],p -, Tht. , 509) A B F 125 xe'ro u xo uv Tv : OUKOUVT� Tht., Dt. 50) Rashi, after he had explained "people" Israel," gives a second explanation: "Through the world's nations will come to this land ... cit. [see n. 47], 174f. as referring to the "tribes of Zebulun's trading, merchants of " (see Rosenbaum-Silbermann, loc. 51) Alternatively bhe translators already expressing the thought which existed. of Deut. may have used a Hebrew now appears in Deut.--if such a recension recension

20 Introduction If we interpret the correction as pkv, it restores the form which is used more commonly than pkt and pktov. Hence it is not likely to be the reading of Deut. Since pktov is much better attested (Rahlfs; a link between 847 the replacement of take (cf. p. 12 on than pkt, pktov is generally accepted for Deut. Wevers) . The false pk v, however, does not constitute and the groups the unusual by pl. 50 col. II, which ofer the same reading, since the usual form is a frequent mis- 5. The interpretation of a mispelling Deut. On the Hebrew of the abscondi tos Hebrew, no ktv could be the reading of represent the the abundance (et thesauros renders the epk ' tv would end the would have left blank (4) a large blank tion is correct, seas) and harenarum Vulg. ) eo to6vv benediction 6f the rest of the the original of a common other hand, 3 ] D]: "(for the treasures monizing reading is ambiguous. p- type and stand for IJ,T6p C a. Ta. pa.k , cO'U might . If could follow. Zebulun, line. At they shall hidden in the Because precedes the blessing of Dan. 847 would bear witness to another peut. with the Hebrew. receive s and" indeed IT6pc the first scribe pl. 53 col. VIII, 52) If this explana- attempt of har- (20) ]Tu: after and above T appears what It does not, however, serve any function. As promised, the text revealed a number of its relations with the Hebrew might be represent Deut., but in attempts restore the true meaning of since the handwriting this papyrus congruences to date of the textual scholarship, looks like a huge apostrophe. of the three columns of 847 studied here characteristics (see p. 8f. ). The ambiguous stressed again. In part, the other cases they are secondary the precedes discloses In 848 this regard (see p. lf. in the work undertaken in Jewish circles. results of the textual analysis of tion it might be stated that, at least the type 847 and Hebrew parent the period of of textual confirms n.4f.).In particular text; and Christian the addi- case of 52) It should, however, be noted that in tory formula: xCc T00[c Av &&v. Here, the 33,22 blank the blank follows the introduc- would precede [xu] T00u [ r6

3. P.Fouad 266 Addendum 21 847, the results of the textual work undertaken in Jewish circles did not directly influence the text of Deut. as transmitted by Christians. 3. P. F O U A D 266 ADDENDUM An additional group of fragments originating from a Ptolemaic by the photographic mission in 1976. According these papyri are registered under roll was seen Professor Z. Ali, as rolls 942, 848, and 847: P.Fouad 266 (cf. n.39). ments are published here on any reasonable suggestions plates 56 and 57. for identification. an impression the dating of the of the 2nd cent. to 53) fair guess. I am No of to the same nu.mber The tiny frag- unable to offer which the handwriting. Therefore of certainty. The middle cent. B.C. seems to be a are long enough to provide words are extant the character of hand lacks any degree the middle of the 1st 4. ON THE ARRANGEMENT OF THIS EDITION This of tion tion. which inferences may be drawn. basis for the placement is frequently uncertain. ters per line varies line reaches blanks letter variation of width indicating the line photographical edition has the original columns, as is Such an arrangement allows is lost between fragments; But the been arranged the case with conclusions in cases of as on the limited reader should of the fragments, since the The number considerably. a reconstruc- Dunand' s edi- amount of text of lines Sometimes far Only safe, carefully per the last word losses textual check the exact position column and of let- into the intercolumnation, or the scribes where margins survived provided we allow for of a left does some 54) at the ends of verses. count become reasonably for each numbers particular line. Different methods of within columns show the grade of cer- tainty we felt in placing the fragments (see p. 25). Occasionally fragments contain the end and the beginning of two columns. When the 53) See, e.g., in fr. 42; K in fr. 8; in frr. 7 and 13; v in frr. 29, 31, 38 and 44 (?); T in frr. 31, 33, 34, and 43; and in frr. 2, 19, and 50. 54) See the description of the rolls mentioned above, A. Henrichs thought it of columns (loc. cit. [see n. 2] 45). pp. 3ff. Because of the difficulties prudent to refrain from reconstruction

22 Introduction it was not graphs an overlapping part of order to indicate the example possible to print both columns on one plate, the photo- in all such cases on both plates in two columns. For pl. 10 shows col. 1 2 had to be broken up for two plates. But the photograph is shown relative position of the fr. 16 contains parts of coll. 12 and 13; plus the beginning of col. 13, and pl. 11 exhibits the right portion of col. 12 in addition to col. 13. The intercolumnation and the ad- jacent writing always appears on both plates. A series of columns exists for with several columns missing 942 (Genesis) the letters A-F roll of Deuteronomy) the Roman avoid the impression of continuous series the 848 able only in between. and for the numbers I- IX roll 848 55) For columns the of are of used in columns. ( Deuteronomy)' columns roll 847 order fragments, one progressive series covers rolls 942 (1st roll of Deuteronomy), because the hands are not F. Dunand (1-113), I the letter from roll 847 (2nd clearly written by 1-84 of roll (2nd to In numbering (Genesis) and distinguish- (1-183; see the List of Fragments, p. XIf.). Since, however, used two separate series for Genesis (1-3) and Deuteronomy distinguish G to the numbers her three fragments (G 1-G 3). As roll of Deuteronomy) and different hands, from Genesis by adding the fragments originating P .Fouad 266 addendum are two separate series of numbers are used for them (947: fr.1-49; P.Fouad 266 add. : fr.1-53). The critical notes facing the plates serve several They begin with information concerning the column as proceed to line by line notes very much in the style cal Lines are able are of which transcribed readings generally purposes. a whole and of a criti- apparatus, though they include palaeographical information. little is extant or decipherment appears difficult for the convenience of the reader. Wrong or debat- previous editions and discussions was not thought worthwhile to in- dicate and statements in corrected, but it each change in the placement of brackets or to add dots be- neath letters certainty of old ones, the un- to the cases where the previous editor did not indicate the the reading. Wherever new fragments were added old brackets' have become meaningless. In some 55) Discrepancies between Dunand's and the new calculation of columns begin at pl. 47 col. 78.

4. On the Arrangement of this Edition 23 however, we to small traces of ink which otherwise This is normally done without explicit thought that the readers attention should be directed would easily reference to escape notice. the previous editions. Reading by other scholars are, however, quoted in cases where a previous edition or the transcript which G.D. Kilpatrick kindly gave to the photographic mission imply that visible. The chances time. In addition, in places of textual other manuscripts are given so that tion in easy reach. For the present more are that those letters got interest lost the the reader has selected purpose, however, it sufficient to refer the remainder conclusions e ings. It is assumed valuation of the variants critical apparatus, as I McLean and, for the (see nn. 28 and 44). 56) ical edition. to a few manuscripts by sigla and to by statements like "and other (most) mss." silentio are not possible. I that whoever wants to letters are in the mean- variants from informa- is usually summarize Therefore did not record misspell- proceed to a careful e- will base her or his research on Wevers' did Hebrew, For the with occasional from J. Hempel' s Vulgate text I help from Brooke- Stuttgart edition used H. Quentin' s crit- In enjoy work. short, the notes do not aim at completeness: the reader may them as an apritif; the main dish will be her or his own 56) Biblia sacra, III Numeri et Deuteronomi, Rome 1936.- I received a copy of M. K. H. Peters, An Analysis of the Textual Character of the Boharic of Deuteronomy, Septuaginta et Cognate Studies 9, Scholars Press 1979, unfortunate- ly after the master pages of the present book were already typed for photomechani- cal reproduct ion.

25 SIGNS AND ABBREVIATIONS (a) P 1 a t e s 1, 2, 3 etc. 1(?), 2(?) etc. ( 1 ) , (2) , (3) etc. (1)?, (2)? etc. 32,39 after least appearing affecting line numbers is extant or the number of a column (e.g.72*): at one new fragment has been added to those in Dunand's edition, or another change the text has been made. the column is numbers line in cases where the upper the position of the line reasonably certain. margin within in cases where the position of the the lines within the column figures is questionable. line numbers extant line fragments; column is as indicated by of the not which are counted from the first a fragment or a sequence of position of the line within the determined by such numbers. line extant line of a sequence the number of lines lost is not certain. numbers which are counted from the first of fragments when between fragments Chapter and verse of Deut. within the same column are repeating the number of the Subsequent verses indicated without chapter. separates columns. (b) Notes The S igla and abbreviations used for o f Gen. Wevers ' the manuscripts and editions and Deut. and the grouping of the mss. into families follow edition (J.W. Wevers, Genesis, Gttingen 1974 [ see n. 11 ] and Deuteronomium, Gttingen 1977 [ see n. 28 ] ) . In some cases, however, a larger part of the abbreviated word is preserved in order to

26 Signs and Abbreviations facilitate typing and recognition by the reader (for for Masoretic text; "TargOnkelos" instead of Wevers' fiti" instead of his TarP). All such deviations from example "Mas" Tar � "TargNeo Wevers' system of notation should be self-explanatory. As in Wevers' edition, dashes between scripts sigla form a for manuscripts (o-oI - 58) indicate that these manu- family in Genesis or Deuteronomy respectively. His mi- nus sign indicating the mss. which deviate from the reading repre- sented by their family (for example, c- ! 6 ) has been replaced by the abbreviation "exc" = except [ as c (exc 16) ] But it should be noted that this notation is also used in cases where a whole pas- sage is not contained by that mss. Deut. (without number of chapter and verse) refers to the critical or original text, whether this text was the result of a single translation from the Hebrew or the final revision of a number of earlier translations (cf. n. 5). In addition, the following abbreviations are used. Dunand, Introd. Fr. Dunand, Papyrus grecs bibl. du Deut. (Introduction), Rech. d' et d' hist. 27, Cairo 1966. Vol. de la Gen. et arch., de phil. Hanhart R. Hanhart, OLZ 73, 1978, 39-46. Wevers see above on J.W. Wevers' editions. Wevers, Bei tr. J .W. Wevers in Beitrge zur alttestamentlichen Theo- logie, Festschrift for W. Zimmerli zum edited by H. Donner, R. Hanhart 1976, 498-505. G6ttingen, 70. Geburtstag, and R. Smend, Wevers, Text Hist. J.W. Wevers, Text History of the Greek Deuteronomy, Mitteilungen des Septuaginta-Unternehmens XIII, Abh. Akad. Gttingen 1978. The symbols used that have become [ ] in reference customary in lacuna of to the text of papyrological estimated three deletion in the papyrus. the papyri editions: letters. are those < > omission in the papyrus.

Signs and Abbreviations 27 superfluous letters on the papyrus. resolution note that different these parentheses purposes when the of an abbreviated Greek word; but are also used for context prevents mi sunders tanding. doubtful letters. traces of unidentified letters. beginning or end small number (I ! of the following beginning or end of line in 3) indicates line (cf. p. of a column. the papyrus. The the line number 57) 25). 57) It should be noted that, refers to the beginning of in the transcripts verses of Deut. given on pp. 10ff. this sign

NOTES AND PLATES

30 Notes on Plate 1 Col. A: Gert. 3,10-12 The identification of fr. 114 is tentative. (1)ff. xe gXpu]]v (>0 (T. 569 Arm : T. ovou all other photograph the fibres look as if sXpu3rlv mss.) [Xceu see below) no ou (ukou ou sveeukv u ou]ou eq u sxev (2) ]vo 5 eu [q, on the they are slightly displaced; o tends to have a straight horizontal upper stroke, short upward stroke� gu; ] �uvoq SU &u (= &u; &, 1]) all and occasionally the foot of & ends in a e�&q; Rahlfs (hesitant question), cf. Mas mss. Col. B: Gen. 4,5-7 (2) gunsIv o 0goq, as Clem. Rom. 4 and Lat. E : gungv upoq o Ogoq all other mss. Col. C: Gen. 4,23 21(?)f. Teuq &euTou �uveuu]v ASeu [xeu %&kke(u) exou- cTg Du Tq c0vq] �ovux[gq Ag X. Instead of A6u, the mss. offer A66 or AS,; in Gen. 4,19 A6e6 and A6e6e also occur� A6eu is probably an erroneous dative

COL. fr. 114 3,10 (1) 11 (4) ' 12 fr. (1) (4) Plate 1 (942) COL. B 124 4,5 fr. 21(?) COL. 125 C 4,23

32 Notes on Plate 2 Col. D: Gert. 7,17-20 There might be a kollesis 0.6-1.2 cm. from the left edge, though the photograph is insufficient proof; cf. Introd. n. 20 1 &]u, on the photograph taken in 1973 the u is not damaged 4f. &[&p&To xuro &ev (&. Tou uSeToq all mss.; apparently too long for 942, see Dunand and Wevers' app.) To 66 uS]p &&KpeT[&U Col. E: Gen. 37,34-38,1 (1) 5u]pp[ri,v (?) (5) keo]v cpc[xcX.o"cu (7) ]o[o]q [ov uuov (8)f. &xX]euo&v [uTov o T0 (&. e. o . euou al- most all mss., apparently too long for 942 : &xX&&v urov 458) ou 422) Iop almost all mss. : T. I. om. 25 (12)

Plate 2 (942) COL. 7,17 18 4 19 8 D 35 36 38,1 COL fr. 126 (1) (4) fr. 127 (7) E 4 37,34 (12)

34 Notes on Plate 3 Col. F: ning of a vertical a photograph Dunand, but (3) 246 and others) (9) Gen. 38, 10-12 stroke taken in 1973) 1(6) Syh] Zeus 15-376' (l)f. O]T[U (this piece with a serif and the begin- was found broken off in 1976; here it is attached from see Kilpatrick, �alZc,.p ] 'r'rlu (5)f. o3 [ 0' &p 961 O' [exc. t5 Arm Syh, Et. Pap. 9, 1971, 225 and Hanhart 40 n. 3) (4) [ouz0 ou ]poq (4)f. uul (5)oq ou (as in most and 72], most =Mas) r] �uvr] mss. or c0[q as in 53' 458 : xeOc0q 376 : � ] mss belonging to C'' b d 53'-56[corr - (6)f. xO[ql(?)o v u ouxcou (Z. most mss. : 0uyT]p [0. under aster� xcu I (9) a. ]pa. xX'r10 uc; [ Io] ustc:

Plate 3 (942) COL. F* () (4) (8) fr. G 2 fr. G 3 fr. 128 38,10

36 Notes on Plate 4 Col. 1: Deut. 17,14-16 The position of fr. 2 is uncertain; see below on line 6 f. 1 suos]kSrs sug (cf. Wevers, Text Hist. 65) : &uo&XSuq ] &uS Dunand Hist. 76) 2 8s]oq (8. 707 52 319 : 8u800ouv (m0o&u 616 Lat.100 8. oou most mss. Dunand; cf. Wevers, Text Am Bo Sa3) oou (. B 58 Lat. 100 Arab Arm Bo : au xk0oo 509 : o. sv xk0t0 [s.xk. obelized G Syh] all other mss.; cf. Wevers, Text Hist. 78 8 8]oq u Aeth Arm) A F and Beitr. 502) 6f. either 6 [SuoT]o0v or 7 ePXo[vTe] o� as in B : 8. oou euTov (eur0v 29[orig.] 500 53 : e. om. V 72 75 121 M V and all other mss; cf. Wevers, Text Hist. 76 10 5]uvcsu [ (2nd. pers. sg., pace Dunand); the iotacism (Wevers, Text Hist. 65) is influenced by the analogy of o6ksu (see nn. on pl. 15 col. 20, 5; pl. 27 col. 40,19); other iotacisms are rare in 848 (pl. 14ff. probably suq Auyul15[Tov o0q (possibly o0q ev as others) Ov kSU]V Ur (?; kUV Ov mss. [o 88 &usv uuv] u 9o1171o&8, the s11 pl. 20 col. 29,23; and 22 col. 32,11 eo&u.[�) in V 106 WI-54' and [with variants]) I ! vertical stroke at the end of line 17, above the damaged p, is probably a shadow; if it is ink, it has no meaning. Col. 2: col. 1 most mss. and most Aeth : Deut. 17,18-19 fr. 2 may 6 (?) f. o6] I 7 (?) 08 [ : 8. oou B b and others) ukoo&o (. other mss. : x ukooscSu [x. ukTTsuv Tou ukeooscSeu 426 : xeu d[exc. 106]) 18 (?) be placed one line lower; cf. n. on 8sou (8. 72-476 : 8. euTou [ukescSeu 376corr.] A M 72] B F b 58-72 b 127 527 [oq q 6vokq. The V precise reading of 848 remains uncertain; after 8sou either cou (less likely eurou) or xeu could have followed; the omission of however, results in a rather short line (even assuming 8soy eurou wu); cf. Wevers, Text Hist.68

Plate 4 (848) 17,14 15 fr. COL. fr. fr. 129 1 COL. ) C.Acu c T co 2 17,18 2(?) 4(?) 19 8(?) fr. 131 16 13 16

38 Notes on Plate 5 Col. 3: Deut. 18,3-8 3f. TOO o[uT]ou xeu Too [ouvou (o. Dunand, cf. Wevers, Text Hist� 71 and 78) xeu I b T]u : or Too o[uu]ou xu Tu xu I b T]v : Tou ouTou (T. . 29-72 46-550" 125 246[orig.] : T. . oou [ all others, =Mas) xeu Tou ouvou (o. 125 : o. oou all others, =Mas) xeu Tou oou (. T. &. . om. 127[orig.] AethM; xeu T. &keuou cou xeu T. ouvou oo u accord- ing to Lat.100) xeu Tv mss. 5 5c0o[g]uq : bc0ouq Dunand o s B Kyr. I 861 Lat.100 Arm : xupuoq o 86o cou all others, =Mas; cf. Wevers, Text Hist. 76 and Beitr. 502 7 ee.crcveu as in 58 b 30'-85[marg.]-321' [marg. ] 59, cf. Dunand's note (her text is incorrect) 15f. as inBC" bdn(exco767) (with variants) : xeu k. all others; cf. Wevers, Text Hist. 79 and Wevers, Text Hist. 79 and many others 17 o]u e- traces are insignificant 20 Tq peo&]?q. [Tq or Tq peo&c0q T].q. [ the bottoms of . and ? and the top of ? are extant, though the

Plate 5 (848) COL. 3' 2 4 fr. 3 18,3 - 4 14 16 17 2O fr. 133 fr. 4 7 8

40 Notes on Plate 6 Col. 4: Deut. 18, [10]. 15-16 The gloss at the lower margin xc]epo6- q mu0oq) is written by a different hand; it comments on Deut. 18,10 20(?) eS]6k[qx]v 21(?) evecrcqo6u o[ou] 1[% o] 86oq cou, oou om. 376- 706 cI' 44-610 129 and many others : placed after oou B; cf. Wevers, Text Hist. 73 Col. 6: Deut. 19,3-5 This column 17(?)f. T[ Toy 118(?) mkouov had relatively long lines C'' b and many others : m. euTou most mss.; cf. Wevers, Text Hist. 77) exouoq (e. A F M V and almost all others : oux &uSq B) ] xo, u ouro[q (o. A B F and most others : our00q M V 618-707 19 125 and a number of others : ou X q 68'-120[corr.] : om. 72) ou ucz0v 119(?) uov (e. almost all mss. : om. 53 30) mpo Tqq 6X86 x]eu TpuTq (T. B 72 : mpo Tqq � . 60e 72 : mOO Tq T. all others), the omission is not recorded in Wevers' app.; the addition of moo q seems to smooth the text; for this reason, the shorter phraseas in 848 B 72 may represent Deut. 19(?) x[ oq v I 2�(?) 6uo&ku 6T Tou m]kouov (m. B and many others : . uTou all others; cf. Wevers, Text Hist� 77) su[q Toy (TOy om. b, and it may have been omitted in 848 too) �u121 (?)ov ouv�6uv (. 707 54-75' 392 Dunand : ouveye�suv A B F M V and most mss.; if Toy was omitted in the preceding line, the longer form 121(?)our - eye�suv could have followed here) uke Keu] sKxpouo[8u Xsup 122(?) eurou qu euvru X]OTOVTO[(; (x. most mss. : TOU XO[TOVTOq 57-528 68'-83-120) TO UkOV II

Plate 6 (848) COL. 4' 18,15 16 9(?) 20(?) fr. lCOYu -' occo' fr. 134 fr. 5 COL. 6 5(?) 6(?) 20(?) fr. 7 19,3 5

42 Notes on Plate 7 Col. 7: Deut. 19,5-9 1 I [xeu sxsoov TO Ou60uO]V 2 k]OUOV most mss. : . ewrou 0" (exc.G) d t and others, =Mas; cf. Wevers, Text Hist� 77 4 mss.; osu is probably the fut. formed from 6c0/c0; see Mayser I 2, 117 and Khner-Blass I 2, 436 and cf. Wevers, Text Hist. 65 (Dunand thought of a scribal error, 00[o&Teu instead of osTeu probably not followed by oou (omitted also in oI[exc. 64]-58-72 552 b 799 319); cf. Wevers, Text Hist. 76 16 the little hole in oq was smaller when the photographs were taken in 1973 17 5]c0u cot, cf. Wevers, Text Hist. 77 b and others) rather than soeKouo] .q palaeographically also oos]q (as 19 exouou] (as B[marg.] 72'' B[txt] F M V and many others); ([or -cg(;] in 71') is possible though not likely; not suoxouo as A 376 121 68'; cf. Wevers, Text Hist. 73 Col. 8: Deut. 19,10-1'1.13-14 5(?) lqv, cf. Wevers, Text Hist. 67 u00o[uv sv Dunand and Wevers, Text Hist. 66, though text 5o[v oo &v (as all mss.); occasionally the margin (see introd. n.21); see Hanhart 40 n. 3 stead of wnu oux soTeu (all mss.); the omission of 848 might have had the full scribe writes well into the 6(?) xg go (=Mas) in- oux in 848 is a common type of Hanhart 4 3 cf. Text yr] scribal error and and for this reason Wevers, Text Hist. 70 M V O' [exc.G]-707 b : [= (most mss. : q [=] B might be unrelated to Mas; cf., however, 20(?)ff. ]v T [ xk121(?)0ovo] and others [848 is not recorded in Wevers' app.; B F and most mss.) and many mss.)] NqN% o [Osoq oou 55ov

COL. 7' fr. 8 19,5 1 4 CNTOC 7 fr. 135 14 16 2O 9 fr. 9 4(? 8(?) Plate 7 (848) COL. 8 19,1o 17(?) fr. 11 13 20(?) 14 fr. 115

44 Notes on Plate 8 Col. 9: Deut. 19,14. 18-20 lff. (cont. from col. 8)I11oo A[corr.] C'' s 18-83 and other mss.) xk]po]vo]oe[u eurv (. e. A[corr.] O'[exc.G]- 707 C'' s 18-83 and many other mss., =Mas : . e. gv xk]p03 M : . b Lat.100 : and all other mss.; 848 offers no evidence for the omission of pace Wevers, Text Hist. 79) oux ggl2vg pu�] xaa [e0ou (x. 58 : variants]; cf. xT (xT 2 46' variant]; cf. indicates the :,pTupr]ou. , 82 : Wevers, Text Hist. 79 -52'n 646 and others Wevers, Text Hist. 65) :,pn:'uprlo, obelized Syh] x. e. all other mss. [with and Beitr. 502) xcI3Tc neomv] e6uxuev [xeu � � � � � end of Deut. 19,15; e:,p'Vq[J,o, .o,l,, .o.'rc all others [with uev, the following vacat � lv ,v (:,t:)'r (A B F M V and almost all mss. [with variants]) was omitted in 848 as in 59 and Arab (not recorded in Wevers' app.) 18(?) I[ouo]eu T0u e6&k00'u' etwrou [, cf. Dunand's note, Wevers, Text Hist. 79f., and Hanhart 44 20(?) I [xeu ou g]ukouou, pace Dunand (see Hanhart 40 n. 3) 21 instead of Dooeoououv, the latter being the reading of all mss. (with variants); cf. Dunand's note and Wevers, Text Hist. 65 22(?) I[Ou&uv (. A M and a]'most all mss. : ouoeu B F V and a few others)] To p, T.p. n Lat.100 Luciferos of Cagliari, Athan. I 7 Aeth Bo, cf.72 : xee. D. all others; cf. Dunand's note andWevers, Text Hist. 66f. Col. 10: Deut. 20,3-5 20(?) Touq [gX8oouq u0v xeu 8ueo00ou uuq]l xeu om. B and a number of mss.; whether 848 had this xeu or not cannot be de- termined, pace Wevers, Text Hist. 73 and Dunand, who both assume its omission 22(?) Toy k[ov Dunand; on first sight, the following text (kyovsq. Tq o [o om. Mcorr. 29-72-376 73 610orig. 246 71-121-527 319] vSpoq o ouxo5ooq) is too long; either ksyovTsq could be omitted in 848 as in Aeth or o In both cases the first line of fr. 13 would be the first line of col. 12. If so, col. 12 would have only 20 lines, so far as this can be established by counting the letters. This seems to be unacceptable. One might rather assume that the text of 848 was longer, reading ksyovsq q oq (oq V) vD011 loq o ooo]- oeq ouxo5oosv (o. added V; cf. 121, where 10 letters were deleted after ouxo- 5ooq)l ]. Alternatively, 848 might have had the very text of the majority of the mss., leaving the end of line 22 blank after ks�ovTsq and beginning line 1 of col. 12 with Tuq o evSD00uo q (K.T.X.). Cf. Dunand's note.

18 19 19 ,I4 15 6(?) P late 8 COL. fr. 136 4 -(848) 9 12b 'o COL. 20 20(?) ] '0 C - ' 12a fr. lO 20,3 17(?) 4 20(?) 5

46 Notes on Plate 9 This column probably had 21 lines; cf. on pl.8, col. [, of the remainder of the line (wvv oux v- seems to have omitted either euvv (as in 602) or Col. 11: Deut. 20,5-8 10,22(?) 2(?) I ov Svogv urv) the scribe (Dunand and Wevers, Text Hist� 65); or Keuv]v preceded ouxuev (cf. the textual problem at the end of col� 10). gxeuvuosv (414) instead of gvsxeuvuosv is not short enough; cf. also line 6(?) 4(?) I v ouue[v 6(?) ?[u]v[u].su [ 6(?)f. ev]17(?)Spc0.o.q.o.[oTuq 16(?) ] oks0. .[eu 22 opsu]goSc0 [eu euo-l 1 or sulllopg[gTc0, see n. on col� 12,1

Plate 9 (848) COL. I 1' 2(?) 4(?) fr. 13 20,5 6 15(?) 7 fr. 14 16(?) 20(?) 8 fr. 137 ' t3

48 Notes on Plate 10 Col. 12: Deut. 20,8-9 (cf. Et. Pap. 9, 1971, 227). 12-14) Fr. 16 ex- tends into col. 13 1 (cont. from col. 11) (o- or su)111 oTp&$[&T0 (We- vers, Text Hist� 67) : eo- or &uoTpesrr almost all other mss.; the present tense [intr. act.] is caused by assimilation to opsu&o80; cf. Deut. 20, 5 and 6 15 I q ] uexoua[av xe oqaov, or -oOUOUv A F M V and ob0ob(v) B and a number of other mss. (with va'riants) Dunand mss. : most 16 spuxe9usuq eur[]v, pace Dunand; see Hanhart 40 n. 3 17 [s0q] epeS0u [oou], cf. s0q ev (so Dunand, but too long) . o. B; for the omission of v see Blass-Debrunner-Funk � 383 : Keu epef0osu A F M V and almost all mss. [K. 8o(su) 56; x. epef0 oou in 376 b is rather a misspelling of x. epeS0osu than a subjunctive used instead of the future]; for a different explanation of the reading of 848 see Wevers, Text Hist� 66 21 K]eu evTe

Plate 10 (848) COL. 12 ': fr. ! 38 20,8 -r C fr. 15 fr. 16 12 15 13

50 Notes on Plate 11 Col. 13: Deut. 20, 16-19 Fr. 16 begins with col. 12; frr. 18 and 19 also con- tain part of col. 14 1-9(?) 68]vc0v, the text of 848 was apparently 1 or 1 1/2 lines longer than Deut. 10 assimilation to Deut. 16,5; 17,2; and % o 96oq oou 6uf0o]uv oou (o. 848 by 20,24 [see Wevers, Text Hist. 67 and 89] o u xk.'ripovop,6 u 'r'riv �'r]v e urcov all 11(?) ekk]e 12(?)ff. probably � xeu (x. B 125 54'-75-767, Rahlfs mss. [with variants]) eveeszusuzs eu]zouq ou I 1 3 ( ? ) -roy X'vre ov : x. om. d 71' : x. Toy almost all other mss., Wevers) Aoppe uov xe u (x. B almost all other mss., Wevers) M oi-29 and many others mss., n[exc. 458], Rahlfs : x. om. 71, =Mas : X]el14(?)[veveuov (x. X. om. d t) xeu (x. A B � Rahlfs : x. om. d 71': x. Toy almost all other mss., Wevers) a0aeuov (x. . om. C" 131 646) xeu (x. C"[exc.131] 646 : x. om d 71'Mas L] Toy IT. V 58-82-376 and many others, Wevers : om. A B F M and all others, Rahlfs] Euauov xeu [x. om. 71'] Toy IT. 376 C''exc.131marg. and many other mss., V and all others [with variants]) Wevers : om. A B F M V and all other mss., Rahlfs] A B F Iou] 115(?)oe.u[ov (x. Iouoeuov om. 44 751) (x. AB[corr.] FMV and many other mss., Rahlfs many other mss., Wevers) [ap�&ouov (x. T. : x. om. d 54 : x. Toy F. obelized in Syh m 58 422 Arab, =Mas) ov Tponov][ 16(?) avaTaukeTo [ oou ]% o 376 C"[exc.422] and om. in o� oou, M and B [txt] the omissions of all but the first Toy as witnessed by 848, B, and partly by others, seems to represent Deut. (so Rahlfs; for the opposite opinion see Wevers who, however, was not aware of 848's reading of this passage); the omission of xeu E6eZov 2t(?)f. is haplography, r]] IJ. 6122 ( ? ) peg � � cf. the similar omissions in other mss. as recorded above.

Plate 11 (848) COL. 1 3 20,16(?) 9(?) fr. 2(?) 139 20,17 fr. 18 fr. 16 5(?) 6(?) 20(?) 18 19 fr. 19

52 Notes on Plate 12 Col. 14: Deut. 20,19-21,3 Frr. 18 and 19 begin with col. 13. Kollesis seems to appear c. 0.4 cm. from the right edge of 5(?) sv 6(?)ff. I ?(?) Xe0 ekke ukov o To o]k0p[ul 9 (?) cuq Tu] .�0u as B 376 � b d n and others; probably ] &uq TOy [X1 ? (?) px kk see ukov �vo) o]T uko[v I � (rather than &o]k&Sp6[ul 9 (?) fr. 19 4 (?) xxo] .s.q..[ Wevers, Text Hist. 73 o &nucrueou (or ] &uq Toy [ ou xcpno3p03]TOV &C[TUV TOU- most mss. (incl. A F M V) add To before the first ukov (T. om. B n and a few others; see Wevers, Text Hist. 73). No other ms. has ukov after oTu, but cf. Mas (F'. P � D D ) and Philon II 97 (var.) dkk Okov voq Tu fruuv auto (see Wevers, Text Hist. 70; Hanhart 43). oksSp6u- Dunand (for 848) : soks0psucsug A F M V and most mss. (with variants) Wevers (for 848) pala eo graph i cal 1 y 9(?)ff. difficult) I !� (?) (the final np.o[� cs noksov, n. 72 C" s 318 28 646, = Mas : Toy n. A B F M V and all mss., also Dunand erroneously for 848 13(?) v (=v) A F M V and most mss.: (=) B 15' C 246 and others; cf. Wevers, Text Hist. 67 and 88f. 20(?) I q �6poucue Tqq nok6� (n. 848 only : n. 6xsuvq all q is i(?) other other mss. ), cf. Wevers, Text Hist. 65

Plate 1 2 (848) COL. 14 fr. 17 20,19 4(?) fr. 139 5(?) 16(?) 20(?) fr. 140 8(?) ;4c fr. 18 --- 12 (?) 21,1 2O fr. 19

54 Notes on Plate 1 3 Col. 15: Deut. 21,4-8 4f. �&upol 5xo0o] Dour 7f. ssk& [s� !s o Qsos, Q. B 246 71 '-527 and other mss. : 0. cou A F M V and most mss.; cf. Wevers, Text Hist. 76 13 Tpcuqi] ucu xcu, x. om. by all others; cf Wevers, Text Hist� 65 14 .i0v 20 ou]$ &kuTQ00 [', as xups in $ sx 7g AuyuuTou xupu& (x. partly om.) or o. &. 30 Aeth(most mss.) Arm(var.) ] xs [X] v 18 oOXFo sx y% Ayumou (s. y. A. obelized in Syh) all others [with variants]; cf. Dunand's note and Wevers, Text Hist� 80 and Beitr. 502f.

Plate 1 3 (848) COL. 15 21,4 fr. 2O 8 6 14 7 16 8 2O fr. 21

56 Notes on Plate 14 Col. 16: Deut. 21,9-12 1 I [euTouq To eu ou 8e ]ep[q To]! 3 TO epsoov xe To xe]kov, cf. Wevers, Text Hist. 80 6 epeS]u oou, . B C''(exc. 320) b and others : om. A F M V and all others; cf. Wevers, Text Hist. 73 7 xeu [pov]o.sucuql 13xsekv Dunand, but cf. her pl. V. Col. 17: Deut. 21,14-18 2f. v �svv]l vg[ 6uo �uvxS, hardly Dunand's �6v]lTeu (as 767[orig.] 319) 4 eurv and 5 uoousv., the first letters of both words were undamaged when the fragments were studied in 1973 7 o uuo o [payoToxo as 767 Philon xo.q all other mss.; cf. Wevers, Text Hist. 80 76[orig.?] 318[with misspelling I 209(var.) : uuoq 8ff. xeTe] 19xkqpov.[oqu (x. Philon I 209[var.], Wevers : B almost all other mss. [with variants], Rahlfs) Touq uuouq (u. Philon II 220 Ambr., Cain I 13 : Touq uuouq eurou A B F M V and almost all other mss. [with vari.ants] Wevers, Text Hist. 78) Te XOV'rc uTou Touq uuouq omitting eurou) , =Mas, Rahlfs; in 848, eurou would be too long; cf. uuepXov] I10Te euro[u (e. om. Philon II 220), Te uuep- euTou 426 C" b and many other mss. incl. Philon I 209(var.; 10f. payoToxu] Illu Tu u[uu, T. u. B b n 407' Philon I 99 and 209, II 220 : Tv uuc0v 344(marg.) : Toy u uov A F M V and all other mss.; the mss. bear witness vers, Text Hist. 73 [ uuoq ueq to their uncertainty about 16 6up6[] the construction; see also We- 18f. &ev 6& Tuvu] 119 .r] (=)

Plate 14 (848) 1 4 6 8 12 11 21 9 lo COL. 16' fr. 22 COL. 17' fr. 21,14 141 __,.co. OCCOt 'i I ,,, 15 '.,<A fr. 23 ' t 12 ' ' 17 fr. 24 18

58 Notes on Plate 15 Col. 19: Deut. 22,1-3 The columns 17-19 must have had 23 lines each (if the text was not shortened) 2(?)f. 5ocTp6%6uq 5]uT5 T[u 586kqxou oou (T. 5. o. A[txt] 0'(exc. G 82]-58-707[txt] C-52-73'-551-761 d f[exc.56] 54-75' and many others, = Mas : T. 5. o. xsu 5KoS6uq 5uT5 Tu 56kx0u ou [F M 29-82 46' 57'-414-417-422-528-550'-615 56 WI-127-767 and others] or T. 5. . x. 5. 5uT0 A[marg.] B V 72 b and others [with variants]; cf. Wevers, Text Hist. 80 and Beitr. 503) esv 68 s]yyu 5(?) euq T� OuXuS]. oOU 6(?) O 5- s]k[oq] oou 5u5, as V0(exc.G) d 53' t : 5ur5 o 5skoq oou all others; cf. Wevers, Text Hist� 72 and 80 7(?) 5uT00]u ou'T' (o. rather than ou'T00[q]'; if q would have followed, a bit of its bottom should be visible) OUr d 53'-246 54 tLexc. 602] and others : xsu our00 458 (Mas Tar) : OUTq all others; Wevers records only the original reading of 848. Col . on the write 20: Deut. 22,6-14 The precise position of fr. 28 has been restored photograph if. II 'rcov ou X.'qr I 'r'qv irl.[mp5 'rov ms- 5 ssu, cf. n. on col. 1,10; Mayser I 2,90. The scribe started to immediately after scs u, but, after writing the s, decided to leave a blank after ed out the mss. : x. xsu &&u in order to mark the beginning of verse 8; therefore he wash- (cf. introd. n. 26) 6 xsvv [o]sq A F b M V and many ou&uq B F 0(exc. G)-15(corr.)-58 and many others; cf. Wevers, Text Hist� 79 9 TOy 5]skv5 cO[U 11 'x.5, TO �sv[5 (848 only; cf. Mas) instead of t&T To% T&vToq results from comparison with the Hebrew; cf. Hanhart 43 and Wevers, Text Hist� 70f. 14 xu8]kuv, error for xu8kov, cf. Dunand, ad loc. (not recorded in Wevers' app.) 17 s0oks]v [oou 5] &5v s..Q.[sku, F_,5v as B 82 Rahlfs (848's reading is not recorded in Wevers' app. nor was it recognized by Dunand) : 5v almost all other mss.; for (v/d.v in Deut. see Wevers, Text Hist. 99ff. 20 po5<u>cruuxouq, cf. Wevers, Text Hist. 65 21

Plate 1 5 (848) 22, I 2(?) 2 4 6(?) 3 8(?) COL. fr. 19' fr. 25 142 fr. 1 4 8 12 16 2O COL. 20 26 fr. 27 . )1 c 22,6 7 8 lO II I3 fr. 28 14 l'OC z, . 4

60 Notes on Plate 16 Col. 22: Deut. 22,21-22.23 [?] nandothers : s. T. 8. (ou) ouxou ographically also possible; cf. see on (22)ff. (1) ! [S]uu Te[q 8u0eq A F M V and almost all others, =Mas; T.[ the rare variants T 8upe ouxou (529) asBb610 is palae- and Super ouxou (527 August., Deut.33); cf. Wevers, Text Hist. 80 (2) apparent- ly I[ku]8ook.[oouou (as 376-707 730 Arm[var.] : k. eur A B F M V and all others; cf. Wevers, Text Hist. 65) o 5s Tq oks I () ].uTq (. A F M V and most mss. : Teurq C''[exc.77 529] s 28 [cf. Tq ok6q ou evSp6q eurq 77; in 848, Teurq would reach into the left margin; this could have been the case oly if Teurq was a correction placed into an original vacat as in col. 33,15] : 6xsuq 381 618[marg.] : eorum Lat.100 : om. b d[exc. 106] 75 and a few others LB and 630 [corr.] om. the whole phrase o. e. T. . e.]) x[eu (x. 72 5OO[orig.] 125 55 Lat. 1OO Arm Bo : 6 ku8ouq xeu A B F M V and almost all mss. [cf. 8oq o. cvSpecj � ok6q euq reap. TeUTq C''exc.413 5OOorig. b n s exc.30' and others] : x. [and 344 (marg.) t and the following euoSevs uTeu] (5) I Toy o[ ,xov (5) f. others : 6p6q A B F M V om. 72; cf. Wevers, Text 6fj0]l(6)p6,T& [, as b d and most mss.; cf. Wevers, 67 (9) I vSpu [ (10)ff. ! Toy 0vS[p0 Toy xo,pot6vov t6 veux[Oq (. T. �. as 392 and Lat.100 : . T. �. xeu Tv �uveuxe A B most mss.; cf. Wevers, Text Hist� 65) or I Toy evS[pe zov xoupmlZSVov ! (11) �uvx[ (cf. Dunand's note), followed by x 6p66 [.. 6. V and others : x. sepsuq A B F M and most mss.; cf. on 6] I (2) A B M V 58-376'-707 b and many tification I�p� [ri]X, I. of ft. 18 is tentative; mind the frequency Hist.67) 85 (marg.)- Text Hist. r](; ] I ( ) yu F M V and b 53 ' -246 ovripov The iden- line (5)f.] Toy mss. (12) ff. of 6voTsuv in this context and the abundance of other words suiting tification is correct, read 6] I (l)v.[OTsusv the three letters. If the iden- and I (14) e[uTqv Col. 23: Deut. 22,24.26-27 (1) eppoTspou]q 6u[ (J6T6 (as 53'-767, =Mas) fits mss. : k u8ookoTeu 46-616 44 75 85 and others cf. Dunand's note (3)f. e]oSeo[uTeu Tv 6eu 6u koou txt]-15 Orig., Cels. I 170 fit; instead of better than kuOo[3okriGr]oovTcu (A (2) B F M V and most te u(Juv WI 318 Arm ), (6. k. O[exc. G 376 Syh, =Mas : om. in all other mss.) oTu] oux may just v6evuv 6u koou, Dunand considered v6evuSe (6) �uveux]e Tou [kouov (14) rather 8eveT]ou q [ than eeveT]ou oTu [q, omission of oq;u as B oI dLexc. 106] 630[corr.]; Dunand's 8ev]eTou [q does not suit the traces (15)f. k.[uov (. as B C'' n s and others : . euou A F M V others; cf. Wevers,. Text Hist. 77) xeu �ov6uoqqu %uxv (x.�.%. c b, cf. D3 and al 1 ["and should kill him with respect to the breath of life"] Mas; so Dunand for (cont. p.. 62)

Plate 16 (848) (1) fr. 144 (4) (8) (12) (13) COL. 22 fr. 143 22,2i 22 fr. 29 fr. 118 COL. 23 fr. 30 (1) (4) 22,24 fr 31 (14) 22,26 27

62 Notes on Plates 16 (cont.) and 17 [cont. from p. 60] 848) or x.. euTov (as 458 Arm); x.. eurou wlv IT. partly om.] uXu A B F M V and all other mss. (too long for 848). If we could be sure that 848 had in fact x. . uX� , this reading would deserve serious considera- tion (cf. P. Oxy. 7, 1033,11 x&uvSu&Ooe& &[ ux; also see Antiphon 2,1,4); may be a secondary stylistic simplification of the Greek (cf. P. Oxy. 6, 903,33 k[v Tv ux ou ["threatening my life"]; P. 4ich. ined. 1681 [de- ciphered by H. W. Witt III; A.D. 381], line 5 k& [= keu] ' &Oex& [= Oex&; "he would have murdered me a long time ago"]). See, how- ever, Wevers, Text Hist. 68f. Col� 24: Deut. 23,3-5 (1) sx].koue[v ]% sq (s. 82 C'' and other mss. : xeu sq A B F M V and most other mss.) or sx].koue[ xeu sq (so Dunand) see Wevers, Text Hist. 71 (2)f. o]u scsk[suosTc sq sxxklqoc s03 (s. 407 Lat.104, =Mas: om. 54 : xcu sa all other mss) or o]ux suosk[suosTcu s. s. xcu s03 (Dunand); see Wevers, Text. Hist. 71 (8) scOOTC]u[c (?) Col. 25: Deut. 23,7-11 Fr.33contains also part of col� 26 8 u]0ov as b : s. oou all others, =Mas; cf. Wevers, Text Hist� 66 9 u o evTo 0eTo ]ovoou, . O(exc.G)-72 f(exc.129) 128-630' and others : xeu . x.!.k. A B F M V and most mss.; see Wevers, Text Hist. 79 10 [o oS oux s]cTeu I! x ouc[]q euro[u 14 so]s[0ev (so][sOeu Dunand)

Plate 1 7 (848) COL. 24 23,3 (1) fr. 32 fr. 145 (8) 23,4 5 COL. 25 fr. 33 2 23,7 4 8 8 9 fr. 34 lO lo 12

64 Notes on Plate 18 Col. 26: Deut. 23,14-17 This column probably had 23 lines ]T[6 (6v[pT Dunand) v T p&ok CyOul (O. om. a F see Wevers, Text Hist. 74 2f. I 6sksoeu os xicom, napa6ouvau X0POV, accordinq to Wevers (Text Hist. 74), oou (A M oi-58-707 was omitted (as B F V and almost all others); the evidence of is inconclusive; cf. Dunand 3 eo [pocz0ou (e. . as'125 : b d[exc. 125] and many others : &uq Teq X6cpeq oou . . 68'-120 Teq X&upeq all others; see Wevers, Text Hist. 67) oou 7f. [ ou 6e Tu % (xuou B Phil. I 156 : xupu eurou A F M V and for the probable omission of e uTou in 848 see Dunand and Wevers, though the evidence is not conclusive) oq I 8 pooT&S&uTeu oou [ euTou, according to Dunand, 848 also omitted the second eurou; possible. 1 I []v[6p.- b and others), o-roy] I 129 and others) the letter count po poc:z0ou B Aeth : q all others; Text Hist. 78, To.,OCZ '"'SOU "i i'1'"" this is quite 376'

Plate 1 8 (848) COL. 26 fr. 33 TO 4 8 23 15 16 12 17

66 Notes on Plate 19 Col. 27: .Deut. 23,21-22.24-24,1 5(?)f. 6ev 86 [ 6ulS(?)u 8(?)f. ms- p[ oou I 9(?) x 6OT oo FupT, oO 407 Tht, Deut. Arm Bo : 6v oo A B F M V and almost all other mss. : om. Ps. August., Speculum 65; rather 6v than left out in 848 (if anything); for the omission of &v oou see Dunand and Wevers, Text Hist� 65 (cf. ibidem 81) and below on next line oux 9(?)f. cont. F M N and many other mss. (see Wevers, Text Hist. 81) : &v oouABO(exc.G) 16 bandmany others; the omis- sion of the entire sv oou (considered by Dunand) is unlikely, while it cannot be ruled out that 848 in fact had &v oou 16(?) ze]q XePOV, in 1973 I saw the first two letters which are now broken off 19(?) Toy v 20(?) oT?.ukv 23(?) uvo]x][c u]T. (. . 12168' : T [with variant spellings] all mss.; see Wevers, Text Hist. 65) Col. 28: Deut. 24,4. 13(?) 'Suv]&T]eu [ oev0 I, extremely thin and ambiguous traces 16(?)f. oTc 816Xul17(?)�e &vevTc (6. A F a M and the majority of mss. : &ueYTuo� B V 426[orig.] C'' s 44 and others : om. F 29-72 610 767 59 319; see Wevers, Text Hist. 115ff.) ]Dh Tou 0&o]u oou &oT[cv xeu I, the mss. place e after S&kuyFu; see Wevers, Text Hist. 67

Plate 19 (848) 5(?) 8(?) 6(?) 20(?) COL. 27 fr. 35 23 21 22 fr. 36 24 :'l . 25 24,1 13(?) 6(7) COL. fr. 28 37 24,4

68 Notes on Plate 20 Col. v (e. 29: Deut. 24,7-11 as AF and many mss. (cf. Et. Pap. 9, 1971, 227). 14(?) To]v [vobov ov or sev as B V d n[exc. 75] t [cf. Wevers, Text Hist. 99 ff.]; av om.M 58-72-381' traces of v and o (the 52'-313-417-528-552 and others) ]eve[7-fsuk]o[uv , the latter on the bottom of fr. 38) are extremely thin and insignificant 16(?)ff.u[ke]os (.as B64-707, C" [most mss.] and many others [cf. also ukeeo9eu 82-381' 46-529'-739[orig.] and many others] or as AF F MV Wevers, Text o5[ (o6 F MV M V and most mss. [cf. also ukeso9eu 376 WI 68' 458 : bvqo9r[s 413 : om. Sa 3) [cusuv 117 (?) [o] 9[soq] (8. C'' [exc. 52'] 53'-56[marg.]-129 and and most mss. : om. AethM; the evidence for 848 is Hist� 76], though the omission is likely) [ Zaki aly) sx]opsuoD[s-] 119(?)vv OHz0v 0" (exc.G) C" and many others : x qq (or zq) many mss. : O. oou A B inconclusive [pace Ayumou., . A. A B, A b and others 19(?)ff. cont. &v 6 (6. F M V and many mss. many others; cf. Wevers, Text Hist. 81) 120(?) 848 alone kLr]]121(?) Text Hist. ! 23(?) p]ov c B 15-58-618 [orig� ] g T][v] 122 (?) o : kqouov oou [and variant] all mss.; OTUOOV TC (TU om. 81) oux [koo] for oTo6u see on col. : om. A B 82 [orig. ]-376 '-707 and Xrlp: ri. v [Xrioo]v (. f. Wevers, Text Hist. 77) o6u- and a few others; see Wevers, xcv 22(?)f. To] 6vgX[o- 1,10

P late 20 COL. (848) 29 24,7 8 fr. 38 11 (?) fr. 39 13(?) 14(?) 9 lo fr. 147 16(?) 20(?) fr. 146 fr. 40

70 Notes on Plate 21 Col. 31: Deut. 24,19-21 The preceding column lines; frr. 43 and 44 belong to coll. 31 and 32 mss., cf. Mas (so Dunand) : To3u ;o3Xo3u xeu (3. as apparently contained 23 8(?)ff. T[u poo19(?)ku- B 426 and a number of other obelized Syh) T pooku- latter reading is too long, even 125, = Mas.); cf. Wevers, Beitr. A F M V and most mss. (with variants); the if both xeu were om:]tted (XeUl omitted in 44- 503� T[u TI9(?)Xu xeu Tu opge]vu (dis- cussed briefly by Wevers, reading Text Hist. 81) would also be a possible though unique 17 &vT&kko]eu ou (. V Lat.lOO Arm Syr : &yt0 [om. Sa ] oou &vT&XXobt, A B F M and all other mss.) [ou&uv 40 p].p. I 19(?) O]UX 20(?) ] poo[XUZ]u

Plate 21 (848) COL. 31 w fr. 43 24, 19 5(?) 8(?) fr. 41 ' -2'Co ,4.A.y 2O 21 5(?) 16(2) 20(?) fr. fr.42 0 fr. 44

72 Notes on Plate 22 Col. 32: Deut. 25,1-5 Frr. 43 and 44 contain coll. 31 and 32; frr. 46 and 48 extend well into col. 33 4(?) ev0[pv 6(?) Toy 5u[xeuov xeu xeTe- � � A B F N V 957 and most mss. or xoov as b 121 68' 407, hardly xeTeyvcouou(v) as 72-82(orig.) 130-321' (so Dunand; cf. xeTeyoo 76' 71) : xeTeyvcsu (-o 107'-125) d[exc. 106) 799 9(?) svev[Tu as the majori- ty of mss. (so Dunand) or vv[Tov as B V and many mss; cf. on col. 28,16f. and here on line 9f. 9(?)f. uoTyt0]110(?)OOuv [urov vvT (s. as 16-414- 529 630[corr.] [so Dunand] or svevTuov as A B F M V and all other mss.; cf. the preceding note) eurv, e. as A B F M V and most mss. : eurou 426 Kyr. I 572 August., Deut.45, =Mas Dunand) epu00, e. A F and Rahl fs; Hist. 68 11(?) ecsue[v (cf. on col. 1,10) eurou (e. om. 71; M V and most mss. : xeu e. B (according to Brooke-McLean Sixt.); Dunand restored xe for 848 19(?) o eSskoq o is a peculiar 13 (?)f. � addition in 848, see Wevers, Text rather eurv as B 29 Lat.100 than eurv as all other mss., though the restoration of 848 remains hypothetical; see Wevers, Text Hist. 81 20(?) scrceu 23(?) �

Plate 22 (848) COL. 32 3(?) 4(?) 8(?) 12(?) fr. 43 ft. 46 25,1 ,,tA. ' 6xt' , 3 %" - 5 2O(?) :., - - fr. 44 fr. 48 fr. 45

74 Notes on Plate 23 Col. 33: Deut. 25,6-10 Frr. 46 and 48 contain parts of fr. 49 extends into col. 34 2f. &] 13x asBO(exc.G) dn coll. 32 and 33; and a number of mss. (partly T6x6u), = subj. aor. act. (pace Wevers, Text Hist. 74) : T&XS A F M V and all other mss. 5 euro[u & Iopek, for the length of this line (see Dunand's note) cf. lines 13 and 14 of this column 9 xe[u 13 []ou[v 15 ur exdented; this word seems to be a correction written by the original scribe. When he was uncertain about his text, he left a blank of three letters; later, after checking the text, he filled the missing word into the blank; but since the available space was too small, he started the word at the left margin. This type of later addition in prepared spaces is a common scribal practice [cf. p.5f. and on col. 38,17(?); the addition of ' by a second scribe in a blank left by the first scribe is a similar procedure]. There is no trace of an erasure (pace Wevers, Text Hist. 65 and app.) 18(?) x om. 72 C'' b d and others; see Wevers, Text Hist. 74 ur[ou o] 6v 20 6uq To A F M V explains this omission as and most mss. : xeTe To B 630(corr.) Arm 20f. om. F 29-72 d[exc. 106] and a number of mss. Wevers an attempt to avoid tautology; this is hardly correct as is a ly translates the Hebrew Greek idiom notwithstanding (similar to k7%0v 66v etc the fact that it precise- .; the form :oxput96oc is koine ) ; see Text Hist. 74 .

Plate 23 (848) COL. 33' fr. 46 fr. 47 fr. 3 25,6 4 7 8 12 14 8 16

76 Notes on Plate 24 Col. 34: Deut. 25,14-18 Fr. 49 contains part of coll. 33 and 34. Compared with col. 33 (23 lines; Dunand, Introd. 4), col. 34 is shorter (22 lines). The line numeration is also established by the letter count of the lost upper portion of col� 34 [13 lines lost; but mind the difficult decipherment of line 14(?)] 14(?)f. I xcu l'rpov ?y[a:ri!uxpov crrc(!uov] ! ]5(?) a:X'r](vov xa:u 6u[xa:uov o'rc ou uva:]! 16(?)okupoq asB(txt) F29 73' d(exc.106) 129 75'-127.602-799 18-630 Philon III 37 Kyr IV 549 Ps.August., Speculum 64 Aeth : x. . . . . 5. xeu 8uxeuov - o u oo u xe u Dpov eke0 uvov x. 8. . �. u. . A B(marg.) M V and all other mss. (with variants), = Mas. It should, however, be noted that Dunand read line 14 (?) as I xeu &uxeuov e[ou xk. ("traces des lettres extre'ment mutiles") . If she were right, 848 would follow the reading of the majority of mss., and line 14(?)ff. would become 15(?)ff. adding a 23rd line to the column (see above). The traces, however, are incompatible with her reading (definitely not euov&). 848 and Philon prove that the haplography, which entered into a substantial part of the tradi- tion, is very old, but it might have occurred independently several times. We ve r s did not yet know the new reading of line 14(?) when he published his edition and Text Hist. 18(?)f. SF_,.o'y'l, followed by a blank marking the transition to a new verse and paragraph (see p. 5; pace Dunand). For the omission of Tu 0&u oou after ]'% (also omitted in Philon III 37 Lat.100 Ps. August., Speculum 64 Arab : 4. . (7. all other mss., = Mas) see Wevers, Text Hist. 76 (Wevers regards the shorter text as Deut.). It would, however, be palaeographically possible ] [u Ou oou]l 1 g (?) aura aq to reconstruct lines 18(?)f. as Fouu eSuxe [eq ouc0v, though this text would hardly represent a Greek sentence nor fit into the pattern of the attest- ed variants (el-end of verse B F M V and most mss. [with variants] : om. d : ouv TeuTe A[orig.] : . . euxov [euxe 52'] A[corr.] 58-72 52' and oth- ers; xeu added before q2 b 246 75' 509 Aeth Arm) 19(?) eSuxe O[exc. G] C'' b 106 246[orig.] n and a number of others (perhaps Deut., pace Wevers, Text. Hist 66) : eSuxOV A B F M V and almost all mss. 21(?) &x[o]o&uo&vou � � 22(?) I x qq A. uyuF[TOU as B Aeth : AuyuTou almo